Abstracts and Poster Sessions

Poster Sessions:

Abstracts Alphabetical by Last Names

A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z

A

COMPLEX REFRACTIVE INDEX OF VOLCANIC ASH RETRIEVED FROM FOURIER TRANSFORM INFRARED SPECTROSCOPY

Jo-Dawn Adams^1,2, Ansel Lavitz¹, Adriana Rocha Lima¹

¹Department of Physics, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250.

²Department of Physics, University of Texas at Austin, 2515 Speedway, Austin, TX 78712.

Around 50-70 volcanoes erupt each year, releasing ash into the atmosphere that can reflect sunlight and, in extreme cases like the 1991 Mount Pinatubo eruption, cause the average global temperature to cool by 1°F. However, global climate models generally do not predict the effects of ash from eruptions due to the uncertainty of optical properties, especially the Complex Refractive Index (CRI). Deriving CRI is critical for understanding how light interacts with particles in the atmosphere, which drives weather patterns and Earth’s climate. This project aims to establish an optical baseline that can improve upon the predictive abilities of global climate models to ensure that radiative effects of volcanic ash are accurately simulated based on measured properties rather than generalized approximations. We used Fourier Transform Infrared Spectroscopy (FTIR) with Attenuated Total Reflectance (ATR) in the 2.5µm-25µm range to derive the CRI of distinct volcanic ash samples. Original and crushed ash samples were placed on top of a diamond crystal to achieve the total internal reflectance necessary for ATR, which was sensitive to particle size. Larger particles were crushed to reduce heterogeneity at the interface between sample and crystal. The CRI was estimated from the Kramers-Kronig relation and fitting modelled Fresnel reflectance to the measured FTIR absorbance data iteratively. The resulting CRIs revealed significant spectral variations among the samples depending on their origin. Integrating these volcanic ash optical properties into climate models may lead to more accurate simulations of ash radiative effects.

This work was supported by the National Science Foundation, grant number 2349543, project title: REU Site: EXPeriments in Earth and Atmospheric Science: Learning Opportunities and Research Experience (EXPLORE).

**CHARACTERIZING THE ROLE OF PLEXIN A AND SEMAPHORIN-1A IN DROSOPHILA MELANOGASTER BORDER CELL MIGRATION**

Titilayomi Adesanya, Christopher Welsh, and Michelle Starz-Gaiano

Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Collective cell migration is integral for various biological processes. We investigate border cell migration in Drosophila melanogaster. Drosophila are an exemplary model for identifying required conserved genes for collective cell migration. Identifying conserved genes important for border cell migration may be useful in understanding their roles in multiple cell types. Border cells travel to the oocyte of the egg chamber, an essential process that occurs during oogenesis. Plexin A (Plex A) is a transmembrane receptor, and Semaphorin 1a (Sema-1a) is a ligand for Plex A. The relationship between Semaphorins and Plexins has been shown to facilitate cell guidance in neuron development but has yet to be studied in border cells. We hypothesize that overexpression of Plex A and knockdown of Sema-1a may cause border cells to migrate individually instead of migrating as a cluster. The knockdown of Plex A may hinder the cluster from detaching from the anterior end of the egg chamber and/or moving forward. If our hypothesis is correct, we expect border cell clusters with Plex A and Sema-1a expression irregularities to have decreased migration distance due to lack of cohesion and difficulties detaching from the anterior end.

To investigate these hypotheses, we employed genetic methods, immunofluorescence staining and microscopy, and imaging analysis software. We altered the levels of Plex A and Sema-1a genetically in a variety of cell types in the egg chambers. We dissected the experimental flies to extract the ovaries and egg chambers. We fixed and stained the egg to investigate differences in border cell migration, adhesion, and detachment. We are quantifying this data by measuring the distance of migration relative to the stage of egg chamber development. We have examined Sema-1a protein expression, and its expression is concentrated in the membrane of multiple cell types, which supports our hypothesis.

This research is funded by the NSF grant IOS-2303587 and the Meyerhoff Scholars Program.

**GALLERIA MELLONELLA LARVAE (WAXWORMS) AS A MODEL FOR STUDYING ACINETOBACTER BAUMANNII PATHOGENESIS**

Aakriti Adhikari, Gulnur Tuluoglu, Khandra T. Sears

Department of Natural and Physical Sciences, Baltimore City Community College, 2901 Liberty Heights Avenue, Baltimore, MD, 21215

The larval stage of Galleria mellonella, waxworms, is gaining increasing use as a tool in biomedical research. G.mellonella larvae have been proven to be a useful and reliable model for studying the pathogenicity of multidrug-resistant Acinetobacter baumannii, an opportunistic pathogen that is difficult to kill. We are developing a model of A.baumannii infection in fifth instar G.mellonella waxworms to assess virulence of clinical strains and test interventions against them. Fifth instar larvae measuring approximately 2-3 cm were used for mortality assays. Groups of 10 larvae were injected with 10l of A.baumannii (10 to 1.3 x10⁸CFU) delivered into the hemocoel through the last proleg. Two control larvae groups were PBS only (injection without bacteria) and non-injected. Larvae were monitored for melanization and death for five days post-infection (p.i.). Most untreated larvae remained alive throughout the experiment with two deaths likely due to natural mortality. Similarly, the PBS-treated control group also showed two deaths throughout the experiment. We observed a dose-dependent response to A.baumannii infection. Larvae exposed to the highest treatment dose began dying within a few hours, with 100% mortality observed within 24 hours p.i.. The second-highest dose resulted in 80% mortality, while the third and fourth highest doses each produced approximately 40% mortality (p.i). The fifth- highest dose resulted in 20% mortality and no deaths were observed in the remaining lower dose groups within the first 24 hours (p.i). Surviving larvae continued development into cocoons. Future goals are to use this model to assess the virulence of clinical isolates and interventions – vaccine candidates or antibiotics – that may limit infections. Importantly, we are able to replicate dose-dependent responses to bacterial infection in this model using simple and easily accessible tools.

This research was supported by Baltimore City Community College.

DECIPHERING THE ADIPOSE TISSUE EXTRACELLULAR MATRIX COMPOSITION ALTERATIONS DRIVEN BY DIET-INDUCED OBESITY

Zaira A. Ahmed¹, Marveline Akinola, BS.¹, Benjamin T. Cole, BS.¹, Diana M. Elizondo, PhD.¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Obesity is characterized by chronic low-grade inflammation and metabolic dysfunction that contribute to insulin resistance and increase the risk of type 2 diabetes (T2D). In adipose tissue, excess nutrient intake drives rapid tissue expansion, placing mechanical and metabolic stress on adipocytes and the surrounding microenvironment. A major consequence of this expansion is pathological remodeling of the extracellular matrix (ECM), which promotes fibrosis, sustains inflammation, alters tissue architecture, and impairs insulin signaling. Although ECM remodeling is recognized as a hallmark of obesity, the specific compositional changes that occur during adipose tissue expansion remain incompletely understood.

The objective of this study was to identify obesity-associated changes in adipose tissue ECM composition using a diet-induced obesity mouse model. Mice were maintained on either a high-fat diet or a chow diet to generate obese and lean cohorts, respectively. Adipose tissue was collected to evaluate collagen expression and histological features associated with obesity-induced tissue remodeling.

Expression of fibrosis- and ECM-associated collagen genes was quantified in whole adipose tissue by quantitative polymerase chain reaction. Histological analyses were performed to measure adipocyte size and number, enabling direct comparison of molecular ECM alterations with structural changes in adipose tissue.

Compared with chow-fed controls, high-fat diet-fed mice exhibited reduced expression of Col1a1 and Col3a1 and increased expression of Col6a3, indicating a shift in ECM composition. Histological analysis further revealed fewer but significantly enlarged adipocytes, consistent with obesity-associated adipocyte hypertrophy. Together, these findings demonstrate that diet-induced obesity is accompanied by substantial remodeling of the adipose ECM and alterations in adipose tissue architecture. Defining these obesity-associated structural and molecular changes provides insight into mechanisms underlying metabolic dysfunction and supports the identification of ECM-related therapeutic targets to improve insulin sensitivity and develop new treatments for obesity-associated T2D and related metabolic diseases.

This work was supported by the NIDDK grant 4R00DK136921-02 and UM FIRST awards to Dr. Elizondo.

INVESTIGATING THE THERAPEUTIC UTILITY OF METFORMIN IN OVARIAN CANCER

Adedoyin Ajibade, Megha J. Pandya, Ayokunnumi Ogunsanya, Achuth Padmanabhan

Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Ovarian cancer is an aggressive gynecological disease that is frequently diagnosed at advanced metastatic stages because early progression is largely asymptomatic and reliable screening strategies are lacking. Although primary treatment strategies for ovarian cancer patients, such as cytoreductive surgery and chemotherapy, show initial signs of effectiveness, over 85% of ovarian cancer patients experience relapsed tumors within their first 6 months of treatment. Relapsed tumor cells are resistant to extant therapeutics. To address this unmet clinical challenge and to develop new clinically translatable therapeutic strategies for metastatic ovarian cancer, we investigated the therapeutic potential of the FDA-approved anti-diabetic drug metformin in ovarian cancer cells. Metformin significantly reduces cell viability across five ovarian cancer cell lines, suggesting potent cytotoxicity against a broad spectrum of ovarian tumors. Further, metformin significantly impairs several metastatic phenotypes in ovarian cancer cells, such as the ability of these cells to migrate, invade through the extracellular matrix, resist anoikis, and form multicellular aggregates. Systematic screening of an FDA-approved drug library identified that metformin sensitizes ovarian cancer cells to multiple clinically approved small molecules, many of which have not previously been investigated for the treatment of cancer. Future studies will determine the utility of these new drug combinations involving metformin as a therapeutic strategy in ovarian cancer using clinically relevant in vitro as well as in vivo models. This research will enable the clinical repositioning of metformin as an adjuvant therapy to overcome chemoresistance in metastatic ovarian cancer.

Support for this research was provided by research grants from the following sources to Dr. Achuth Padmanabhan – Ovarian Cancer Alliance of Greater Cincinnati (PRIV0201 and PRIV0219), UMGCCC American Cancer Society Institutional Research Grant (IRG-18-160-16), the Department of Defense (HT9425-23-1-0351 and HT9425-23-1-0232), and the National Institutes of Health (R03CA282712).

INVESTIGATING CONSERVATION OF THE HIV-1 CORE ENCAPSIDATION SIGNAL

Aaron Johnson^1,2; Sofia Shevchenko^1,2 ; Zeynep Akan^1,2; Vianney Tanifor^1,2; Horasa Ji²; Michael Summers^1,2

¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

²Howard Hughes Medical Institute, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

Human immunodeficiency virus type 1 (HIV-1) selectively packages its genomic RNA through a highly structured region known as the Core Encapsidation Signal (CES), the minimal RNA element required for genome recognition and packaging into new viral particles. Unlike the full HIV-1 leader RNA, the CES lacks the TAR, poly(A), and primer binding site (PBS) regions while retaining key structural elements, including the dimer initiation signal (DIS), splice donor (SD)-associated region, Ψ packaging signal, AUG region, and the U5:AUG interaction. Previous studies of the HIV-1 NL4-3 laboratory strain demonstrated that the SD region adopts a continuous structure rather than folding back on itself in the 1G conformation. This project investigates whether the same structural arrangement is conserved in the HIV-1 MAL strain. Using nuclear magnetic resonance (NMR) spectroscopy, we aim to characterize the three-dimensional structure of the MAL CES RNA and compare its folding with that of NL4-3. Determining whether this conserved architecture exists across strains will improve our understanding of the structural mechanisms underlying viral genome packaging and may provide insights into conserved RNA features that could serve as future antiviral targets.

Funding for this research is supported by the Howard Hughes Medical Institute (HHMI) and NIAID #5R01AI150498

**THE IMPACT OF NEBULOSA KNOCKDOWN IN THE FAT BODY ON AGE-DEPENDENT IMMUNOCOMPETENCE IN DROSOPHILA MELANOGASTER**

Casey Duhon¹, Evangeline Chen², Pranati Denduluri¹, Zahraa Al Aanizy¹, Jeff Leips¹

¹Department of Biological Sciences, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

²Department of Biology, UMCP, 1204 Biology-Psychology Building, College Park, MD 20742

Immunosenescence, the age-related decline in immune function, is a complex trait shaped by both environmental and genetic factors. Previous genome-wide association studies (GWAS) identified the gene nebulosa (nebu), a gene encoding a carbohydrate transmembrane transporter, as a contributor to natural variation in age-dependent bacterial clearance ability in Drosophila. While initial functional validation demonstrated that hemocyte-specific knockdown of nebu significantly reduces clearance, the role of this metabolic gene in other primary immune tissues remains unexplored. In this study, we test the hypothesis that nebu provides metabolic support for the immune response within the fat body, the central organ for antimicrobial peptide production and nutrient storage. Using the GAL4-UAS system, we generated tissue-specific RNAi-mediated genetic crosses to knock down nebu expression in the fat body using a fat body driver line. We crossed this driver with a line containing a UAS-RNAi nebu construct and an isogenic control line with normal nebu expression. This is an ongoing project. Our plan is to infect virgin female flies aged 1 week (young) and 5 weeks (old) with streptomycin-resistant Escherichia coli (E. coli). Bacterial clearance ability will be measured 24 hours post-infection through individual homogenization and colony counting to estimate remaining infection titres. Knockdown efficiency of nebu will be validated using quantitative PCR (qPCR) to measure nebu transcript levels relative to the housekeeping gene eIF1a. This research will contribute to our understanding of the metabolic basis of immunosenescence and help identify potential targets for mitigating age-related immune decline.

Support for this research was provided by the Department of Biological Sciences at the University of Maryland, Baltimore County.

ENHANCING T7 RNA POLYMERASE TRANSCRIPTIONAL HOMOGENEITY FOR STRUCTURAL STUDY USING TEMPLATE MODIFICATIONS

Sridhar Alagar Ramanujam¹, Eunice Ewusie¹, Geoffrey Okonkwo¹, Brandon Fonseca¹, Michael Summers, Ph.D^1,2.
¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250
²Howard Hughes Medical Institute, 4000 Jones Bridge Rd, Chevy Chase, MD 20815

Human Immunodeficiency Virus (HIV-1), is a retrovirus affecting nearly 41 million people globally. While anti-retroviral therapies (ART) suppress viral replication, strict regimen adherence is difficult due to off-target effects and burden of treatment. To address these limitations, further structural characterization of conserved functional areas such as the 5’-leader (5’-L) is needed. However, high resolution nuclear magnetic resonance(NMR) studies of HIV-1’s 5’ leader requires homogeneous RNA samples. Achieving this proved challenging as T7 RNA Polymerase (RNAP) induces 3’ transcriptional heterogeneity through promoter-independent self-templated extension, driven by T7 RNAP’s affinity for RNA oligonucleotides. Previous strategies meant to combat this such as incorporating methylation (2’OME) into the final two 3’ nucleotides has been proven insufficient, as T7 RNAP frequently introduces 3’ heterogeneity. This project evaluated two approaches to improve T7 transcriptional homogeneity. 1) Using a T7 RNAP enzyme mutant which was designed to reduce 3’ heterogeneity, increasing the yield and reducing purification needed of the desired RNA. 2) Using commercially available template modifications that terminate transcription cleanly through steric or hydrogen bonding considerations. Through these two approaches, our work aimed to establish a practical method for increasing T7 RNAP transcript homogeneity. In doing so, we provided a robust framework to overcome transcriptional heterogeneity bottlenecks in high-resolution structural techniques such as NMR.

Funding for this research was supported by the Howard Hughes Medical Institute and the National Institute of Allergy and Infectious Diseases (NIAID, 5R01AI150498).

INFLUENCE OF THE CERVICOVAGINAL MICROBIOME ON FIBROBLAST ACTIVATION IN A CERVICAL CANCER MICROPHYSIOLOGICAL SYSTEM

Hanif Ali¹, Sudarshan Bollapragada¹, Corine Jackman Burden¹

¹Department of Chemical, Biochemical and Environmental Engineering, University of Maryland Baltimore County, 1000 Hilltop Circle, Baltimore, Maryland, USA 21250

Cervical cancer is the fourth leading cause of cancer-related death among women worldwide, causing over 350,000 deaths annually despite preventative widespread human papillomavirus (HPV) vaccinations and screenings. While persistent high-risk HPV infection is the primary driver of cervical cancer, bio-chemical and -physical factors, including pH, O₂availability, and stromal architecture of the surrounding stromal microenvironment, also facilitates disease progression. Inflammation has been shown to promote tumor progression, in part through the activation of fibroblasts into cancer-associated fibroblasts (CAFs). CAFs are major regulators of stromal remodeling, cytokine production, and immune cell recruitment. Lactobacillus crispatus is associated with vaginal health, whereas anaerobic bacteria, like Gardnerella vaginalis, are associated with dysbiosis (the imbalance of microbes in the female reproductive tract), chronic inflammation, and increased cervical cancer risk. However, despite evidence linking cervicovaginal dysbiosis to HPV persistence and inflammation, it remains unknown whether distinct microbial community state types directly regulate fibroblast activation and CAF formation. We hypothesize that G. vaginalis is more likely to promote fibroblast activation and CAF differentiation than L. crispatus through sustained inflammatory signaling. Currently, conventional monolayer cultures lack the ability to recapitulate the multicellular cervical microenvironment, whereas microphysiological systems (MPSs) enable controlled interrogation of epithelial-stromal-microbial interactions. By combining an MPS-based approach with HPV-positive cervical epithelial and fibroblast cell cultivation, we will determine how microbial composition influences fibroblast activation. We have shown that L. crispatus and cervical cancer cells both grow in DMEM medium, a preliminary result that is necessary for co-cultivation. Future work will assess CAF activation by α-SMA immunofluorescence, fibroblast activation protein gene expression by qPCR, and IL-6/IL-8 secretion measured by Luminex. This work will establish a mechanistic link between HPV and cervicovaginal microbial composition and stromal remodeling, providing a foundation for microbiome-directed therapeutics to prevent cervical cancer progression.

EFFECTS OF ZINC EXPOSURE ON NEURODEGENERATION IN DROSOPHILA MELANOGASTER

Amina Ali¹, Justine Anne Guevarra¹, Fernando Vonhoff¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000

Hilltop Circle, Baltimore, MD, 21250

Alzheimer’s disease is a neurodegenerative disorder associated with the accumulation of amyloid-beta (Aβ) plaques, which result from abnormal processing of Amyloid Precursor Protein (APP). Although APP is known for its role in Alzheimer’s disease, it has been shown to be involved in neuronal growth, synapse formation, cell communication, and maintenance of neuronal function. APP also contains zinc-binding domains that help regulate zinc homeostasis in the brain. Zinc is an essential trace metal required for neuronal signaling, enzyme activity, and synaptic function. However, dysregulated zinc has been implicated in abnormal APP processing, Aβ plaques, and neurodegeneration. Understanding the relationship between APP and zinc may provide insight into the underlying mechanisms of Alzheimer’s disease. This study explores the role of appl in response to zinc exposure using Drosophila melanogaster. Wild-type (w¹¹¹⁸) flies and appl null (applᵈ) flies were maintained on diets containing different concentrations of zinc throughout 30 days. Brains were dissected at various intervals and examined for visible signs of neurodegeneration to compare the effects of zinc exposure in flies with and without functional appl, the fly ortholog of APP. Current progress has focused on w¹¹¹⁸ flies to establish how varying levels of zinc affect normal brain anatomy and neuronal health, helping to provide a guideline for comparison. Future work will analyze the applᵈ flies to determine whether the absence of appl changes the neuronal response to zinc exposure and increases susceptibility to neurodegeneration.

Support for this research was provided by the Graduate Student Association (GSA) Research Grant. Amina Ali was supported in part by a grant to the UMBC Meyerhoff Scholars Program from the Howard Hughes Medical Institute (HHMI).

INVESTIGATING THE ROLE OF RNAI SNR1 ON CLIMBING SPEED AND ENDURANCE IN DROSOPHILA MELANOGASTER

Avyukth Kasukurthi¹, Srinika Allala², Dr. Jeff Leips³

¹Marriotts Ridge High School, 12100 Woodford Dr, Marriottsville, MD 21104

²River Hill High School, 12101 Clarksville Pike, Clarksville, MD 21029

³Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

What genetic factors contribute to variation in physical performance? Genome-wide association analyses identified Snr1 as a candidate gene associated with natural variation in age-dependent climbing speed and endurance in Drosophila melanogaster, but its tissue-specific role in regulating locomotor performance remains unclear. This study investigates whether reduced Snr1 expression in the brain and muscle affects climbing speed and endurance. Tissue-specific RNA interference (RNAi) was achieved using the GAL4-UAS system by crossing virgin females from brain- or muscle-specific GAL4 driver lines with males carrying either a UAS-Snr1 RNAi construct or an isogenic control line with normal Snr1 expression. The resulting offspring were tested using climbing assays to measure speed and endurance, allowing comparison between experimental and control groups. We hypothesized that decreased Snr1 expression would decrease locomotor performance by decreasing climbing speed and endurance. Analysis of brain-specific Snr1 reduction revealed no significant effect on locomotor performance (p > 0.05), indicating that decreased expression in the brain alone does not significantly affect these traits. Muscle-specific expression experiments are currently underway and climbing speed and endurance data are being collected to determine whether decreased Snr1 expression in muscle contributes to differences in physical performance. Ongoing analyses will further clarify the tissue-specific role of Snr1 in regulating locomotor performance in Drosophila melanogaster and improve our understanding of the genetic basis of natural variation in physical performance.

Support for this research was provided by the Department of Biological Sciences at the University of Maryland, Baltimore County.

PARTICLE SIZE DISTRIBUTION ANALYSIS OF GARDEN SOILS AT PRINCE GEORGE’S COMMUNITY COLLEGE

Eno Amana, Lynne Heighton

Prince George’s Community College, Department of Natural Sciences, 301 Largo Rd, Largo, MD 20774.

Soil plays a critical role in construction as it serves as the supporting medium for structural loads, making its proper evaluation essential for stability and performance. This study analyzes two soil samples collected at Prince George’s Community College (PGCC) to determine their suitability for construction based on particle size distribution. The samples, identified as Garden 1 (G1) and Garden 3 (G3), were previously used in a comparative study of cultivated and undisturbed soils. G1 represents an amended soil used for agricultural purposes, while G3 was collected directly from compacted ground with no prior cultivation. Sieve analysis was conducted on both samples to determine particle size distribution, and the coefficient of uniformity (Cu) and coefficient of curvature (Cc) were calculated. The results show that G1 is poorly graded (Cu = 3.93, Cc = 1.02), indicating a limited range of particle sizes and reduced structural stability. In contrast, G3 is well-graded (Cu = 7.92, Cc = 1.95), reflecting a wider distribution of particle sizes and improved load-bearing capacity. These findings suggest that while soil amendment may enhance conditions for agricultural use, it may reduce suitability for construction due to changes in particle size distribution. This study highlights the importance of soil classification and gradation analysis in determining appropriate land use.

Support for this research was provided by the Department of Natural Sciences and the Student Research Club at Prince George’s Community College and funding from the NASA MUREP Curriculum Awards (MCA) grant under NASA Award Number 80NSSC23M0194.

INVESTIGATING THE INFLUENCE OF POLYMER COMPOSITION ON THE PROPERTIES OF GEL ELECTROLYTES

Owen Rubin¹, Adim Amanyeiwe¹, Rahuldeb Roy¹, Deepa Madan²

¹Department of Mechanical Engineering, University of Maryland Baltimore County, 1000 Hilltop Circle, Catonsville, MD, 21250

Lithium-Ion batteries are extensively used throughout the world, as they have some of the highest energy capacities and lifespans among other batteries. However, they hold numerous problems, such as overheating that leads to fires, and capacity loss that leads to poor performance. Batteries made with a combination of zinc (Zn) and manganese oxide (MnO₂) have the potential to be an efficient, safer and cheaper alternative. In this study, we investigated the influence of polymer composition on the electrochemical and mechanical properties of gel polymer electrolytes by varying the ratio of poly(vinyl alcohol) (PVA) and polyacrylamide (PAM) in both acidic and alkaline media. We hypothesized that the ideal ratio of PVA: PAM would have an optimal balance between ionic conductivity and mechanical properties. We prepared electrolytes with PVA: PAM ratios of 1:0, 7:3, 1:1, and 3:7, and evaluated the electrochemical properties via electrochemical impedance spectroscopy (EIS), linear scan/sweep voltammetry (LSV), Tafel polarization, chronoamperometry, and overpotential measurements to assess ionic conductivity, electrochemical stability, corrosion behavior, current retention, and polarization behaviors. We found a trade-off between mechanical strength and ionic conductivity with varying polymer composition. Increasing the PAM content enhances the ionic conductivity and reduces the mechanical robustness, whereas increasing the PVA has the opposite effect. In acidic environments, we were able to determine the optimal balance for Zn-MnO₂ battery performance falls between 1:1 and 3:7 ratios. In contrast, the alkaline gel electrolytes show poor electrochemical stability due to the formation of zinc oxide (ZnO) and precipitation of zincate species, promoting electrode degradation and inconsistent data. Further studies focus on optimizing longevity of the gel electrolytes for acidic media. In alkaline medium, more research is required to develop strategies to suppress the parasitic reactions while preserving the role of hydroxide ions.

CHARACTERIZING THE ROLE OF NEUROPEPTIDE Y (NPY) ON DIET-INDUCED OBESITY-LINKED POLYENDOCRINE METABOLIC OVARY SYNDROME.

Dzifa Ameko¹; Marveline Akinola, B.S.¹; Benjamin T. Cole, B.S.¹; Diana M. Elizondo, PhD¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Polyendocrine metabolic ovary syndrome (PMOS) is an endocrine and metabolic disorder characterized by hyperandrogenism, ovarian enlargement, fibrosis, insulin resistance, and irregular menstrual cycles. Obesity is a risk factor for PMOS and is associated with inflammation, extracellular matrix (ECM) deposition, and impaired reproductive function. Macrophages contribute to obesity-associated remodeling, but mechanisms linking obesity to ovarian fibrosis remain poorly understood. Neuropeptide Y (NPY), an immunometabolic mediator regulating inflammation and extracellular matrix remodeling, may contribute to obesity-associated ovarian remodeling relevant to PMOS. We characterized obesity-induced changes in periovarian adipose tissue (POAT), ovarian fibrosis, and macrophage-associated NPY expression using a diet-induced obesity mouse model. Female mice were fed chow or high-fat diets. Metabolic status was assessed by weight, glucose and insulin tolerance tests. POAT remodeling was evaluated using hematoxylin and eosin staining, while ovarian fibrosis was assessed by Picrosirius Red staining and hydroxyproline assays. Flow cytometry was performed to identify macrophage populations expressing NPY within POAT, ovary, and spleen. High-fat diet-fed mice exhibited increased collagen deposition surrounding the ovary and within periovarian adipose tissue (POAT), consistent with obesity-associated fibrotic remodeling, despite no significant differences in ovarian weight, suggesting that extracellular matrix remodeling may precede overt changes in ovarian morphology. Flow cytometric analysis demonstrated that NPY expression was predominantly associated with pro-inflammatory (M1) macrophages rather than anti-inflammatory (M2) macrophages, implicating NPY-expressing inflammatory macrophages as potential contributors to obesity-associated tissue remodeling. High-fat diet-fed mice exhibited a modest increase in NPY+ M1 macrophages within POAT, whereas no significant differences were observed in ovarian or splenic macrophage populations, suggesting that obesity may preferentially promote localized inflammatory responses within the periovarian microenvironment rather than systemic or ovarian immune remodeling. Together, these findings establish obesity-induced ovarian and periovarian remodeling as a foundation for future mechanistic studies investigating macrophage-derived NPY as a regulator of fibrosis and reproductive dysfunction relevant to PMOS.

This work was supported by the NIDDK grant 4R00DK136921-02 and UM FIRST awards to Dr. Elizondo.

THE EFFECTS OF PARTICLE SHAPE, SIZE, AND COMPOSITION ON LIGHT SCATTERING IN THE ATMOSPHERE

Keenan Anwary¹, Pengwang Zhai¹

¹ Department of Physics, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, 21250

Since its inception, NASA has launched over 60 satellites dedicated to Earth observation. The data collected from these satellites is used to support numerous fields of research, including, but not limited to, meteorology, biology, and geography. In order for all of this data to be used effectively, we must interpret satellite measurements with physics models of light-matter interaction, i.e., light scattering theory. The fundamental physics parameters in light scattering include scattering extinction, absorption, and efficiency, phase matrix, and albedo. The Mie scattering model is well researched and documented and is a very valuable tool in showing the effects of refractive index and size parameters in the aforementioned scattering parameters, but it is only applicable to spherical objects. We must then find ways to model light scattering by nonspherical particles, for example, the recently published ZheJiang University’s (ZJU) database of scattering of a large range of particle shapes. The ZJU scattering model uses more sophisticated techniques such as the invariant imbedding T-matrix method, to create a prediction-based model that can handle inhomogeneous groups of aerosol particles and ice. Using this in python models with volume size integration techniques, we are able to model a collection of different sized and shaped particles to see what their scattering parameters would look like. This will allow us to predict the chemical composition, shape, and the size of the aerosols by comparing the data received from space to our predictions. We do this in order to better understand and model Earth’s ever-changing atmosphere.

EXPERIMENTAL ASSESSMENT OF LATENT HEAT ENHANCEMENT IN MEPCM-DIELECTRIC SLURRIES FOR MICROCHANNEL ELECTRONIC COOLING

Yanyu Arias¹, Roshith Mittakolu¹, Damena Agonafer²
¹Department of Chemical, Biochemical and Environmental Engineering, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250
²Department of Mechanical Engineering, 7809 Regents Drive, College Park, MD 20742

Dielectric fluids used in the direct-contact electronic cooling process have inherently low specific heat capacity, limiting their ability to manage high heat fluxes in a compacted microchannel heat sink. Combined with dielectric fluids, microencapsulated phase change materials (MEPCM) offer a promising alternative by storing thermal energy through latent heat absorption during melting. Prior numerical work has shown that dispersing Nextek37D MEPCM in TMC-7300E dielectric fluid can increase the fluid’s effective heat capacity by nearly two orders of magnitude near the phase transition temperature. To quantify latent heat utilization in these slurries, two enthalpy-based metrics were developed: outlet melt fraction (OMF) and effectiveness, which measures the fraction of particles that have undergone phase change at the channel exit and compares the slurry’s performance against the baseline working fluid TMC-7300E.

No experimental study has validated this combined system or confirmed whether the numerically predicted operating regimes, which include underengaged, optimal, and exhausted, match real flow conditions. This work provides the first experimental demonstration of MEPCM-enhanced dielectric cooling in an aluminum microchannel heat sink housed within a Teflon enclosure. The heat sink features a parallel-channel design to maximize convective surface area while maintaining a manageable pressure drop. MEPCM mass fractions of 5-15% are tested under 5-30 W/cm² and inlet velocities of 0.1-0.8 m/s. Bulk temperature rise and pressure drop are measured to quantify latent heat utilization, identify transport limitations, and evaluate deviations from computational prediction. The results will help establish the first experimental validation of OMF and effectiveness for dielectric‑based MEPCM slurries, and identify the optimal combinations for mass fraction and velocity to maximize thermal performance while keeping acceptable hydraulic penalties. The contribution of this work will help advance the understanding of MEPCM-based cooling and provide design guidelines for high-power microelectronic applications.

THE MICROBIOME-GUT-BRAIN AXIS IN ATTENTION DEFICIT HYPERACTIVITY DISORDER: ANALYZING THE IMPACT OF SPECIFIC NUTRITIONAL MODULATOR (CAPSAICIN).

Zainab Arowona, Paria Parto

Department of Natural Sciences, Prince George’s Community College,

Largo, MD 20774.

The human microbiome is a complex ecosystem with a genetic diversity 100 times that of the human genome. A healthy microbiome is defined by high microbial diversity and functional stability. However, industrialized lifestyles, such as the widespread use of antibiotics and decreased fiber consumption, have caused microbiome erosion and microbial imbalance. This literature review examines the Microbiota-Gut-Brain Axis (MGBA) as a critical interface connecting the gut microbiome to Attention-Deficit/Hyperactivity Disorder (ADHD) and how Capsaicin, a bioactive molecule from chili peppers, could regulate some neurotransmitter functions related to the symptoms of ADHD by increasing serotonin and dopamine functions while reducing oxidative stress. A systematic approach was used to synthesize existing research. The search focused on identifying relevant peer-reviewed articles and case studies published between 2010 and 2024. Various databases were utilized, with 80% originating from PubMed and ProQuest. The search terms included a combination of keywords: Capsaicin, Microbiota-Gut-Brain-Axis, ADHD, biological pathways, and central nervous system. The synthesis employed thematic analysis, focusing on the biological pathways that facilitate communication between the gut and the central nervous system and their relevance to ADHD. Twelve papers were reviewed, and 80% suggested that microbial imbalance is a hallmark of ADHD, often shifting tryptophan metabolism toward the kynurenine pathway and inducing neuroinflammation. While animal models suggest CAP can improve ADHD symptoms via the MGBA, there is no clinical evidence that capsaicin can effectively treat ADHD on its own; these animal models point to potential benefits for those with ADHD and a need for more comprehensive gut microbiome studies in humans

Support for this research was provided by the NASA MUREP Curriculum Awards grant

This research is supported by the University of Maryland Louis Stokes Undergraduate Research Fellowship (LSURF) initiative and the Nanoscale Energy and Interfacial Transport (NEIT) Lab.

DIURNAL EVOLUTION OF CLOUD FRACTION AND SURFACE METEOROLOGICAL VARIABLES ACROSS CONVECTIVE REGIMES OVER THE AMAZON RAINFOREST

Tim August¹, Dr. Henrique Barbosa^2

1Department of Earth Sciences, Millersville University, 18 E. Frederick Street, Millersville, PA, 17551
²Department of Physics, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

The Amazon rainforest plays a key role in the global climate cycle, releasing moisture into the atmosphere and influencing rainfall patterns across South America and beyond. Due to the region’s complex atmospheric processes, tracking and understanding convection remains a challenge. Using data from the GoAmazon 2014-15 field campaign, we investigate how cloud properties, surface meteorological conditions, and the large-scale environment differ among shallow, congestus, deep, and mesoscale convective system (MCS) regimes. Ceilometer measurements, surface observations, and ERA5 reanalysis were combined to examine whether large-scale environmental conditions are already established prior to shallow and deep convection. A fog indicator was also developed utilizing both ceilometer and surface-based observations to investigate low-level cloud and fog occurrence. Cloud classifications were assigned based on a pre-existing daily convective regime classification method. Deep and MCS days exhibited higher cloud fraction as well as higher relative humidity, precipitation rate, and dewpoint temperature, while shallow and congestus regimes exhibited higher surface temperatures and boundary layer heights. ERA5 composites indicate deeper regimes are associated with greater total column water vapor and total column liquid water, as well as lower cloud base and planetary boundary layer heights. These ERA5 composites suggest that many large-scale environmental differences are already present on the day preceding shallow and deep convection. This work was supported by the National Science Foundation, grant number 2349543, project title: REU Site: EXPeriments in Earth and Atmospheric Science: Learning Opportunities and Research Experience (EXPLORE).

B

VALIDATION OF MUSCLE-SPECIFIC GAL4 DRIVER LINES FOR AGING RESEARCH IN DROSOPHILA MELANOGASTER

Sariah Rimmer^1,3, Elise Delaporte^1,3, Rose Mupende^1,3, Kallaina Basnet¹, Apolline Nurit², Dr. Jeff Leips¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

²Miami University, 501 High St, Oxford, OH 45056

³Meyerhoff Scholars Program, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

Aging is a continuous biological process that occurs after development and affects virtually all organisms. Although universal, many of the genetic mechanisms underlying aging remain poorly understood. Drosophila melanogaster serves as an excellent model organism for aging research because of its short lifespan, extensively studied genetics, and the conservation of many aging-related genes with humans. Before investigating the function of specific genes involved in aging, it’s important to validate the genetic tools that will be used to manipulate gene expression.

To accomplish this, we used the GAL4-UAS system to evaluate two muscle-specific driver lines (Mef2-Gal4 and Mhc-GAL4). We crossed each driver line with a UAS-Green Fluorescent Protein (GFP) reporter line. GFP is a fluorescent biological molecule originally extracted from a jellyfish. In this context, it allows us to visualize where the GAL4 is being expressed and to confirm that a driver is expressed in the intended tissue. We crossed virgin females from each GAL4 driver to a line carrying the UAS-GFP and an isogenic line without the GFP as a control. Once the offspring emerged, they were examined under a fluorescence microscope to assess GFP expression in muscle tissue in comparison with control flies.

Fluorescence microscopy confirmed strong GFP expression in the muscle tissue of offspring carrying the Mef2-GAL4 driver. In contrast, the Mhc-GAL4 driver produced little to no detectable GFP signal. These results indicate that the Mef2-GAL4 line is an effective muscle-specific driver, whereas the Mhc-GAL4 line exhibits much weaker activity under the conditions tested. Based on these findings, the Mef2-GAL4 driver was selected for future studies.

CASE STUDY ON SMOKE EVOLUTION AND AEROSOL OPTICAL PROPERTIES OVER THE SOUTHWESTERN UNITED STATES

Clara Bax^1,2, Xiaoguang (Richard) Xu¹, Anin Puthukkudy^1

1Department of Physics, Earth and Space Institute, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250
²Department of Atmospheric and Oceanic Sciences, University of Colorado Boulder, 914 Broadway, Boulder, CO 80309

Increased wildfire activity in the southwestern United States affects air quality, human health, and Earth’s radiative balance. NASA’s PACE HARP2 instrument uses multi-angle, multi-spectral polarized observations of reflected sunlight to determine aerosol optical properties. Data from this instrument can provide information about aerosol concentration, size, and composition. Studying smoke evolution can help improve air quality forecasts and public health alerts. This project analyzes four aerosol optical properties— aerosol optical depth, single-scattering albedo, fine-mode fraction, and Ångström exponent—from six wildfires that occurred on June 29, 2026. Given the spatial and temporal similarities, the evolution of aerosol optical properties across the smoke plumes was expected to be similar. Using RGB imagery, six plumes were mapped, with a cross-section and bounding box selected for each. Then, quality-controlled pixels within 7.5 km of the cross-section were studied to understand the optical properties of the individual plumes. Pixels near the fire source are often flagged and excluded because heavy aerosol loading from biomass burning can cause a false cloud mask; thus, only five of the six plumes were analyzed. Each of these five plumes showed aerosol optical depth that peaked near the source and decreased with distance. The Ångström exponent was higher than 1.5 near fire sources, which indicates the presence of fine-mode aerosols, commonly found in biomass burning events. However, differences in the single-scattering albedo and Angstrom exponent were observed between individual plumes, emphasizing variation among individual plumes. These results show that regional smoke evolution does not follow a single pattern, highlighting uncertainties in observing wildfire smoke development and measuring aerosol optical properties. This adds further challenges to quality forecasts and public health alerts.

DESIGNING STRUCTURAL MEMBERS USING BIO-INSPIRED DESIGN PRIMITIVES

Ben Bazarsuren¹; Linnea Hesse, Ph.D.²; Noah Knorr²; Shane Borowski¹; Paris Von Lockette, Ph.D.¹

¹Department of Mechanical Engineering, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

²Department of Biology, University of Hamburg, UHH, Mittelweg 177, Hamburg, Germany 20148

Bio-inspired design is the process in which innovation aims to emulate nature’s patterns, processes, and/or materials to develop efficient solutions to design challenges. Although there exist numerous engineering advancements that have taken inspiration from biological systems, the underlying structural principles that govern how nature organizes fibrous architectures remain insufficiently examined. Therefore, this project investigates whether nature has an innate ability to optimize for a set of structural elements when designing fibrous structures.

Three-dimensional micro-CT scans of plant species belonging to the Smilax and Monstera genera were translated into cross-sectional images and were analyzed using a MATLAB-based image-processing computational pipeline. This pipeline identifies mechanical trends throughout the stems of these plants as they approach a branching node by extracting the following geometric characteristics: fiber area, number of fibers, fiber and slice centroids, stem bending resistance, fiber symmetry, and locations of different-sized fibers within an individual cross section. Statistical analyses were performed to compare trends both within and across species, enabling the identification of recurring structural behaviors.

Preliminary results reveal several recurring structural trends across multiple plant specimens. As branching nodes are approached, fiber area consistently increases while the number of fibers decreases. Changes in fiber distribution potentially suggest species-dependent strategies for balancing bending resistance, with some plants shifting fibers farther from the centroid to increase stiffness, while others redistribute fibers to produce directional reinforcement and asymmetric behavior. These findings indicate that, despite differences in morphology, plants exhibit repeatable organizational patterns that may reflect common structural design principles. Identifying these trends provides a quantitative foundation for advancing topology optimization methods and developing bio-inspired structural designs that leverage nature’s strategies for optimizing mechanical performance.

This research was partially funded by the Meyerhoff Scholars Program alongside the USM LSAMP program, supported by NSF LSAMP Award No. 2207374

CELLS BREATHE TOO: RE-ENGINEERING THE T-FLASK FOR IMPROVED DIFFUSION AND OXYGEN TRANSFER

Callista Bernhardt^1,2,3, Alexandria Slokan¹, Garima Sharma¹, Nathan Gull¹, Michael Tolosa¹, Venkatesh Srinivasan¹, Govind Rao¹

¹The Center for Advanced Sensor Technology, University of Maryland, Baltimore County, UMBC Technology Research Center (TRC) Bldg, Baltimore, MD, 21250

²Howard Hughes Medical Institute, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

³Department of Mechanical Engineering, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

T-flasks are widely utilized in cell culture to conduct in vitro studies. Current T-flasks only allow oxygen to enter through the cap, thus limiting the fill volumes. In recent developments, our lab has commercialized breathable shake flasks that were shown to improve bacterial cell cultures. Extending that innovation to T-flasks, this study uses a modified T25-flask—the bottom side is replaced with silicone to enhance oxygen diffusion for cell cultures that thrive in oxygen-rich environments. We sought to ameliorate the silicone by increasing the surface area with indentations, thereby improving diffusion and oxygen transfer. We evaluated both silicone casting and laser engraving methods and noted the barriers faced in both methods. The most efficient and reproducible method was found to be laser engraving using ideal settings to obtain a pattern without puncturing the silicone. The optimized pattern on a 350 micron thick silicone sheet increased the surface area by approximately 1.5 times compared to an unpatterned silicone sheet. Further feasibility testing of this membrane in T25-flasks is underway. We will be extending this study to T75-flasks and a larger variety of cell cultures in future experiments.

FAILURE-AWARE EPISODIC MEMORY FOR PENETRATION-TESTING AGENTS

Shrikant Bhatnagar¹, Dr. Keke Chen¹

¹Trustworthy and Intelligent Computing Lab (TAIC), University of Maryland Baltimore County, 1000 Hilltop Circle, Baltimore, Maryland, 21250

Large language model (LLM) agents are increasingly applied to autonomous penetration testing, but they show a well-documented weakness: over long, multi-step tasks they lose earlier context, pursue a single line of attack, and over-commit to one strategy instead of exploring alternatives. Prior work has largely addressed this by distributing the workload across multi-agent frameworks, where each agent is assigned to a narrow role within the penetration-testing pipeline. Yet even with these advances, agent performance on standard benchmarks remains limited, and the tendency to persist with an ineffective strategy before pivoting is still pronounced. This project asks whether failure-aware episodic memory can reduce repeated, ineffective exploration and improve the performance of autonomous penetration-testing agents. The premise of the study is that the failures are largely caused by the absence of retrievable, structured knowledge about which techniques have already failed and why. To test this, we reimplement a baseline agent to log its failure trajectories, and evaluate the process based on defined metrics. We then equip the same agent with FAEM (Failure-Aware Episodic Memory) and compare the two conditions on established capture-the-flag benchmarks. If FAEM measurably reduces redundant actions and improves recovery after failure relative to the baseline, it would indicate that failure-structured memory helps address this weakness, with implications extending beyond security agents to any long-horizon agentic task.

Support for this research was provided by the COEIT Student Summer Project Award.

CORRELATION OF EASTERN BLUEBIRD (SIALIA SIALIS) AND TREE SWALLOW (TACHYCINETA BICOLOR) FLEDGING SUCCESS IN PAIRED VERSUS UNPAIRED NEST BOXES

Rose Bodie¹, Nicholas Denton², Kiersten Newtoff^1,2, Kevin Omland¹

¹Department of Biological Sciences – University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

²Department of Chemical and Biological Sciences – Montgomery College, 51 Mannakee St, Rockville, MD, 20850

Eastern Bluebirds (Sialia sialis) and Tree Swallows (Tachycineta bicolor) are secondary cavity-nesting birds that readily use human-constructed nest boxes. Placing these boxes less than ten meters from each other (or “pairing” them) is commonly used to support multiple cavity-nesting species. For example, bluebirds and swallows readily inhabit paired boxes simultaneously. Paired nest boxes (n = 76 boxes) located in Montgomery County, Maryland, were monitored biweekly during the breeding season (March – August). Using data from the 2025 breeding season, we found that bluebird nests that fledged had, on average, 8.28 days of overlap with swallow nests, which was significantly higher than bluebird nests that were unsuccessful, with only 2.17 days of overlap (p = 0.013). Further, swallows had an even greater benefit, with successful nests having an average overlap of 18.23 days while unsuccessful nests had only 4.42 days of overlap (p = 0.002). In paired boxes with only Tree Swallows, nestlings never successfully fledged (n = 13). However, Eastern Bluebirds were able to fledge if not paired with a Tree Swallow (n = 18). These findings suggest that both species benefit from cooperative interspecific nesting in paired boxes, but to a more significant degree in Tree Swallows. A likely factor that explains these data is the possibility that both species defend against a common predator. Another possible factor could be bluebirds being nonmigratory and therefore being better nest site selectors compared to swallows, which are migratory. These findings support the existence of interspecific benefits between songbird species.

Support for this research was provided by UMBC’s Graduate Student Association, Montgomery College, Washington Biologists’ Field Club, and the Maryland Ornithological Society.

DESIGNING STRUCTURAL AND HYDRAULIC MEMBERS USING BIO-INSPIRED DESIGN PRIMITIVES

Shane Borowski¹, Paris Von Lockette, Ph.D.¹, Linnea Hesse, Ph.D.², Noah Knorr²
¹Department of Mechanical Engineering, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250
²Department of Biology, University of Hamburg, Mittelweg 177, Hamburg, Germany, 20148

Understanding the organizational principles behind fibrous plant structures offers valuable insights into biomechanics, microfluidics, and bio-inspired materials design. This study investigates the coupled structural and hydraulic architecture of climbing and hemiepiphytic stems, specifically Smilax and Monstera. These species possess resilient, flexible stems that balance high mechanical loads with continuous, high-efficiency fluid transport. To analyze these multi-functional dynamics, 3D digital models were reconstructed from high-resolution X-ray computed tomography scans of plant stems approaching branching, segmented in ImageJ, and processed through a custom MATLAB framework. A Navier-Stokes-Korteweg fluid formulation was integrated to evaluate local capillary mechanics alongside an orthotropic composite criterion, specifically Tsai-Wu failure theory, to evaluate structural performance under mechanical loading. Our findings reveal a distinct radial organization in both species, where fiber density and vascular bundle concentration increase near the stem periphery, concentrating structural reinforcement where bending stresses are highest. Hydraulically, this spatial arrangement yields a bimodal distribution of capillary performance: smaller peripheral vessels generate peak suction pressures, driving strong initial fluid uptake, while larger central vessels minimize viscous drag, optimizing bulk volume transport. Fluid dynamic simulations demonstrate a transient ascent velocity profile where peripheral micro-channels achieve rapid initial meniscus velocities before reaching viscous equilibrium, complementing the steady-state transport of the core. Structurally, evaluating the cross-sections under transverse bending loads shows that this peripheral fiber concentration significantly elevates the local buckling resistance, preventing localized compressive failure during stem flexure. These results demonstrate that Smilax and Monstera utilize a functionally graded architecture that optimizes the trade-off between hydraulic efficiency and structural integrity. Applications of this research extend to multi-functional additive manufacturing, informing the design of lightweight load-bearing components with embedded microfluidic cooling channels, bio-inspired fluid transport networks, and advanced composite materials.

This research was funded by the UMBC Mechanical Engineering Department.

DETERMINING THE EFFECT OF SET4’S PHD DOMAIN ON GENE REGULATION

Phoenix Bryant¹, Nathaniel Shelton¹, Maraki Negesse¹, Winny Sun¹, Erin Green^1,2

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Cir, Baltimore, MD, 21250

²University of Maryland Greenebaum Comprehensive Cancer Center, University of Maryland, Baltimore, 22 S. Greene Street, Baltimore, MD, 21201

Set4 is a chromatin regulator in Saccharomyces cerevisiae that is involved in the hypoxic stress response. Its ortholog MLL5 in humans is implicated in cancers and neurodevelopmental disorders such as autism spectrum disorder. The SET domain (Su(var)3-9, enhancer-of-zeste, trithorax) is found in proteins that regulate gene expression through catalyzing the methylation of histones. Yet, Set4 is a part of a subfamily of SET domain proteins that is distinguished by their lack of catalytic methyltransferase activity. Of special interest in these Set proteins is their PHD domain, which is canonically a chromatin reader. However, the function of the PHD finger in Set4 has not yet been characterized. Previously, the lab has shown through chromatin immunoprecipitation-qPCR that in Set4 PHDΔ strains, Set4 loses its localization ability to chromatin, demonstrating that the PHD domain is important in this localization. Additionally, in Set4 PHDΔ strains, the expression of the PAU genes, genes involved in cell survival during oxidative stress, are upregulated. The function and role of the PHD finger in gene expression will further be characterized in this presentation using qPCR and other techniques. In doing this, the function of Set4 and its orthologs implicated in disease, such as MLL5, will be better understood.

This investigation was sponsored in part by the U-RISE Program at the University of Maryland, Baltimore County (UMBC), which is supported by the National Institute of General Medical Sciences of the National Institutes of Health (NIH) under Award Number T34GM136497. This research was also sponsored in part by NIH grant R01GM148698 awarded to EMG.

MASS SPECTROMETRY IN QUANTIFYING SALIVARY BIOMARKERS OF ORAL CANCER

Vinnie Bui¹, Chengpeng Chen, Ph.D.¹

¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, Baltimore, MD 21250

Early detection of oral cancer remains challenging because diagnosis relies on invasive tissue biopsy. Human saliva provides a non-invasive diagnostic fluid that contains peptides capable of reflecting pathological changes within the oral cavity. Advances in proteomic technologies have enabled the identification of salivary biomarkers for oral cancer screening. This study evaluates the application of Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF MS) and Liquid Chromatography Mass Spectrometry (LCMS) for identifying salivary peptide biomarkers associated with oral cancer. Published studies employing matched cohorts of healthy controls and oral cancer patients were reviewed to examine experimental workflows, including saliva collection, peptide extraction through centrifugation, mass spectrometric analysis, and biomarker validation using Western blotting. Retrospective analysis demonstrates that MALDI-TOF MS and LCMS can distinguish oral cancer from healthy controls by detecting differences in peptide mass-to-charge (m/z) profiles. Candidate biomarkers identified by mass spectrometry are validated through Western blot analysis, demonstrating differential protein expression between oral cancer and control samples. These proteomic approaches provide insight into molecular changes associated with oral carcinogenesis. This study highlights the relevance of mass spectrometry techniques for discovering salivary biomarkers that may support earlier detection of oral cancer. Further validation using larger samples may apply salivary biomarker analysis into routine diagnostic screening and improve early intervention for patients at risk of oral malignancies.

This work was funded, in part, by the Alex Brown Center for Entrepreneurship Scholar.

INVESTIGATING DIFFERENCES IN RNA SPLICING BETWEEN MAL AND NL4-3 HIV-1 STRAINS

Tazia Burney¹, Max Chen¹, Jake Han², Katelyn Meyer³, Bersabel Tekle¹, Brian Grossman¹, Michael F. Summers^1,4

¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County (UMBC), 1000 Hilltop Circle, Baltimore, MD 21250

²Department of Biology, The University of Maryland, College Park (UMCP), 3972 Campus Dr, College Park, MD 20742

³Department of Biomedical Sciences, Colorado State University (CSU), 711 Oval Drive, Fort Collins, CO 80521

⁴Howard Hughes Medical Institute, Chevy Chase, MD 20814

The Human Immunodeficiency Virus type 1 (HIV-1) is a retrovirus that depletes CD4+T cells, weakening the host immune system and potentially resulting in Acquired Immunodeficiency Syndrome (AIDS). Current antiviral therapies target HIV-1 proteins that have high mutation rates. Therefore, a greater understanding of the HIV-1 structure is necessary to target more conserved regions as they are less susceptible to mutations. Our lab studies a highly conserved region of the HIV-1 RNA genome, the 5′-Leader (5′-L). This region controls many viral functions such as translation, packaging, assembly, and splicing. The HIV-1 5′L exists in equilibrium between two conformations: a monomer (Cap 3G) and a dimer (Cap 1G). In the monomeric conformation, there is an exposed 5′-cap and an exposed Spliced Donor region (SD). The dimeric conformation has a sequestered 5′-cap and a sequestered SD. Studies suggest that the monomeric conformation favors splicing over the dimeric conformation. Our project focuses on investigating the difference in splicing efficiency of these two conformations. To investigate these differences, we designed a series of RNA splicing constructs representing the MAL and NL4-3 HIV-1 strains with four different 5′-cap sequences (1G, GAAG, GGAAG, and 3G). Site-directed mutagenesis and in vitro transcription using T7 RNA polymerase were used to produce the constructs, which were then capped and fluorescently labeled. Splicing assays will be performed using HeLa cell nuclear extracts, providing the necessary splicing machinery, and these products will be analyzed by electrophoretic mobility shift assays (EMSAs). By comparing splicing efficiencies across cap variants and between the MAL and NL4-3 strains, this project aims to determine how RNA secondary structure and 5′-cap identity influence HIV-1 splicing and provide insight into conserved mechanisms that regulate viral gene expression.

Support for this research was provided by the Howard Hughes Medical Institute and NIAID grant #5R01AI150498.

AUTOMATED MONITORING SENSORS FOR PROCESS ANALYTICAL TECHNOLOGY APPLICATIONS IN CELL AND GENE THERAPY

Francheska Burrola^1,2, Garima Sharma², Yunqian Wei^1,2, Michael Tolosa², Mohan Tulapurkar³, Venkatesh Srinivasan², Govind Rao²

¹Department of Biological Sciences, University of Maryland Baltimore County, 1000 Hilltop Circle, Baltimore, Maryland

²Center for Advanced Sensor Technology, University of Maryland Baltimore County, 1000 Hilltop Circle, Baltimore, Maryland

³University of Maryland Baltimore, Baltimore, Maryland.

Cell and gene therapies have the potential to treat cancer, genetic disorders, and other life-threatening diseases, yet their high cost and manufacturing variability limit patient access. A major challenge is the lack of real-time monitoring of key metabolic variables in static cell culture systems. Dissolved oxygen (DO), dissolved carbon dioxide (dCO2), and glucose regulate cellular respiration and metabolic balance, but static G-Rex workflows typically rely on incubator setpoints and intermittent off-line sampling, offering limited insight into dynamic culture behavior. To address this gap, we evaluated 10 to 12 day human embryonic kidney (HEK) cell cultures in static G-Rex systems to assess the feasibility of continuous DO and dCO2 monitoring under production-relevant conditions. HEK cells were selected as a model system due to their well-characterized metabolic and respiratory profiles. Using real-time DO sensors and a rate-based approach to estimate dCO2, we demonstrated proof-of-concept for continuous, 24/7 monitoring over extended culture durations. Continuous monitoring provided more detailed insight into culture dynamics than traditional methods. These findings support integration of process analytical technology strategies into static G-Rex workflows, enabling greater automation, improved process consistency, and more informative batch records to support regulatory review and scalable, affordable manufacturing.

Support for this research was provided by the Ronald E. McNair Scholars Program Summer Research Institute as well as the Alex. Brown Center For Entrepreneurship and Innovation.

C

DEVELOPMENT OF A CHEMICAL MODEL OF ACCIDENTAL AMMONIA RELEASE USING COUPLED EQUILIBRIA
Matteo Chase, Lisa Kelly, Bindu Abraham
Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 2150

Ammonia (NH3) is a major product of the modern chemical industry, being the second most produced chemical worldwide. Its predominant uses are in the production of fertilizers and other industrial chemicals such as nitric acid; although interest in its applications in energy has been steadily growing due to ammonia’s high energy density and potentially carbon-free sourcing. As such, investigations into the scenario of accidental ammonia release have been undertaken to accommodate its ever-growing demand and transport. However, though these studies have uncovered important information on ammonia cloud formation and dynamics, the underlying chemistry associated with such events remains unclear. Therefore, the goal of this study is to develop a simple model of the chemistry of ammonia with carbon dioxide (CO2) and water vapor (H2O) using the framework of chemical equilibrium. To achieve this, existing literature on the reactions between NH3, CO2, and H2O was used, and all reactants were assumed to begin in the gas phase, homogeneously distributed throughout a reaction vessel of known volume. All calculations were performed using Microsoft Excel and used published thermodynamic data to determine each reaction’s equilibrium constant. Using this model, it was predicted that elevated humidity and CO2 concentration promoted the sequestration of gaseous NH3 primarily through the formation of ammonium carbonate, with ammonium bicarbonate also being a significant product. This model has potential applications in informing emergency management of ammonia release scenarios when paired with fluid dynamic models of ammonia clouds. Future directions include analysis of the accuracy of the model using existing data and refinements of the model using kinetic analysis of the reactions.

Support in the poster and abstract development process was provided by the Summer Facilitating Opportunities and Collaborations for Undergraduate Scientists (FOCUS) program.

INVESTIGATING DIFFERENCES IN RNA SPLICING BETWEEN MAL AND NL4-3 HIV-1 STRAINS

Max Chen¹, Jake Han², Tazia Burney¹, Katelyn Meyer³, Bersabel Tekle¹, Brian Grossman¹, Michael F. Summers^1,4

¹University of Maryland, Baltimore County (UMBC), 1000 Hilltop Circle, Baltimore, MD 21250

²The University of Maryland, College Park (UMCP), 3972 Campus Dr, College Park, MD 20742

³Colorado State University (CSU), 711 Oval Drive, Fort Collins, CO 80521

⁴Howard Hughes Medical Institute, Chevy Chase, MD 20814

The Human Immunodeficiency Virus type 1 (HIV-1) is a retrovirus that depletes CD4+T cells, weakening the host immune system and potentially resulting in Acquired Immunodeficiency Syndrome (AIDS). Current antiviral therapies target HIV-1 proteins that have high mutation rates. Therefore, a greater understanding of the HIV-1 structure is necessary to target more conserved regions, as they are less susceptible to mutations. Our lab studies a highly conserved region of the HIV-1 RNA genome, the 5′-Leader (5′-L). This region controls many viral functions such as translation, packaging, assembly, and splicing. The HIV-1 5′L exists in equilibrium between two conformations: a monomer (Cap 3G) and a dimer (Cap 1G). In the monomeric conformation, there is an exposed 5′-cap and an exposed Spliced Donor region (SD). The dimeric conformation has a sequestered 5′-cap and a sequestered SD. Studies suggest that the monomeric conformation favors splicing over the dimeric conformation. Our project focuses on investigating the difference in splicing efficiency of these two conformations. To investigate these differences, we designed a series of RNA splicing constructs representing the MAL and NL4-3 HIV-1 strains with four different 5′-cap sequences (1G, GAAG, GGAAG, and 3G). Site-directed mutagenesis and in vitro transcription using T7 RNA polymerase were used to produce the constructs, which were then capped and fluorescently labeled. Splicing assays will be performed using HeLa cell nuclear extracts, providing the necessary splicing machinery, and these products will be analyzed by electrophoretic mobility shift assays (EMSAs). By comparing splicing efficiencies across cap variants and between the MAL and NL4-3 strains, this project aims to determine how RNA secondary structure and 5′-cap identity influence HIV-1 splicing and provide insight into conserved mechanisms that regulate viral gene expression.

Support for this research was provided by the Howard Hughes Medical Institute and NIAID grant #5R01AI150498.

**THE IMPACT OF NEBULOSA KNOCKDOWN IN THE FAT BODY ON AGE-DEPENDENT IMMUNOCOMPETENCE IN DROSOPHILA MELANOGASTER**

Casey Duhon¹, Evangeline Chen², Pranati Denduluri¹, Zahraa Al Aanizy¹, Jeff Leips¹

¹Department of Biological Sciences, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

²Department of Biology, UMCP, 1204 Biology-Psychology Building, College Park, MD 20742

Support for this research was provided by the Department of Biological Sciences at the University of Maryland, Baltimore County.

UNLOCKING HEALTHIER CELL CULTURES: HOW BREATHABLE SHAKE FLASKS ENHANCE MAMMALIAN CELL CULTURES

George Churchill^1,2, Garima Sharma², Mohan Tulapurkar³, Venkatesh Srinivisan², Govind Rao².

¹Department of Biological Sciences, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

²Center for Advanced Sensor Technology, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

³University of Maryland, Baltimore, MD 21201

Breathable shake flasks are designed to enhance gas exchange by allowing oxygen and carbon dioxide to diffuse through the flask walls. Improved gas transfer have been shown to promote healthier and improved bacterial cell cultures. However, there are minimal data available on the breathable shake flasks with suspension mammalian cell cultures. In this work, we test the feasibility of adapting breathable shake flasks for suspension type mammalian cell culture as they are widely used in commercial production of biopharmaceuticals. We evaluated the performance of breathable shake flasks in comparison to the conventional non-breathable shake flasks using multiple suspension mammalian cell lines. Cell growth was assessed by measuring viable cell counts, while changes in media color were monitored as an indicator of cellular metabolism and metabolic byproduct accumulation across the breathable and non-breathable shake flasks of different volumes. We observed that some mammalian cell lines adapted well to breathable shake flasks and exhibited improved growth, whereas others showed little or no benefit. These findings indicate that the effectiveness of breathable shake flasks is cell line dependent. Future work will focus on optimizing breathable shake flask conditions for a broader range of suspension mammalian cell lines. Improving culture conditions could accelerate cell growth, enhance cell health, and increase the efficiency of biopharmaceutical manufacturing.

BEYOND THE MAP: AI-MEDIATED COUNTERMAPPING AND COMMUNITY NARRATIVES IN BALTIMORE

Talisha Clinton¹, Rebecca Williams²
¹Department of Computer Information Systems, Baltimore City Community College, Baltimore, MD
²Department of Computer Science and Electrical Engineering, University of Maryland, Baltimore County, Baltimore, MD

Maps have long shaped how people understand cities and influence decisions about investment, housing, and opportunity. In Baltimore, federal redlining maps created during the 1930s classified neighborhoods according to perceived lending risk, reinforcing discriminatory housing policies whose effects remain visible in patterns of wealth, environmental quality, health, and neighborhood investment today. While these historical maps have been widely studied, they largely represent the perspectives of government agencies and financial institutions rather than the lived experiences of residents. This research asks: How can AI-mediated countermapping enable Baltimore residents to challenge historical narratives created by redlining maps and contribute community-driven representations of their neighborhoods? Prior research in critical cartography and countermapping has demonstrated that maps reflect social and political power and that participatory mapping can challenge dominant spatial narratives. However, little research has examined how artificial intelligence might support these community-driven mapping practices. Building on this, this project explores whether artificial intelligence can support more inclusive mapping by integrating historical redlining data with residents’ stories, experiences, and visions for their communities. As an ongoing study, the project draws on historical redlining maps, demographic and socioeconomic datasets, and literature on critical cartography, countermapping, participatory mapping, and artificial intelligence to inform the design of an interactive AI-mediated mapping platform. Rather than replacing historical evidence, the proposed platform seeks to complement it with community knowledge and lived experience. If successful, this research will demonstrate how AI-mediated countermapping can help challenge the legacy of discriminatory mapping practices while creating more equitable, community-centered representations of Baltimore’s neighborhoods and providing a model for using AI to advance spatial justice.

This work was supported in part by the UMBC College of Engineering and Information Technology (COEIT) Student Summer Projects fund.

D

INTEGRATED SENSORS AND CONTROLS TO ACCURATELY GAUGE UAV POSITION

Cleon Davis, Ankit Goel

Department of Mechanical Engineering, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore MD 21250

Autonomous unmanned aerial vehicles (UAVs) rely heavily on sensors and control systems to accurately determine their position, orientation, and motion. Relying on a single tracking source can lead to drift, lag, and reduced navigation accuracy. Building on these positioning improvements, the project focuses on orientation and motion tracking by integrating an MPU9250 inertial measurement unit (IMU) into the system. Preliminary testing revealed tracking limitations with the NEO-6M GPS module, prompting its replacement with the higher-performance NEO-M8N GPS for improved positioning accuracy.

The MPU9250 supplied raw accelerometer and gyroscope measurements, which were transmitted from the Arduino to MATLAB through serial communication. By refining the Arduino firmware and adjusting the serial baud rate, a robust, instantaneous data link was established for the integrated sensors. Because raw accelerometer and gyroscope data require mathematical processing to determine the sensor’s orientation, a rotation matrix was applied to the raw IMU measurements to transform the data into roll, pitch, and yaw angles. MATLAB then processed the transformed IMU data to calculate the sensor’s motion and estimate its speed in three-dimensional space.

The processed sensor data will be plotted in MATLAB to visualize the UAV’s roll, pitch, and yaw in real time. This visualization serves as a foundational step for analyzing orientation data, which will subsequently inform the development of control algorithms necessary for autonomous navigation.

Support for this research was provided by Dr. Ankit Goels CELL lab

**THE IMPACT OF NEBULOSA KNOCKDOWN IN THE FAT BODY ON AGE-DEPENDENT IMMUNOCOMPETENCE IN DROSOPHILA MELANOGASTER**

Casey Duhon¹, Evangeline Chen², Pranati Denduluri¹, Zahraa Al Aanizy¹, Jeff Leips¹

¹Department of Biological Sciences, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

²Department of Biology, UMCP, 1204 Biology-Psychology Building, College Park, MD 20742

Support for this research was provided by the Department of Biological Sciences at the University of Maryland, Baltimore County.

CORRELATION OF EASTERN BLUEBIRD (SIALIA SIALIS) AND TREE SWALLOW (TACHYCINETA BICOLOR) FLEDGING SUCCESS IN PAIRED VERSUS UNPAIRED NEST BOXES

Rose Bodie¹, Nicholas Denton², Kiersten Newtoff^1,2, Kevin Omland¹

¹Department of Biological Sciences – University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

²Department of Chemical and Biological Sciences – Montgomery College, 51 Mannakee St, Rockville, MD, 20850

Support for this research was provided by UMBC’s Graduate Student Association, Montgomery College, Washington Biologists’ Field Club, and the Maryland Ornithological Society.

GEANT4 SIMULATION OF Y-90 IMAGING: A DEEP LEARNING FRAMEWORK FOR 3D IMAGE ENHANCEMENT

Ertan Dogan¹, Sidhya Pathak², Victor Sandrin³, Xueying Sun⁴, Ayan Kabaria⁵,

Muhammad Khalid⁶, Sophia Lopez⁷, Abby Nam⁸, Ehsan Shakeri⁹, Harrison Lewis⁹,

Hussam Fateen⁹, Farshad Safavi¹⁰, Ananta Chalise¹⁰, Lei Ren¹¹, Stephen W. Peterson¹²,

Jerimy C. Polf¹³,and Matthias K. Gobbert⁹

¹A. James Clark School of Engineering, University of Maryland, College Park

²Department of Computer Science, University of Virginia ³Department of Neuroscience, University of Arizona ⁴Department of Information Technology and Management, Illinois Institute of Technology ⁵River Hill High School, Howard County, Maryland

⁶Department of Mathematics, Baruch College, City University of New York

⁷Department of Mathematics and Statistics, University of North Carolina at Greensboro

⁸Department of Psychology, Lafayette College

⁹Department of Mathematics and Statistics, University of Maryland, Baltimore County

¹⁰Department of Radiation Oncology, University of Maryland School of Medicine

¹¹Department of Radiation Oncology, Northwestern University

¹²Department of Physics, University of Cape Town, South Africa ¹³M3D, Inc.

Accurate 3D imaging of radiopharmaceutical distribution is critical for personalized dosimetry and monitoring tumor characteristics. This study presents a comprehensive pipeline for simulating and reconstructing Yttrium-90 (Y-90) emissions using a one-stage Compton camera comprising four crystal modules. We developed a Geant4-based Monte Carlo simulation of an isotropically distributed Y-90 source within a synthesized regular tumor, separated from the detector by an air gap. To emulate a tomographic scan, the camera swept in a circular path, capturing the positions and energy depositions of gamma photons from 60 distinct orientations around both the X and Z axes of the centrally placed tumor. These readings were then reconstructed into 3D images using a kernel weighted back projection (KWBP) method. Because this physics-based reconstruction is limited by noise, blurring, and artifacts due to angular sampling restrictions, we developed an advanced 3D U-Net to denoise the reconstructed volumes. The model was trained using a hybrid loss function that combines mean squared error (MSE) and the structural similarity index measure (SSIM) to ensure both pixel-level accuracy and structural fidelity. We evaluated multiple network configurations, ultimately optimizing performance by integrating a dynamic learning rate, an additional residual U-Net block, and online data augmentation during training. Evaluated on a held-out test set, our best-performing configuration effectively mitigated reconstruction noise and artifacts, achieving a hybrid loss of 0.02730, an MSE of 0.00663, and an SSIM of 0.89002. These findings demonstrate that our two-step pipeline, combining physics-based KWBP reconstruction with optimized, data-driven 3D denoising, is an effective method for refining Compton camera images for clinical dosimetry and tumor monitoring applications.

This work is supported by the grant “REU Site: Online Interdisciplinary Big Data Analytics in Science and Engineering” from the National Science Foundation (grant no. OAC-2348755). Co-authors Safavi and Ren additionally acknowledge support by NIH. We acknowledge the computational resources in the UMBC High Performance Computing Facility (hpcf.umbc.edu) and the financial contributions from NIH, NSF, CIRC, and UMBC for this work.

**THE IMPACT OF NEBULOSA KNOCKDOWN IN THE FAT BODY ON AGE-DEPENDENT IMMUNOCOMPETENCE IN DROSOPHILA MELANOGASTER**

Casey Duhon¹, Evangeline Chen², Pranati Denduluri¹, Zahraa Al Aanizy¹, Jeff Leips¹

¹Department of Biological Sciences, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

²Department of Biology, UMCP, 1204 Biology-Psychology Building, College Park, MD 20742

Support for this research was provided by the Department of Biological Sciences at the University of Maryland, Baltimore County.

E

EVALUATING ANTIMICROBIAL ACTIVITY IN SOIL SAMPLES: A COMPARATIVE ANALYSIS OF MICROBIAL RESISTANCE

Darius Eggleston, Michelle Harmon, Sandy Fox-Moon, Ph.D.

Department of Biology, Anne Arundel Community College, 101 College Pkwy, Arnold, MD, 21012

As global rates of antimicrobial resistance increase, the discovery of new antimicrobial compounds remains of critical interest to the field of microbiology. A majority of clinically significant antibiotics originate from natural products or their derivatives. Soil serves as one of the most plentiful reservoirs of antimicrobial-producing microorganisms. The objective of this study was to investigate the microbial diversity of soil samples collected at Anne Arundel Community College (AACC) while evaluating field-sampling methods for identifying potential antimicrobial-producing organisms. Soil sample collection conditions were recorded, and soil samples were cultured and serially diluted onto Tryptic Soy Agar (TSA) and Sabouraud Dextrose Agar (SDA) plates. The colonies were analyzed for cultural characteristics, with four being selected for further analysis based on the presence of potential antimicrobial activity. Gram staining indicated the colonies were Gram-positive bacilli. These findings suggest that the soil surveying methods used can be effective in discovering potential antimicrobial-producing microorganisms. Future research includes verifying the identities of the unknown organisms and isolating and characterizing any potential antimicrobial agents. Additional biochemical testing and a larger sample pool could provide insight for future surveys.

Support for this research was provided by Joseph Quagraine, Team Proteus members (Ally Ayres, Drew Jones, Chris Chambers, and Melody Cheng), and student mentors (Bethany Bayer, Saachi Chopra, and Emily Sawyer).

OPTIMIZING T7 TRANSCRIPTIONAL HOMOGENEITY VIA TEMPLATE MODIFICATIONS AND MUTANT ENZYMES TO APPLY SEGMENTAL LABELING WITH MUTANT ENZYME TGK FOR STRUCTURAL STUDIES

Eunice Ewusie¹, Sridhar Alagar Ramanujam¹, Geoffery Okonkwo¹, Brandon Fonseca¹, Michael Summers, Ph.D^{1, 2}.
¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250
²Howard Hughes Medical Institute, 4000 Jones Bridge Rd, Chevy Chase, MD 20815

Structural analysis on RNA with high-resolution techniques is greatly affected by heterogeneity. Heterogeneity is an inevitable product of transcription by T7 RNA polymerase (RNAP) which can lower the efficacy of structural methods such as Nuclear Magnetic Resonance (NMR) Spectroscopy in structural determination. Literature suggests that T7 RNAP suffers from self-templated run-on which results in RNA with varying nucleotide length at the 3’ end. Our work strives to minimize RNA heterogeneity through: Utilizing template modifications that compromise T7’s ability to induce run-on such as hydrogen bonding or steric modifications and employing T7 mutants, T7AG and T7LG to reduce 3’ heterogeneity. Our template modifications will be tested using in-vitro transcription and their effects further evaluated with NMR. By minimizing T7’s 3’ heterogeneity, a mutant enzyme, TGK DNA polymerase (DNAP), will be used to segmentally label RNA. TGK DNAP and T7 RNAP will work together to synthesize labeled RNA transcripts. Segmental labeling will allow further studies of larger RNA with NMR by reducing spectral crowding. This is made possible by designing a labeling scheme that can be selectively applied to specific sections of RNA. This project aims to not only promote RNA sample purity but to address NMR’s well known size limitations.

Funding for this research was supported by HHMI and the National Institute of Allergy and Infectious Diseases (NIAID, 5R01AI150498). Special thanks to Brandon Fonseca, Dr. Michael Summers, and all who provided guidance in this research.

F

DEVELOPING A METABOLIC MODEL OF A CELL-FREE PROTEIN SYNTHESIS USING BIOSENSORS

Damilola Fapohunda¹, David Garcia¹

¹Department of Chemical, Biochemical, and Environmental Engineering, University of Maryland, Baltimore County, 1000 Hilltop Cir, Baltimore, Maryland, 21250

Cell-Free Protein Synthesis (CFPS) is a synthetic biology method that enables transcription and translation without the need for live cells. Producing proteins in cell-free systems, specifically using the cytoplasmic contents of E. Coli, allows for the bulk production of a wide range of proteins and can be utilized in biotechnology, biosensing, and biomanufacturing. A challenge with using CFPS is a lack of understanding of how cell-free metabolism works, especially with different conditions and reagents. To understand and optimize the cell-free system, we need to develop a metabolic model, starting with glycolysis, the backbone of cellular metabolism.

Traditional analytical methods for monitoring biochemical systems, such as high-performance liquid chromatography, are labor-intensive and do not provide real-time monitoring. Instead, we propose a simpler approach: protein biosensors and high-throughput methodology. Biosensors, such as ApplePy, GreenPy, and Lime, utilize green and red fluorescent proteins to detect pyruvate concentration and pH changes in the system in real time, allowing us to measure metabolism in high throughput. These experiments will be used to train our model, and once a baseline system is characterized, we will be able to add heterologous pathways and expand the cell-free system.

Currently, we are producing the biosensor proteins using CFPS as well as bacterial transformation. In addition, we are developing standard curves for the biosensors by inducing increases in pyruvate and pH in cell-free systems. The preliminary experiments we have done so far demonstrate that the biosensors do fluoresce in the presence of their respective analytes, and moving forward, we plan to expand the data set that we currently have, then begin work on the metabolic model.

The funding for this project is provided by the Department of Chemical, Biochemical, and Environmental Engineering at UMBC.

STRUCTURAL AND FUNCTIONAL ANALYSIS OF THE KH-DOMAIN PROTEIN FLK IN REGULATING ARABIDOPSIS DEFENCE AND DEVELOPMENT

Faridah Folorunsho, Dennis Lee, Geneva Mon, Leah Vrydagh, and Hua Lu
Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250, USA.

Plants are under constant threat from pathogens and pests, leading to billions of dollars in annual agricultural losses. Understanding the coordination between plant defense and development is critical for enhancing the productivity of economically vital crops. The Lu lab recently discovered a role for the canonical flowering regulator, FLOWERING LOCUS K HOMOLOGY DOMAIN (FLK), in pathogen defense. FLK encodes a putative RNA-binding protein featuring three KH domains (KH1, KH2, and KH3). These domains typically span ~70 amino acids and contain a conserved GXXG motif, where the “XX” residues are primarily positively charged and essential for RNA binding. We hypothesize that replacing the GXXG motif with negatively charged aspartic acids (GDDG) will impair RNA binding without destabilizing the KH domain structure. Using site-directed mutagenesis, we generated FLK variants with GDDG substitutions in individual KH domains, both KH1 and KH2 domains, and all three KH domains simultaneously. These constructs were introduced into flk-1 mutants via Agrobacterium-mediated transformation. Following the collection of T0 seeds, we are currently screening for independent T1 transformants. These will be further screened in the T2 generation to identify homozygous transgenic lines. Once established, these plants will undergo a panel of defense and developmental assays to determine how specific KH domain mutations impact FLK’s dual functionality. This research aims to elucidate how FLK coordinates growth and immunity in Arabidopsis, providing insights to engineer crop disease resistance without compromising yield.

This work was partially supported by grants from the National Science Foundation, NSF 2223886 and NSF 1923069, to Hua Lu.

STRUCTURAL CHARACTERIZATION OF RNA DIMERIZATION SITE DOWNSTREAM OF HIV-1 5′-LEADER

Alexandra Foster¹, Ariana Rodriguez¹, Pengfei Ding¹, Michael Summers¹

¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, 1000 Hilltop Cir, Baltimore, MD, 21250

An estimated 40.8 million people worldwide are afflicted by HIV-1 to date. Drug resistance to the current protein targets has become a recurring issue due to the high mutation rate of HIV-1 during reverse transcription. The HIV-1 RNA genome contains many structured regions that play critical regulatory roles; uncovering the structure and function of these RNAs could further our understanding of the virus and inform the development of new treatments. A stretch of RNA sequence downstream of the HIV-1 5’-leader was discovered to be a hotspot for strand transfer during reverse transcription, which directly results in diversification of the viral genome. Our preliminary results suggest that this region also contributes significantly to HIV-1 RNA genome packaging. Literature proposes that this region dimerizes through the formation of an intermolecular G-quadruplex, however, the detailed structure of this dimer in the context of the 5’-leader remains elusive. We seek to characterize the interactions that form the structure of this G-quadruplex dimer, both in isolated RNA fragments and in the context of the full 5’-leader. Native gels have confirmed the formation of a dimer in a 60-nucleotide fragment containing our region of interest. Furthermore, we’ve confirmed that this dimer is dependent on the presence of potassium chloride and suppressed by the presence of lithium chloride, confirming the G-quadraplex dimeric conformation. The identity of the two stretches of guanosine nucleotides from each RNA strand and the resulting tetrad structure will be further probed using nuclear magnetic resonance (NMR) spectroscopy. The biological role of these key guanosine residues and the G-quadruplex structure will be examined using competitive packaging assays. Determining the 3-dimensional structure of this region could provide the detailed mechanism of its role in strand transfer and RNA packaging. Additionally, this work can facilitate the development of RNA-based antiretroviral therapies.

Support for this research was provided by NIH/NIAID #U54-Al1-700660, NIH/NIAID #5R01Al150498, the Howard Hughes Medical Institute (HHMI), and the Arnold and Mabel Beckman Foundation.

THE ROLE OF GPR155 IN MECHANISTIC TARGET OF RAPAMYCIN COMPLEX 1 SIGNALING IN MICE

Tyra Frimpong¹, Emmanuelle Palmieri¹, Laurie Sutton¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, Maryland 21250

Mechanistic target of rapamycin complex 1 (mTORC1) signaling is implicated in numerous neurological pathologies, including addiction and Parkinson’s disease. In the brain, mTORC1 signaling regulates synaptic growth, neuronal activity, and neurogenesis. Previous studies identified GPR155, also known as LYCHOS, as a lysosomal transmembrane protein that mediates cholesterol-dependent activation of mTORC1. Although GPR155 mediates mTORC1 signaling in cell lines, its role in regulating mTORC1 signaling in vivo and in the brain remains unexplored. To investigate this, we used brain tissue from the cerebellum, hippocampus, and dorsal striatum of transgenic GPR155 knockout (KO) and wild-type (WT) mice to perform western blot analysis. mTORC1 activates S6 kinase (S6K) through phosphorylation, making phosphorylated S6 kinase (pS6K) a marker of mTORC1 activity. We compared the ratio of pS6K to total S6K as a measure of mTORC1 activity. By comparing the pS6K-to-S6K ratio between GPR155 KO and WT mice, we can assess whether GPR155 modulates mTORC1 signaling. This study provides insight into the role of GPR155 in mTORC1 signaling and advances our understanding of potential therapeutic targets for neurological disorders.

EVALUATING AIRHARP2 AND HARP2 MULTI-ANGLE MEASUREMENTS DURING NASA’S PACE-PAX CAMPAIGN

Adrita Fuad¹, Brent McBride²

¹Department of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA, 30332

²Department of Physics, University Of Maryland, Baltimore County, 1000 Hilltop Cir, Baltimore, MD 21250

During the NASA PACE-PAX campaign in fall 2024, PACE satellite data was validated using instruments onboard two aircraft and a ship. Onboard the ER-2 high-altitude aircraft was the AirHARP2 polarimeter, an aircraft version of the HARP2 polarimeter onboard PACE. Validation ensures that PACE and HARP2 are collecting data correctly and that the data can be used to make accurate predictions and observations. Our research aims to see how well AirHARP2 can validate HARP2 by identifying similarities and differences in their collected data. First, we processed HARP2 and AirHARP2 Level 1C data from the same overpass on September 26, 2024, over a scene with both desert and lake terrain. Then, we matched the nearest corresponding single pixel between datasets and extracted radiance, polarization, and degree of linear polarization (DoLP) values across four spectral bands. Since AirHARP2’s spatial resolution (~200 m) is far finer than HARP2’s (~5 km), we then averaged AirHARP2 pixels into a superpixel matching HARP2’s footprint, and compared both datasets using RMSE and Spearman correlation across view and scattering angles. HARP2 and AirHARP2 agreed closely on brightness patterns in both the single-pixel and superpixel results, with Spearman correlations above 0.8. However, AirHARP2 consistently read lower in magnitude, an offset not resolved by superpixel averaging. The near-infrared band showed the clearest disagreement, with a negative Spearman correlation for DoLP. A single-pixel comparison of a separate, primarily ocean scene, without superpixel averaging or statistical analysis, showed visually similar patterns. These results suggest that AirHARP2 is an effective validation tool for HARP2, though the near-infrared DoLP outlier may reflect a calibration issue, a geometric or surface effect, or both. Further studies may explore additional scene types, such as clouds, to determine the outlier’s true cause and persistence.

G

INVESTIGATING THE EFFECTS OF AGE AT INDUCTION AND INDUCTION PERIOD ON PROSTATE INFLAMMATION USING A CHRONIC INFLAMMATORY MOUSE MODEL

Shrinidhi Gokulakrishnan¹, Rafe Beckert¹, Alexandria Howe¹, Apurv Rege¹, Charles J. Bieberich¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

Chronic inflammation is a prolonged immune response characterized by continuous infiltration of inflammatory cells into affected tissue. It is estimated to be a risk factor for approximately twenty-five percent of all human cancers. However, there is limited evidence to support the hypothesis that chronic inflammation acts as a precursor to prostate cancer over time. To investigate this, we utilized a doxycycline-inducible mouse model of prostate inflammation (RIG) to characterize how age at induction and induction period influence prostate inflammation and its effects on prostate tissue. Following doxycycline administration, the Tet-On system induces prostate-specific expression of the proinflammatory cytokine Interleukin-1 Beta (IL-1β) and Green Fluorescent Protein (GFP), allowing inflammatory activation to be monitored. RIG mice were induced at varying ages and for different induction periods to evaluate differences in prostate inflammation using GFP as an ex vivo marker. Prostate lobes and distal colon tissues were collected through necropsy for GFP imaging and histological and immunohistochemical analysis. Initial findings indicate a correlation between prostate inflammation and GFP expression. Future studies will evaluate differences in inflammatory activation, tissue effects, induction responses, and immune cell populations among RIG mice of varying ages and induction periods to determine how chronic inflammation influences prostate tissue and its potential implications for prostate cancer biology.

This work was supported by NIH grants 2R01CA200900-06A1 and 1U54CA274370-01.

INVESTIGATING CERAMIDES AND THEIR METABOLITES IN THE BRAIN REGIONS USING MASS SPECTROMETRY IMAGING

David Antonio Granados Martinez, Nimalee Jayasekera, Herana Kamal Seneviratne, PhD

Department of Chemistry and Biochemistry, University of Maryland Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Ceramides are important lipid molecules, and their dysregulation has been associated with different neurological diseases, including Alzheimer’s and Parkinson’s disease. It is known that the brain is a highly heterogeneous organ that contains a range of neuroanatomical regions. Our previous matrix-assisted laser desorption/ionization mass spectrometry imaging analyses indicate that key lipid molecules such as phosphatidylcholines and sphingomyelins exhibit highly region-specific localization. However, the spatial localization of ceramides and their downstream metabolites is yet to be elucidated. In this work, we carried out data mining to determine the localization of ceramides that contain specific acyl chains and their downstream complex metabolites, including hexosylceramides (HexCer), ceramide phosphate (CerP), and ceramide phosphoethanolamine (CerPE) in the brain. Mass spectrometry imaging data were analyzed by SCiLS software, and Allen Brain Atlas was used for brain region annotation. We further generated Intensity Box Plots to compare the abundance of the ceramide molecules across different regions of the brains obtained from mice with different age groups. From these, we observed increased levels of CerP(d18:1/26:0), CerP(d18:1/22:0), HexCer(d18:2/18:0), and HexCer(d18:1/22:0) in the brainstem of adult mice compared with aged mice. However, aged mice exhibited higher levels of CerPE species such as CerPE(d14:0/18:0), CerPE(d14:1/19:0), and CerPE(d16:1/18:1) in the cerebellum and brainstem area. In sum, our findings indicate that ceramides and their metabolites exhibit region-specific localizations in distinct brain regions. The work is currently underway to increase the coverage of the detection of ceramide molecules and their metabolites.

H

INVESTIGATING DIFFERENCES IN RNA SPLICING BETWEEN MAL AND NL4-3 HIV-1 STRAINS

Jake Han¹, Max Chen², Tazia Burney², Katelyn Meyer³, Bersabel Tekle², Brian Grossman², Michael F. Summers^2,4

¹The University of Maryland, College Park (UMCP), 3972 Campus Dr, College Park, MD 20742

²University of Maryland, Baltimore County (UMBC), 1000 Hilltop Circle, Baltimore, MD 21250

³Colorado State University (CSU), 711 Oval Drive, Fort Collins, CO 80521

⁴Howard Hughes Medical Institute, Chevy Chase, MD 20814

The Human Immunodeficiency Virus type 1 (HIV-1) is a retrovirus that depletes CD4+T cells, weakening the host immune system and potentially resulting in Acquired Immunodeficiency Syndrome (AIDS). Current antiviral therapies target HIV-1 proteins that have high mutation rates. Therefore, a greater understanding of the HIV-1 structure is necessary to target more conserved regions, as they are less susceptible to mutations. Our lab studies a highly conserved region of the HIV-1 RNA genome, the 5′-Leader (5′-L). This region controls many viral functions such as translation, packaging, assembly, and splicing. The HIV-1 5′L exists in equilibrium between two conformations: a monomer (Cap 3G) and a dimer (Cap 1G). In the monomeric conformation, there is an exposed 5′-cap and an exposed Spliced Donor region (SD). The dimeric conformation has a sequestered 5′-cap and a sequestered SD. Studies suggest that the monomeric conformation favors splicing over the dimeric conformation. Our project focuses on investigating the difference in splicing efficiency of these two conformations. To investigate these differences, we designed a series of RNA splicing constructs representing the MAL and NL4-3 HIV-1 strains with four different 5′-cap sequences (1G, GAAG, GGAAG, and 3G). Site-directed mutagenesis and in vitro transcription using T7 RNA polymerase were used to produce the constructs, which were then capped and fluorescently labeled. Splicing assays will be performed using HeLa cell nuclear extracts, providing the necessary splicing machinery, and these products will be analyzed by electrophoretic mobility shift assays (EMSAs). By comparing splicing efficiencies across cap variants and between the MAL and NL4-3 strains, this project aims to determine how RNA secondary structure and 5′-cap identity influence HIV-1 splicing and provide insight into conserved mechanisms that regulate viral gene expression.

Support for this research was provided by the Howard Hughes Medical Institute and NIAID grant #5R01AI150498.

EVALUATING ANTIMICROBIAL ACTIVITY IN SOIL SAMPLES: A COMPARATIVE ANALYSIS OF MICROBIAL RESISTANCE

Michelle Harmon, Darius Eggleston, Sandy Fox-Moon, Ph.D.

Department of Biology, Anne Arundel Community College, 101 College Pkwy, Arnold, MD, 21012

INVESTIGATING THE EFFECTS OF COPPER EXPOSURE ON APP GENE FUNCTION AND ALZHEIMER’S DISEASE RELATED NEURODEGENERATION IN DROSOPHILA MELANOGASTER

Noelle Harris¹; Justine Anne Guevarra¹; Fernando Vonhoff, Ph.D.¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

In humans, the APP gene encodes for amyloid precursor protein (APP), which is a transmembrane protein involved in communication between neurons, calcium signaling, and neuronal stability. When processed abnormally, it can overproduce amyloid beta (Aβ) and result in plaques associated with Alzheimer’s disease. Alzheimer’s disease is an example of neurodegenerative disorder that is characterized by vacuoles in the brain. Recent research suggests that APP functions beyond amyloid beta production and may play an important role in maintaining neuronal health. Also, APP has a copper binding domain, which is important for proper protein folding, maturation, and processing. Interestingly, studies have shown that copper binding to APP can reduce the production of amyloid beta. Using Drosophila melanogaster as a model organism, wild-type ( w¹¹¹⁸) and appl null (applᵈ) genotypes were examined for neurodegeneration following exposure to 3 copper concentrations mixed into their food over 30 days. Flies were dissected at varying time points to observe anatomical changes in the brain. Studies investigating the relationship between APP and copper homeostasis suggest that disruption of either pathway may contribute to neurodegeneration. Therefore, understanding the relationship between APP and copper may provide further insight into the mechanisms underlying Alzheimer’s disease.

Support for this research was provided by the Meyerhoff Scholars Program and the Graduate Student Association.

DETERMINING MECHANISMS THAT REGULATE USP15 LEVELS IN OVARIAN CANCER CELLS

Valarie Hlover, Ayokunnumi Ogunsanya, Achuth Padmanabhan, PhD

Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Ovarian cancer is the most lethal gynecologic cancer affecting hundreds of thousands of women worldwide. Factors that contribute to the high lethality include late diagnosis, lack of early biomarkers, ineffective therapeutic methods, and high recurrence rates. To improve clinical outcomes, it is important to identify factors that drive ovarian cancer progression and determine the mechanisms that regulate them. Through gene expression dataset, we identified a deubiquitinase, USP15, whose levels are elevated in ovarian cancer. Furthermore, high USP15 levels are correlated with poorer survival outcomes. There are very few studies exploring the role of USP15 in ovarian cancer progression and how its levels are regulated in ovarian cancer cells. Our project aims to identify the pathways that regulate USP15 levels and identify novel upstream regulators of USP15. To do this, we fused both USP15 isoforms (USP15-1 and USP15-2) with mCherry and cloned both into a lentiviral vector. We have obtained positive clones for both constructs. In the future, we will stably express the mCherry-tagged isoforms in ovarian cancer cells and transfect them with siRNAs targeting Kinases and Deubiquitinases. Proteins that affect USP15 levels will be further characterized to determine the mechanisms by which they regulate USP15 levels. This project could aid in the development of treatment methods that target USP15, thereby increasing the survival rates of ovarian cancer patients.

Support for this research was funded by the Padmanabhan Start-Up.

SUCROSE DENSITY GRADIENT CENTRIFUGATION VERSUS ANION EXCHANGE FOR EXOSOME SAMPLE PURIFICATION

Sydney Hofstetter, Melanie J. Nelson, Dr. Luis Pinzon Herrera, Dr. Jorge Almodovar

Department of Chemical, Biochemical, and Environmental Engineering, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Exosomes can be effectively isolated from samples using sucrose density gradient (SDG) ultracentrifugation, but ion exchange methods may result in higher levels of exosome sample purity. Exosomes are extracellular vesicles secreted by cells with biomedical applications that include drug administration and immunotherapy. They are small, non-reproducing cellular compartments with varying proteomic and nucleic acid profiles. Purified exosome samples allow us to use samples for downstream applications. We explore the use of ion exchange methods to remove debris and contaminants from exosome-containing samples and compare these methods to SDG centrifugation to determine which yields the highest sample purity levels. To use ion exchange, we utilize anion exchange resin columns and allow the samples to flow through via administration with a syringe. For SDG ultracentrifugation, we manually create a sucrose density gradient in a tube to be ultracentrifuged, which allows sample components to migrate to their respective density levels. Protein content of density layers has been evaluated via BCA assays, and sample purity has been verified via nanoparticle tracking analysis.

This investigation was sponsored by the U-RISE Program at the University of Maryland, Baltimore County (UMBC), which is supported by the National Institute Of General Medical Sciences of the National Institutes of Health under Award Number T34GM136497. We thank Dr. Ryan Pearson at the University of Maryland, Baltimore for assistance with this project.

I

DEVELOPMENT OF DIRECT INK WRITING TECHNIQUES OF A FE₃O₄-SILICONE BOUND MAGNETIC SLURRY USING A FE₃O₄ VOLUME FRACTION OF 20%

Chidube Ibe¹, Levi Niaba¹, Paris Von Lockette¹

¹Department of Mechanical Engineering, University of Maryland, Baltimore County (UMBC) 1000 Hilltop Circle Baltimore, MD 21250

Permanent magnets (PM) are a valuable class of material with many uses in magneto-mechanical devices. However, the PMs currently used in these applications are rigid materials making them difficult to integrate into environments where elasticity is preferred such as soft robotics and biomedical implants.Our goal is to use Direct Ink Writing (DIW) techniques on varying compositions of Magnetoactive Elastomer (MAE) to manufacture flexible magneto-elastic composites with high volume fractions of magnetic materials that also hold shape when printed.While the work at this stage only intends to fabricate flexible magnetic composites that hold shape. We have devised a composition of MAE that can be used as a curable ink for paste-extruding 3D printers following the work of other researchers. This is done by mixing up to 20%v/v carbonyl Fe3O4 particles (<1micron) treated with Polyethylene Glycol 400 (PEG) into a Sylgard 184 binder. These compositions will be printed using a Hyrel 3D printer outfitted with a paste extruder, following a template made with Onshape and sliced in PrusaSlicer. Their shape fidelity is then compared using a Structural Similarity Index Measure (SSIM) obtained through Fiji image processing. We have currently been able to produce a 10%v/v Fe3O4 MAE, resulting in a MAE of high uncured viscosity. A test disk made with a MAE consisting of 1% v/v Fe3O4 was cured, displaying exceptional shape retention, but a low pre-curing viscosity. Analysis of the MAE composition revealed that nozzle sizes between 9-14 mm are ideal for its printing due to the diameter size of the Fe3O4 particles. Currently we have concluded that a 3D-printed PM can be formed into a consistent geometry, and that higher %v/v Fe3O4 MAEs will need to have their initial viscosity managed through PEG treatment or alternating the polymer used.

Support for this research was provided by the UMBC Mechanical Engineering Department and College of Engineering and Information Technology

INVESTIGATING THE ROLE OF PERIOD ON CLIMBING SPEED AND ENDURANCE IN DROSOPHILA MELANOGASTER

Nyle Imtiaz², Dr. Jeff Leips¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

²River Hill High School, 12101 Clarksville Pike, Clarksville, MD 21029

The period (per) gene, a central regulator of the circadian rhythm in both Drosophila and humans, has been identified as a candidate gene affecting physical performance in Drosophila melanogaster through a genome-wide analysis. Per encodes PER proteins, which function as key transcription factors that control the molecular circadian clock. The circadian clock regulates locomotor activity rhythms, sleep, and metabolism, all of which contribute to physical performance. A prior experiment examined the effect of per underexpression in the brain, finding no significant impact on climbing speed and endurance. This suggested that per expression in other tissues may play a larger role in physical performance. To investigate this, we examined the effect of per underexpression in muscle tissue on climbing speed and endurance. The GAL4-UAS system, a gene manipulation tool, was utilized, allowing for targeted gene expression in the muscle tissue. Virgin females of the muscle-specific GAL4 line were crossed with males of a UAS responder line and a control line. We then measured the climbing speed and endurance of the offspring of these crosses. This experiment aims to determine whether per expression in muscle tissue significantly influences physical performance. We hypothesize that the level of expression of the per gene in muscle tissue directly correlates to the speed and endurance of the flies. The results of our experiment may provide insight into how the human orthologs of per influence physical performance in humans. This could enhance personalized healthcare as it relates to physical decline and related conditions. If the per gene is found to affect physical performance in Drosophila melanogaster, further research is needed to determine the underlying mechanisms behind this effect. Future research should investigate the gene’s effect on other tissues and age groups.

Support for this research was provided by the Department of Biological Sciences at the University of Maryland, Baltimore County.

J

VECTOR-FREE SYNTHETIC CIRCULAR SUPERCOILED DNA (SCSDNA) ENABLES RAPID CELL-FREE PRODUCTION FOR THERAPEUTIC DEVELOPMENT WORKFLOWS

Leen Jawhar¹, Xiang Li², Charles J. Bieberich²

¹Department of Biological Sciences, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

High-yield circular DNA production is currently constrained by the use of bacterial plasmid vectors. These plasmid vectors require time-consuming purification protocols and introduce non-functional genetic sequences, such as origins of replication and selectable markers, that can diminish expression efficiency. This study describes a scalable, enzymatic process designed to generate milligram-to-gram quantities of vector-free, synthetic circular supercoiled DNA (scsDNA) consisting solely of the target open reading frame (ORF). High-yield linear DNA branches are generated overnight via Rolling Circle Amplification (RCA) and cleaved using Type IIS restriction enzymes to create offset, overlapping precursors. These precursors form a heteroduplex that is circularized using a thermostable ligase to produce the final scsDNA. The resulting scsDNA significantly improves transfection and protein expression in a cell-free system. This method facilitates rapid, field-deployable production of medical countermeasures, such as radiation antidotes and therapeutic agents, by eliminating the need for bacterial plasmids.

Support for this research was provided by the National Institutes of Health (R41GM154562) and The Arnold and Mabel Beckman Foundation

INVESTIGATING SEX-DEPENDENT BEHAVIORAL AND MOLECULAR EFFECTS OF CHRONIC ETHANOL EXPOSURE IN MICE AT THE NUCLEUS ACCUMBENS-HIPPOCAMPUS PATHWAY

Noah Johansson¹, Sarah Snider¹, Nick Anderson¹, Tara LeGates PhD^1,2

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

²Department of Pharmacology and Physiology, University of Maryland School of Medicine, Baltimore, MD 21201

Approximately 33% of people suffering from Alcohol Use Disorder (AUD) also suffer from Major Depressive Disorder (MDD). This comorbidity increases relapse rates and makes each disorder difficult to study individually. Although we are aware of this linkage, we aren’t sure exactly what happens in the brain. Many behavioral features of AUD and MDD are mediated by the brain’s reward circuitry, especially the Nucleus Accumbens (NAc), which acts as the central hub for the reward system. The ability of NAc synapses to undergo synaptic plasticity and Long Term Potentiation (LTP), a long-lasting strengthening of synaptic connections, is crucial for adaptive behaviors and reward learning. According to previous work, alcohol reduces LTP at these synapses. This project aims to characterize the behavioral and synaptic effects of chronic ethanol exposure in male and female mice. We will assess reward-related, social, anxiety-like, and anhedonia-like behaviors before and after ethanol exposure to determine how these behavioral changes correspond to alterations in synaptic signaling within the NAc. After behavioral testing and ethanol exposure we will process tissue samples from the NAc and then run western blots. Western blot analysis will be used to quantify expression of L-type voltage-gated calcium channels and NMDA receptor subunits (NR1, NR2A, and NR2B), key regulators of synaptic plasticity. Because previous studies suggest sex-specific mechanisms of synaptic plasticity, both male and female mice will be examined. We hypothesize that alcohol-induced deficits in plasticity are due to changes in protein expression. We further predict that males will exhibit greater reductions in NMDA receptor expression, whereas females will show greater alterations in L-type calcium channel signaling, reflecting sex-dependent molecular adaptations within reward circuitry. Understanding how chronic ethanol exposure alters behavior and synaptic signaling may provide insight into the mechanisms underlying AUD and its comorbidity with depression.

Support for this research was provided by the NSF IOS2402645 grant and the Meyerhoff Scholars Program.

INVESTIGATING CONSERVATION OF THE HIV-1 CORE ENCAPSIDATION SIGNAL

Aaron Johnson^1,2; Sofia Shevchenko^1,2 ; Zeynep Akan^1,2; Vianney Tanifor^1,2; Horasa Ji²; Michael Summers^1,2

¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

²Howard Hughes Medical Institute, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

Funding for this research is supported by the Howard Hughes Medical Institute (HHMI) and NIAID #5R01AI150498

K

EXPERIMENTAL PROMPT GAMMA SCATTER CLASSIFICATION AND REORIENTATION FOR PROTON BEAM IMAGE RECONSTRUCTION

Ayan Kabaria¹, Muhammad Khalid², Sophia Lopez³, Abby Nam⁴, Ertan Dogan⁵,

Sidhya Pathak⁶, Victor Sandrin⁷, Xueying Sun⁸, Ehsan Shakeri⁹, Harrison Lewis⁹,

Hussam Fateen⁹, Farshad Safavi¹⁰, Ananta Chalise¹⁰, Lei Ren¹¹, Stephen W. Peterson¹²,

Jerimy C. Polf¹³,and Matthias K. Gobbert⁹

¹River Hill High School, Howard County, Maryland

²Department of Mathematics, Baruch College, City University of New York

³Department of Mathematics and Statistics, University of North Carolina at Greensboro

⁴Department of Psychology, Lafayette College

⁵A. James Clark School of Engineering, University of Maryland, College Park

⁶Department of Computer Science, University of Virginia

⁷Department of Neuroscience, University of Arizona

⁸Department of Information Technology and Management, Illinois Institute of Technology

⁹Department of Mathematics and Statistics, University of Maryland, Baltimore County

¹⁰Department of Radiation Oncology, University of Maryland School of Medicine

¹¹Department of Radiation Oncology, Northwestern University

¹²Department of Physics, University of Cape Town, South Africa

¹³M3D, Inc.

Prompt gamma rays emitted by cells during proton beam radiotherapy can be detected by Compton cameras and used to reconstruct images to verify accurate dose delivery and Bragg peak location. However, image reconstruction is complicated by misclassified scattering events because of the limited temporal resolution of the cameras. To address this, machine learning models were trained on simulated datasets generated using Geant4 and MCDE and implemented through the Big Data REU Integrated Development and Experimentation (BRIDE) platform to classify and filter scatter events for more accurate image reconstruction. After identifying the best-performing models, predictions using a deep Fully Connected Network (FCN) were created for 24 cases of experimental phantom triples data. Images were reconstructed on CORE using Kernel Expectation-Maximization (KEM) and Simple Back Projection (SBP) algorithms on both raw and repaired data, then compared to assess the accuracy of the Bragg peak estimation and noise reduction. SBP reconstructions had more complete beam reconstructions and Bragg peak profiles than KEM reconstructions, however still included noise in 2-D beam profiles. KEM 1D profile reconstructions improved after a reorientation script was implemented to transform the scatter data before classification, which reoriented the data such that it was consistent with training, and reversed before reconstruction.

INVESTIGATING THE ROLE OF RNAI SNR1 ON CLIMBING SPEED AND ENDURANCE IN DROSOPHILA MELANOGASTER

Avyukth Kasukurthi¹, Srinika Allala², Dr. Jeff Leips³

¹Marriotts Ridge High School, 12100 Woodford Dr, Marriottsville, MD 21104

²River Hill High School, 12101 Clarksville Pike, Clarksville, MD 21029

³Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

Support for this research was provided by the Department of Biological Sciences at the University of Maryland, Baltimore County.

DEVELOPMENT OF A VAGINA-ON-A-CHIP MODEL WITH A PHYSIOLOGICALLY RELEVANT OXYGEN GRADIENT TO STUDY BACTERIAL VAGINOSIS

Gurleen Kaur¹, Garima Sharma^2,3, Venkatesh Srinivasan², Corine Jackman Burden¹

¹Department of Chemical, Biochemical and Environmental Engineering, University of Maryland Baltimore County, 1000 Hilltop Circle, Baltimore, Maryland, USA 21250

²Center for Advanced Sensor Technology, University of Maryland Baltimore County, TRC Building, 1000 Hilltop Cir., Baltimore, Maryland 21250

³Department of Biological Sciences, 1000 Hilltop Cir., Baltimore, Maryland 21250

Bacterial vaginosis (BV) affects roughly a third of women globally and is the most widespread recurrent vaginal condition for women of reproductive age. Women with BV are more prone to health risks, including an increased susceptibility of contracting STIs such as HIV, Gonorrhea, Chlamydia and poses a higher risk of pregnancy complications such as preterm birth. BV is defined by a depletion of protective lactobacilli species and an overgrowth of anaerobic bacteria like Gardnerella vaginalis. While BV is recognized as an imbalance of protective microbiota, mechanisms driving the onset of BV remain unclear. Traditional in vitro models lack the ability to replicate the physiologically-relevant microenvironment of the hypoxic vaginal lumen, essential for growth of oxygen-sensitive and -tolerable microbial communities alongside host cells. To overcome this barrier, we are developing a vaginal microphysiological system (VMS) (also known as vagina-on-a-chip) to establish necessary oxygen conditions. We have designed features for a VMS prototype that integrates gas channels, diffusion barriers, and oxygen sensor patches for continuous measurement and regulation of local oxygen concentrations within the chip. Successful optimization of the VMS will enable the investigation of host-microbe interactions and their implications on women’s health for prevention of infections, diagnostics, and treatment.

Support for this research was funded by the Commercialization and ENTR Research (CENTRE) Proposal Award.

INVESTIGATING THE EFFECTS OF MATERNAL CARE ON DNMT3AB EXPRESSION ACROSS DEVELOPMENT STAGES IN AN AFRICAN CICHLID

Onkarpreet Kaur, Cheng-Yu Li, Ph.D

Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

Maternal care and early maternal-offspring interactions play critical roles in shaping offspring development and behavior across many vertebrate species. Although these effects may persist into adulthood, the molecular mechanisms through which early-life social experiences produce long-lasting developmental changes remain poorly understood. Epigenetic regulation, particularly DNA methylation, provides a potential mechanism by which environmental experiences alter gene expression without changing the underlying DNA sequence. The de novo DNA methyltransferase gene dnmt3ab, which is involved in establishing new DNA methylation patterns during development, is therefore a promising candidate for investigating how maternal care influences developmental programming. This study uses the African cichlid (Astatotilapia burtoni) to examine how maternal care affects the developmental expression of dnmt3ab. Females of this species naturally mouthbrood their offspring for approximately 14 days, providing an opportunity to manipulate early maternal care while minimizing genetic differences among offspring. Offspring receiving either full maternal care or complete maternal deprivation will be examined during the larval, juvenile, and adult stages. In situ hybridization will be used to characterize and compare spatial patterns of dnmt3ab expression across developmental stages and maternal-care treatments. We hypothesize that dnmt3ab expression will decrease as offspring develop from larvae to adults and that offspring receiving full maternal care will exhibit higher dnmt3ab expression than maternally deprived offspring. By determining how maternal care influences the developmental regulation of a DNA methylation-related gene, this study will provide insight into the epigenetic mechanisms through which early-life social environments shape vertebrate development and behavioral plasticity. These findings will also establish a foundation for future studies investigating the downstream molecular and neural pathways linking maternal care, DNA methylation, and offspring behavior.

I would like to thank Dr. Cheng-Yu Li for mentorship and guidance, as well as members of the Li lab for their feedback and assistance with fish room husbandry. The Li lab is supported by the NIH UM-FIRST grant.

DFT STUDIES OF GaS, A 2D QUANTUM MATERIAL

Baran Khosravynia, Yussuf Fasasi, Zeev Rosenzweig, and Joseph W. Bennett
Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

The materials used for sensing and quantum information technology need to exhibit high surface area and well-defined electronic and optical properties. Two-dimensional (2D) layered chalcogenides, like GaS, have a high surface area and are electronically tunable. GaS has already shown sensitivity to changes in surface chemistry and distinct optical properties useful in quantum information science. These properties depend on the crystal structure and impurities of GaS, so we use density functional theory (DFT) calculations to model the structural and electronic effects of vacancies and defects on single-layer GaS. DFT is an efficient and computationally inexpensive way to probe materials at the atomistic level, and here we use it to compare total energies of modified GaS supercell surfaces.

Funding is provided by Merck Data Sciences in the Life Sciences Program and the College of Natural and Mathematical Sciences at UMBC

L

QUANTITATIVE CHARACTERIZATION OF THE TEMPORAL RELATIONSHIP BETWEEN ACTIN DYNAMICS, SEPTATION, NUCLEAR DIVISION, AND HYPHAL GROWTH DURING EARLY DEVELOPMENT OF ASPERGILLUS NIDULANS USING LIVE-CELL FLUORESCENCE MICROSCOPY

Jiyon Lee¹, Walker Huso¹, Mark R. Marten, Ph.D.¹

¹Department of Chemical, Biochemical, and Environmental Engineering, University of Maryland, Baltimore County, 1000 Hilltop Circle, MD 21250

Filamentous fungi are important organisms in biotechnology, agriculture, and human health, yet the cellular mechanisms controlling their growth remain incompletely understood. Hyphal extension requires coordinated regulation of actin organization, vesicle trafficking, cell wall remodeling, septation, and nuclear division. However, the temporal relationships between these processes are difficult to characterize due to limitations in imaging living filamentous fungi, which rapidly grow in three dimensions and are difficult to maintain within a consistent imaging plane. To address this challenge, the Marten Lab developed a live-cell flow chamber imaging system that enables long-term observation with fluorescence microscopy of Aspergillus nidulans during early development. This project aims to use quantitative live-cell imaging to characterize the coordination between actin dynamics, hyphal growth, septal formation, and nuclear division. Using a dual-fluorescent Aspergillus nidulans strain expressing LifeAct-GFP and RFP-tagged nuclear markers, time-lapse fluorescence microscopy was performed under controlled growth conditions. Kymograph-based analysis was used to quantify hyphal extension rates and visualize dynamic cellular behaviors over time. After optimization of growth conditions and imaging parameters, reproducible hyphal growth and fluorescence imaging were successfully achieved across multiple experimental trials. Preliminary observations revealed that septal formation may coincide with changes in hyphal growth patterns and nuclear organization, suggesting that septation may serve as a developmental marker associated with transitions in fungal growth. Based on these observations, ongoing experiments will quantify the temporal relationship between septal formation, nuclear division, actin localization, and hyphal extension using live-cell imaging and image-based analysis. This work will provide a quantitative framework for understanding how filamentous fungi coordinate cellular organization during development. By defining the relationship between cytoskeletal dynamics and developmental progression, this study will contribute to a broader understanding of fungal growth regulation and provide insight into processes relevant to fungal physiology, pathogenicity, and future antifungal treatments.

Support for this research was provided by the National Science Foundation Awards (2006189 and 2527369) and the UMBC Undergraduate Research Award (URA).

DFT STUDIES OF THE 2D QUANTUM MATERIAL InSe

Kylie Leeks, Yussuf Fasasi, Zeev Rosenzweig, Joseph W. Bennett

Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

The materials needed for sensing and quantum information technology need to exhibit high surface area and well-defined electronic and optical properties. Two-dimensional (2D) layered chalcogenides, like InSe, have a high surface area and are stable in air. InSe has already shown sensitivity to changes in surface chemistry and distinct optical properties useful in quantum information science. These properties depend on the crystal structure and impurities of InSe, so we use density functional theory (DFT) calculations to model the structural and electronic effects of vacancies and defects on single layer InSe. DFT is an efficient and computationally inexpensive way to probe materials at the atomistic level, and here we use it to compare total energies of modified InSe.

Support for this research was provided by the Purdue University QSI/QuPIDC

OPTIMIZING ASSEMBLY OF THE HIV-1 GAG-RNA NUCLEATION COMPLEX

Samuel Li^1,2, Eric Schreiter-Zavala², Alexandra Foster², Ariana Rodriguez², Pengfei Ding², Mike Summers^2

1Department of Biology, William & Mary, 200 Stadium Drive, Williamsburg, VA, 23185
²Howard Hughes Medical Institute and Department of Chemistry and Biochemistry, University of Maryland Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

HIV-1 is a retrovirus affecting more than 40 million individuals worldwide, with more than half a million dying from AIDS-related complications per year. While existing anti-retroviral therapies (ARTs) can prevent and treat HIV, drug resistance due to the virus’s high mutation rate necessitates developing new and long-lasting drugs. One possible non-protein drug target is the HIV 5’ leader, a highly conserved and structured regulatory region including the first 350 nucleotides of the HIV genomic RNA. It has been implicated in key processes in the viral life cycle, including the selective packaging of the HIV RNA into progeny virions by the Gag polyprotein. Previous work from our lab identified more than 20 Gag binding sites on the 5’ leader. However, the detailed mechanism of selective packaging of the RNA genome remains poorly understood. We hypothesize that the multiple Gag binding sites on the 5’ leader promotes initial assembly of Gag into hexamers, and the resulting protein-RNA complex is then used as a nucleus to catalyze assembly of thousands of Gag proteins into a full immature capsid shell. Our ultimate goal is to resolve the structural basis of this Gag/5’ leader nucleation complex. Currently, our lab has achieved a low-resolution structure of the complex through cryo-EM single-particle analysis (SPA). SPA requires tens of thousands of homogeneously formed complexes, which has been a bottleneck in our analysis. To resolve additional structural details, we have identified critical factors to optimize our Gag/5’ leader complex assembly conditions and to achieve high-quality, homogeneous complexes suitable for SPA. Determining the 3-dimensional structure of the nucleation complex will allow us to understand the detailed mechanism of HIV genome packaging and potentially lead to the development of a new class of RNA-targeted HIV drugs.

This research was supported in part by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under Award Number 5R01AI150498. Dr. Mike Summers is an Investigator of the Howard Hughes Medical Institute. Samuel Li was supported as a Cech Fellow summer research student by HHMI.

EFFECT OF ANCHOR KNOCKDOWN ON INSULIN PRODUCTION IN THE INSULIN PRODUCING CELLS OF DROSOPHILA MELANOGASTER

Nicolás LLobet¹, Emmanuelle Palmieri¹, Fernando Vonhoff¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000

Hilltop Circle, Baltimore, MD, 21250

When the hormone insulin is faultily expressed in humans and other mammals, it leads to many disorders such as Type 1 and 2 Diabetes. The latter affects an estimated 38 million people in the U.S. alone. The extent of this problem begs research into what genes control insulin levels, so as to further medical intervention. Mechanisms involving insulin are highly conserved in fruit flies, making them a strong model system to better understand its function. Specifically, we wanted to see how the alteration of a particular gene in flies, anchor, would affect the IPCs of flies. Anchor was chosen because previous work in our lab revealed that knockdown of anchor in the IPCs alters IPC-associated behaviors, suggesting that anchor may have an effect on insulin production in these cells. To see how anchor affects insulin levels in the IPCs, we compared fluorescence intensity of IPCs stained for Drosophila Insulin-like Peptide 2 (DILP2) in both the anchor knockdown flies and genetic controls. What we found was a significant decrease in DILP2 levels in the IPCs of the anchor knockdown flies compared to those of the controls. This suggests that anchor plays a key role in the regulation of insulin levels.

This research was supported by the Department of Biological Sciences and the Vonhoff Lab at UMBC.

**SUBMERGED CULTURE VISCOSITY (RHEOMETRY) OF ASPERGILLUS NIDULANS KINASE DELETION STRAINS**

Peter Lombardo¹, Alexander G. Doan^1,, Steven D. Harris², Mark R. Marten^1,

1Department of Chemical, Biochemical, and Environmental Engineering, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250
²Department of Plant Pathology, Entomology, and Microbiology, Iowa State University, 2433 Union Drive, Ames, IA 50011

The model filamentous fungus, Aspergillus nidulans, is used in bioprocess research for its genetic tractability and favorable culture conditions. Aspergilli and other filamentous fungi are highly relevant to the bioprocess industry due to their versatility as well as their ability to produce high titers of product (e.g., recombinant enzymes). High culture viscosities, however, pose challenges in mass transfer of oxygen and nutrients, limiting yields. Marten Lab research on A. nidulans seeks to understand how cell-wall regulatory kinases influence fungal morphology and rheological behavior. Previous work in the Marten Lab suggests that the deletion of specific cell-wall regulatory kinases increases hyphal fragmentation without affecting growth rate, which we hypothesize will reduce culture viscosity, thus overcoming mass transfer limitations and substantially improving yields. In this study, vane impeller rheometry is used to measure the dynamic viscosity of submerged cultures of A. nidulans kinase deletion strains. This project aims to conduct rheological screening of a targeted subset of kinase knockout strains that experience diminished cell-wall integrity compared to the control strain, and identify kinase knockouts that significantly influence culture viscosity. Consequently, this project seeks to inform genetic strategies to simultaneously improve bioprocess efficiency and reduce energy demands.

Funding support provided by an Undergraduate Research Award from the UMBC Division of Undergraduate Academic Affairs, NSF Award no. 2006189, NSF Award no. 2527369

M

INVESTIGATING THE ROLE OF ANCHOR IN FEEDING BEHAVIOR, USING A DROSOPHILA MELANOGASTER MODEL

Hareem Malik¹, Emmanuelle Palmieri¹, Fernando Vonhoff, Ph.D¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Cir., Baltimore, MD 21250

Feeding behavior is crucial to study because it is a primary survival mechanism, and its dysregulation can lead to many human eating disorders. Feeding relies on a complex network of membrane-embedded sensors and receptors across several cell populations, many of which are conserved between mammals and flies. For instance, flies have insulin-producing cells (IPCs), a cluster of 14 neurosecretory cells, similar to mammalian pancreatic beta cells, that also maintain metabolic homeostasis. While IPCs have been linked to feeding, the specific molecules and mechanisms that underlie this behavior remain unclear. One highly expressed protein in IPCs is anchor, an integral membrane protein that contains a G-protein coupled receptor (GPCR) domain and a transporter domain that acts as a nutrient sensor for mTORC1 signaling, a major cell growth pathway highly conserved across species. Previous research in the lab demonstrated that pan-neuronal knockdown of anchor reduces overall food intake. To better understand which specific neuronal population explains this feeding phenotype, this project utilized a model with a knockdown of anchor in the IPCs. We evaluated sucrose consumption using a CApillary FEeding (CAFE) assay, which tracks a fly’s sucrose intake from a liquid solution-filled capillary tube. The total sucrose amount consumed was measured at the beginning and at the end of a 24 hours feeding period. Measured data showed that IPC anchor knockdown flies consumed lower amounts of sucrose compared to their respective controls. The data also showed that female anchor knockdown flies had no significant differences in overall food consumed, leading to the conclusion that anchor in IPCs regulates feeding in a sex-dependent manner. Future research will study the specific molecular mechanism underlying this phenotype as well as investigate the sex-dependent effects of anchor on feeding behavior.

This research was supported by the Department of Biological Sciences and the Vonhoff Lab at UMBC.

UTILIZATION OF WORLD AND HEAD-CENTRIC REST-FRAMES IN A HIGH-SPEED VR ENVIRONMENT

Reuben Marshall Jr¹, Janelle Clark¹

¹Department of Mechanical Engineering, University of Maryland Baltimore County, 1000

Hilltop Circle, Baltimore, MD, 21250

Flight companies rely on virtual reality (VR) simulators for pilot training due to their immersive qualities, though they pose a higher risk of visually induced motion sickness, or simulator sickness, when compared to traditional forms of training. VR-based interventions to reduce motion sickness can increase user comfort, which often leads to increased periods of VR immersion and subsequent training. Rest frames (RFs), a primary intervention type, allow VR users to better interpret motion by creating a virtual object linked to the user’s presence in the virtual environment. In this work, we will compare world-centric (car-hood, visor) and head-centric (face-shield, goggles, cap) reference frames against no-intervention using the Simulator Sickness Questionnaire (SSQ) and Slater-Usoh-Steed (SUS) Questionnaire after navigating a racetrack. Results show the impact of different implementations of reference frames on motion sickness and user immersion.

Supported by the Department of Mechanical Engineering at UMBC.

INVESTIGATING THE ROLE OF MYELOID-DERIVED TRANSGLUTAMINASES IN DIET-INDUCED METABOLIC DYSFUNCTION

Izabella Maria Marzano¹, Benjamin T. Cole, BS.¹, Marveline I. Akinola, BS.¹, and Diana M. Elizondo, PhD.¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Obesity was recognized as a major driver of metabolic dysfunction and was strongly influenced by excess nutrient intake. Diet‑induced obesity provided a robust model for examining how chronic high‑fat feeding altered body composition and systemic metabolic regulation. These changes included increased adiposity, altered energy balance, and impaired glucose handling, which together contributed to broader metabolic disturbances.

Myeloid cells were known to play an important role in tissue inflammation and metabolic homeostasis, and transglutaminase 2 (TGM2) expressed in myeloid cells had been implicated in adipose tissue remodeling. However, the physiological contribution of myeloid‑derived TGM2 to the development of obesity and its metabolic consequences had remained unclear.

This study aimed to define the role of myeloid‑derived TGM2 in basal and diet‑induced metabolic states using a LyzM‑Cre conditional knockout mouse model. Wild‑type (WT) and Tgm2 conditional knockout (cKO) mice were studied under chow and high‑fat diet conditions. Longitudinal measurements of body weight were collected to monitor weight gain, and glucose and insulin tolerance tests were performed to assess systemic metabolic responses. Dual‑energy X‑ray absorptiometry (DEXA) scans were used to quantify body composition, including lean and fat mass, and tissue weights were obtained for epididymal white adipose tissue (eWAT) and liver.

Across chow and high‑fat diets, WT and cKO mice exhibited broadly similar body weight trajectories and glucose and insulin tolerance responses, indicating that myeloid TGM2 silencing did not markedly alter overall weight gain or whole‑body glycemic control. In chow‑fed mice, lean and fat mass were largely comparable between genotypes. Under a high‑fat diet, DEXA and tissue analyses revealed differences in fat mass, eWAT mass, and liver mass between WT and cKO mice, suggesting genotype-dependent effects on adipose and hepatic remodeling during obesity.

By comparing WT and cKO mice under lean and diet‑induced obese conditions, this work characterized how myeloid‑derived TGM2 contributed to physiological adaptations to excess nutrient intake and provided a foundation for linking immune cell-associated enzymes to obesity‑related metabolic outcomes.

This work was supported by the NIDDK grant 4R00DK136921-02 and UM FIRST award to Dr. Elizondo.

INVESTIGATING THE IMPACT OF RADIOSONDE MEASUREMENTS ON AEROSOL RETRIEVALS

Jediah Matthews¹, Nicolas Oliveira¹, Henrique Barbosa¹

¹Department of Physics, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

A radiosonde is an instrument carried into the atmosphere by a weather balloon to measure atmospheric conditions, such as temperature, pressure, and density of the troposphere and lower stratosphere. These measurements are needed for the computation of the molecular backscatter and extinction used in LiDAR (Light Detection and Ranging) inversion methods to obtain aerosol optical properties. However, radiosonde measurements are not always available at the same time as the LiDAR measurements. In this work, we investigate whether climatological averages can be used to replace simultaneous radiosonde data. We perform a comparison of aerosol retrievals using a monthly mean profile, a seasonal mean profile, and an annual mean profile with lidar measurements at the Amazon rainforest. The retrieved aerosol backscatter and extinction coefficients from each method were then compared to determine whether there were any significant differences between them, keeping the other input parameters for the inversion constant. The results showed that the differences between the three methods were minimal, suggesting that monthly, seasonal, or annual mean radiosonde profiles can be used when actual radiosonde measurements are unavailable without significantly affecting the reliability of the retrieved aerosol properties.

Support for this research was provided by the UMBC undergraduate research award (URA).

CHARACTERIZING THE EFFECTS OF DUST DEPOSITION ON PHYTOPLANKTON COMMUNITIES WITH PACE AND VIIRS SATELLITE DATA

Natalie McCourt¹, Andrew Sayer^2,3, Ivona Cetinic^3,4

¹Department of Computer Science and Electrical Engineering, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, Maryland 21250

² University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, Maryland 21250

³Ocean Ecology Laboratory, NASA Goddard Space Flight Center, 8800 Greenbelt Rd, Greenbelt, MD 20771

⁴ Morgan State University, 1700 E Cold Spring Ln, Baltimore, MD 21251

Understanding the drivers of phytoplankton distribution, diversity, and physiology is complex yet essential for improving our understanding of the carbon cycle. Dust storms deposit massive quantities of iron into the ocean, a key and limiting micronutrient for phytoplankton globally. This study evaluated the relationship between the occurrence and intensity of significant dust events and changes in phytoplankton physiology and community composition, focusing on the Atlantic off the coast of Patagonia and over regions of the Mediterranean Sea. Conducted to increase the scope of current understanding of how dust events impact ocean ecosystems, we performed time series analyses of remote sensing products including dust aerosol optical depth (AOD), sea surface temperature (SST), sea surface temperature anomaly (SSTA), chlorophyll concentration, phytoplankton biomass (carbon_phyto), fluorescence line height (nflh), and PACE’s phytoplankton community composition product (MOANA), which estimates the relative abundance of picophytoplankton groups. Lag correlation analyses were used in order to characterize the temporal relationship between dust events and biological responses, identifying time lags for which correlations were strongest and statistically significant. Preliminary results indicate a 10-15 day lag between dust event occurrence and peaks in chlorophyll concentration, phytoplankton biomass, chl:carbon ratio, and nflh for both locations, as well as significant negative correlations between SST and biomass. Results also indicate that picoeukaryotes, and Synechococcus, and Prochlorococcus exhibit distinct temporal responses to dust events, suggesting group-specific sensitivities to atmospheric dust deposition.

INVESTIGATING DIFFERENCES IN GAG BINDING BETWEEN MAL AND NL4-3 HIV-1 STRAINS

Katelyn Meyer^1,4, Bersabel Tekle², Jake Han³, Max Chen², Tazia Burney², Brian Grossman², Michael F. Summers^2,4

1Department of Biomedical Sciences, Colorado State University, 1350 Centre Ave Rm H120, Fort Collins, CO 80523
²Department of Chemistry and Biochemistry, UMBC (University of Maryland, Baltimore County), 1000 Hilltop CircleBaltimore, MD 21250
³Department of Biological Sciences, University of Maryland, College Park (UMCP), 3972 Campus Dr, College Park, MD 20742
⁴Howard Hughes Medical Institute, Chevy Chase, MD 20814

HIV-1 is a retrovirus that depletes CD4+ T cells, weakening the immune system. Identifying more conserved regions of the virus, which are less prone to mutation, could point toward more durable therapeutic targets. The 5’end of the HIV-1 RNA genome, which directs several essential viral processes, including genome packaging. HIV-1 selectively packages two copies of its unspliced genomic RNA into virions through interactions between Gag and structured elements within this region. Studies of the NL4-3 strain revealed a minimal packaging region, the core encapsidation signal (CES), which folds into a tandem three-way junction, in which splice donor residues form long-range base pairs rather than an independent hairpin. Whether this fold is conserved across other HIV-1 strains is still unclear. MAL and NL4-3 are both group M strains but differ in genetic subtype and RNA structural conformation. The MAL strain’s CES appears to be more dynamic, sampling multiple conformations rather than settling into the single dominant structure seen in NL4-3, raising the question of how this variability affects Gag recognition and packaging efficiency.

This work asks whether NL4-3 and MAL RNAs differ in Gag binding, and whether Gag preferentially recognizes one RNA when both are present. Viral RNA was synthesized from both strains, and full length Gag proteins were purified for binding assays. Using fluorescently labeled RNAs, competitive binding experiments will be performed by mixing both RNA with Gag to determine whether one RNA is bound preferential or forms higher order complexes more readily. Gel study assays will be used to distinguish the two RNAs and directly compare their Gag interactions side by side.

Linking these structural differences to binding behavior should clarify how CES shapes genome packaging in HIV-1. More broadly, this approach could help build an understanding of how the CES structure shapes Gag-RNA interactions in different HIV-1 strains.

LATENT SEX DIFFERENCES IN CHRONIC STRESS INDUCED EFFECTS ON SYNAPTIC PLASTICITY

Deeya Mistry^1.2, Nicholas Anderson¹, Cherub Willmott¹, Noah Johannsen^1,2, Tara A. LeGates^1,3
¹Department of Biology, University of Maryland, Baltimore County, 1000 Hilltop Circle,
Baltimore, MD, 21250
²Howard Hughes Medical Institute
³Department of Pharmacology and Physiology, University of Maryland School of Medicine, 108 N. Greene Street, Baltimore, MD, 21201

Females have a higher susceptibility to developing stress-related psychiatric disorders such as depression. However, there is limited knowledge regarding the sex-specific mechanisms that arbitrate behaviors relating to these disorders. This further complicates our understanding and treatment of mechanisms for these debilitating conditions. Motivated behaviors are critically regulated by the bidirectional regulation of the strength of hippocampus-nucleus accumbens (Hipp-NAc) synapses. Increased synaptic strength is associated with reward-related behaviors. Chronic stress weakens synapses and disrupts synaptic plasticity, changes associated with anhedonia. Potentiation of the synapses is rewarding, and the potentiation is mediated by sex-specific mechanisms.

We used a chronic variable stress (CVS) paradigm to determine whether stress induces changes in Hipp-NAc synaptic plasticity and behavior similarly in males and females. Both sexes showed stress susceptibility as measured by stress-induced changes in physiology and behavior, verified by Novelty Suppressed Feeding (NSF). Upon CVS completion, the NAc was extracted from the mice, and the synaptosomes were isolated using a two-filter method. Western blot analysis was then performed to evaluate the impact of chronic stress on synaptic protein expression. Future directions include examining the expression of synaptic proteins in neurons differentiated from patient-derived neurons to offer a translational perspective to this research. In conclusion, this work is significant to further understand the synaptic reasoning as to why there are sex-specific differences associated with stress-induced psychiatric disorders. This is especially important to further the development of personalized medicine to treat stress-induced psychiatric disorders.

This research was supported by the University of Maryland, Baltimore County (UMBC) Startup Fund, Howard Hughes Medical Institute grant (52008090), U-RISE Program at the University of Maryland, Baltimore County (UMBC), which is supported by the National Institute Of General Medical Sciences of the National Institutes of Health under Award Number T34GM136497.

SYNTHESIS AND PURIFICATION OF ENTEROVIRUS D68 NEGATIVE-STRAND CLOVERLEAF RNA FOR CRYSTALLOGRAPHIC STUDIES

Emily Mitroka, Bethel G. Beyene, and Deepak Koirala

Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, Baltimore, MD 21250

Enterovirus D68 (EVD68) is a positive-sense (+), single-stranded RNA virus in the Enterovirus genus of the Picornaviridae family. EVD68 infection primarily causes mild to severe respiratory illness in children, including bronchiolitis and pneumonia, and in some cases, acute flaccid myelitis (AFM), a serious neurological condition. Study of this virus has become increasingly important, as no vaccine or antiviral therapeutics exist yet. Understanding enteroviral genome structure could reveal potential targets for antivirals and therapies.

The enteroviral genome consists of a single open reading frame flanked by 5′ and 3′ UTRs; domain I of the 5′ UTR – the 5′ cloverleaf (5′ CL) – promotes replication. Our lab recently determined the crystal structures of 5′ CLs, revealing a highly conserved H-type four-way junction that mediates interactions with replication-related proteins. The (-) strand, synthesized from the (+) strand template during replication, serves as the template for (+) strand synthesis and is proposed to form a similar cloverleaf at the (-) 3′ end (3′ CL). This project aims to reveal structural and functional insights into the EVD68 3′ CL and to test its interaction with a replication-related protein using X-ray crystallography. The method requires highly pure and concentrated RNA samples, which are prepared by in vitro transcription from DNA templates that are in turn prepared through PCR amplification. After designing and ordering DNA templates, we optimized PCR conditions. Using gel electrophoresis to visualize PCR products, we found that adjusting the DNA polymerase extension time improved amplification across several conditions tested, yielding high-quality DNA templates for subsequent transcription. While the synthesis and purification of RNA samples using these DNA templates are ongoing, future studies aim to grow crystals of these RNAs and determine their three-dimensional structures by X-ray crystallography. Once the structure is solved, we will investigate structure-guided interactions of the EVD68 3′ CL with replication-linked proteins.

This work was supported in part by NSF CAREER award 2236996 to D.K. and by NIH T32 grant GM158458 to B.G.B.

PERSONALIZED LONGITUDINAL ORAL HEALTH MONITORING USING IMU AND AUDIO SENSORS FROM SMARTWATCHES

Kodilinye Mkpasi¹, Dong Li¹, Nirmalya Roy^1,2, Mary Elizabeth Aichelmann-Reidy³, Anuradha Ravi²

¹Department of Computer Science and Electrical Engineering, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Department of Information Systems, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

³University of Maryland School of Dentistry, University of Maryland, 650 West Baltimore Street, Baltimore, MD, 21201

Maintaining optimal oral hygiene is critical across all age groups, yet individuals often neglect certain areas, apply excessive force, or disproportionately focus on specific regions while brushing. Such suboptimal behaviors can lead to plaque accumulation, enamel wear, gingival inflammation, and progressive periodontal conditions. Without timely clinical evaluation, these conditions may worsen and result in irreversible damage. At the same time, dental professionals face challenges in assessing treatment effectiveness between scheduled visits, as they lack continuous insight into patients’ daily brushing practices and disease progression. This research seeks to augment dental care with AI-driven tools that monitor routine brushing techniques and support a personalized modeling framework for adaptive, real-time oral health assessment. By integrating longitudinal brushing analytics with automated plaque and gingival index prediction, the system aims to provide clinicians with actionable insights to evaluate treatment efficacy, detect early signs of disease progression, and deliver more informed, data-driven patient care. With the widespread adoption of smartwatches, there is significant potential to use these devices for monitoring toothbrushing behavior and technique. This research explores the use of consumer-grade smartwatches to classify brushing motions and dental regions using IMU and audio data. Key challenges include data sampling issues, labeling IMU/audio data manually, and inter-personal variability. Linear interpolation was applied to align heterogeneous signals, and built-in orientation filtering supported stable angle estimation. Prior generic models suffered from inter-user variability in brushing speed and biomechanics, achieving only ~60% accuracy. To improve performance, we propose a two-phase personalized framework combining lightweight neural models with user-specific adaptation layers. A global model is trained on pooled data and fine-tuned per user while freezing shared feature extractors, enabling accurate and edge-deployable inference.

Support for this research was provided by the U-RISE Program at UMBC, and funding from the

National Institutes of Health under Award Number T34GM136497.

A CASE STUDY ON THE SHAPE FIDELITY AND ACTUATION OF HETEROGENOUS LATTICES OF MAGNETOACTIVE ELASTOMER CURIOUS SHAPES

Deylon Moy Joseph, Paris von Lockette

Department of Mechanical Engineering, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Magnetoactive elastomers (MAEs) are composite materials consisting of micro- or nanometer-sized magnetic particles embedded into a compliant elastomeric matrix. They belong to the class of smart materials because their macroscopic properties can be changed by the application of magnetic fields. Additionally, under the influence of magnetic fields, their initial geometries conform to the external field, creating magnetostriction and/or other actuation modes. Utilizing Comsol Multiphysics, this begins by performing a convergence study of single unit cells of topology optimized loading cases for axial, bending and shear case MAEs to discover what level of mesh refinement is needed to ensure mesh-independent results. This level of mesh refinement will then be used to explore our first goal, simulating the deformation vs. field response of these cases in increasingly larger homogeneous periodic arrays. Secondly we will combine the mesh-independent unit cells across deformation modes into a heterogenous structure to explore their ability to produce mixed-deformation when combined. Successful mixed deformation will be considered achieved when a metric of 85% or higher shape fidelity is achieved when immersed in a uniform magnetic field. The motivation behind this study is to discover required mesh-refinement characteristics as well as an advance in MAE research through the combining of three loading case specific unit cells that were created utilizing topology optimization in heterogeneous arrays. Tension, shear and bending unit cells are speculated to be used in tandem to create complex MAE structures that could be capable of biomedical applications through biomimetic movement to perform locomotion, peristalsis, and minimally invasive tools that can be present in delicate anatomical structures and environments.

VALIDATION OF MUSCLE-SPECIFIC GAL4 DRIVER LINES FOR AGING RESEARCH IN DROSOPHILA MELANOGASTER

Sariah Rimmer^1,3, Elise Delaporte^1,3, Rose Mupende^1,3, Kallaina Basnet¹, Apolline Nurit², Dr. Jeff Leips¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

²Miami University, 501 High St, Oxford, OH 45056

³Meyerhoff Scholars Program, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

N

FORMALDEHYDE STIMULATES CYTOSOLIC CA²⁺RESPONSE IN OLFACTORY NEURONS INDEPENDENT OF CANONICAL ODOR DETECTION PATHWAY

Favour Nwogu¹, Iskandar Nassar¹, Farhan Augustine¹, Tatsuya Ogura¹, and Weihong Lin¹

¹Department of Biological Sciences, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

The mechanism by which formaldehyde (FA) stimulates olfactory sensory neurons (OSNs) is not fully understood, despite its ubiquity in human society and healthcare settings. OSNs integrate a diverse array of external environmental stimuli and serve as a primary site of contact for inhaled toxicants, including aldehydes, which are common byproducts of electronic cigarette vapor. Prior studies have shown that cellular responses to formaldehyde stimulation can reliably occur in both neurons and non-neuronal cells, suggesting that FA-induced signaling involves intracellular calcium mobilization. To determine whether this response extends to neuronal cells, we examined FA-evoked signaling in primary OSNs. We hypothesized that FA activates OSNs through a signaling pathway independent of the canonical cyclic nucleotide-gated channel alpha-2 odor transduction cascade and instead relies on the mobilization of intracellular Ca²⁺ stores.

To test this hypothesis, we isolated OSNs from adult CNGA2-KO mice, a transgenic animal line in which the gene which encodes function of the cyclic nucleotide-gated channel (CNG)A2 odor transduction pathway is knocked out. We loaded the cells with Fura-2 AM for ratiometric Ca²⁺ imaging. Acute FA stimulation across multiple e-cigarette-equivalent concentrations elicited robust intracellular Ca²⁺ transients in CNGA2-KO OSNs compared to the cytosolic baseline, both in the presence of extracellular Ca²⁺ (0.02% FA: 30.6%±11.8%; 0.08% FA: 75.2%±9.5%) and in the absence extracellular Ca²⁺ (0.02% FA: 37.9%±11.9%; 0.08% FA: 78.8%±34.2%)(Mean±std). The persistence of FA-evoked signaling independent of canonical odor detection raises the distinct possibility that repeated aldehyde exposure may induce calcium excitotoxicity which could lead to apoptosis. Together these findings indicate the presence of a non-canonical mechanism of FA-mediated OSN activation and establish a framework for discovering physiological consequences of environmental and vaping-related aldehyde exposure.

This project is supported by the ATIP pilot Grant and START award to WL.

CLASSIFYING MINERAL DUST USING AEROSOL PROPERTIES FROM PACE OBSERVATIONS

Ashley Nguyen^1,2, Zhibo Zhang²

¹College of Computing, Data Science, and Society, University of California, Berkeley, CA 94720

²Department of Physics, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Mineral dust aerosols play an important role in modulating atmospheric processes and the Earth’s energy budget by scattering and absorbing shortwave and longwave radiation. Accurately distinguishing them from other aerosol types using satellite observations remains challenging because of their variability in size and shape and atmospheric mixing. In this study, we examine dust-concentrated and dust-smoke mixture events from regions surrounding Northern Africa and North America. To begin, data products are queried from NASA’s PACE (Plankton, Aerosol, Cloud, ocean Ecosystem) instruments: Ocean Color Instrument (OCI) and Hyper Angular Rainbow Polarimeter (HARP2). Comparing these instruments allows us to answer how geographic regions, atmospheric composition, or viewing geometry report differences in optical properties, as well as which retrieval variable best separates dust-like from non-dust scenes. OCI Unified Aerosol Algorithm Aerosol Optical Properties (UAAAER) are then collocated to match a common grid with HARP2 Fast Multi-Angle Polarimetric (FastMAPOL). Since HARP2 FastMAPOL only has ocean retrievals, this study focuses on events over marine regions. We then developed a K-means clustering algorithm to quantify aerosol clusters, explain internal properties, and identify feature importance. An initial run of the OCI UAAAER and HARP2 FastMAPOL cluster analysis using Angstrom Exponent and Aerosol Optical Depth features for the Saharan Dust Transport resulted in silhouette scores of 0.538 and 0.457, respectively. While both instruments have three moderately well-defined clusters, OCI UAAAER showed clearer distinctions among clean sky, severe sandstorms, and moderate smoke/dust loading, supported by the centroid points. HARP2 FastMAPOL clusters grouped more closely, with a clear indication of intense sandstorms but vagueness in other potential aerosols. This study is still underway, with opportunities to add more temporal and spatial data containing a mixture of dust, smoke, and other aerosols, validate with in-situ observations, and select additional optical features to better distinguish aerosol types.

DEVELOPMENT OF DIRECT INK WRITING TECHNIQUES OF A FE₃O₄-SILICONE BOUND MAGNETIC SLURRY USING A FE₃O₄ VOLUME FRACTION OF 20%

Chidube Ibe¹, Levi Niaba¹, Paris Von Lockette¹

¹Department of Mechanical Engineering, University of Maryland, Baltimore County (UMBC), 1000 Hilltop Circle, Baltimore, MD 21250

Support for this research was provided by the UMBC Mechanical Engineering Department and College of Engineering and Information Technology

INVESTIGATING THE FUNCTIONALITY OF PAPER-BASED BIOSENSORS USING 3D-PRINTED WAX

Haddijatou Niasse¹, David C Garcia¹

¹Department of Chemical, Biochemical, Environmental Engineering, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Many common diagnostic tools including pregnancy tests, COVID-19 tests, and blood glucose monitors fall under the umbrella of portable biosensors that can be used anywhere with a specific reagent. Unfortunately, there are many areas that do not have access to basic health needs and if they do, they must transfer the samples to multiple testing facilities before obtaining results. This project involves the production of a portable paper-based cell free extract biosensor using Whatman paper that can be used to test various types of samples by adding its specific reagent without outside sources of technology. Currently, there is a lack of existing tools to create these biosensors but recently, using wax filament to 3D print stencils onto Whatman paper has been efficient. Preliminary testing has been done by observing the migration of 3uL of a 1:9 diluted purple food coloring solution onto 3D printed Whatman paper followed by being melted at 130-140℃ for 0, 8, 16, 24, and 32 minutes. Following the fusion process, 6 randomly selected diameters of dye were measured from each ticket. Through testing, it is now known that the ticket melted for 32 minutes, created the most effective hydrophobic border and prevented the spread of dye beyond the printed boundaries on the Whatman paper. This approach shows that 3uL or less of liquid can be contained within printed boundaries. All in all, it supports the future development of paper-based cell-free biosensors for diagnostic applications in spaces with or without advanced technology .

Contribution for this project was given by the Department of Defense (DoD).

OPTIMIZING TRANSFECTION METHODS TO IDENTIFY THE INTERCELLULAR LOCATION OF THE BRADYKININ B1 RECEPTOR.

Nia Nkansah¹, Shannon Chastain¹, Laurie Sutton^1

1Department of Biological Sciences, University of Maryland, Baltimore County, 1000
Hilltop Circle, Baltimore, MD, 21250

The Bradykinin B1 receptor (B1R) is a unique G protein coupled receptor (GPCR) that is activated during inflammatory responses and is associated with chronic pain. Generally, GPCRs are located at the plasma membrane until activated by a ligand and endocytosed. Unlike other GPCRs, B1R does not directly traffic to the plasma membrane, but we hypothesize that under certain inflammatory conditions it will. B1R’s exact intracellular location and trafficking pathways are poorly understood, which leaves a gap in our understanding of how location could ultimately influence the receptor’s signaling activity and function. This project focuses on developing an effective transfection protocol to locate B1R and map out its trafficking pathway.

For this project, HeLa and COS-1 cells were transfected with a plasmid encoding an N-terminal mCherry-tagged B1R to determine its location under naive conditions. We optimized our transfection protocol by testing different transfection reagents and amounts of plasmid. We compared Lipofectamine 2000 to Lipofectamine LTX and the ratios of Lipofectamine to DNA to maximize protein expression. Following our receptor optimization, B1R was co-transfected with specific Venus-tagged lipid markers at various amounts to assess possible colocalization through confocal microscopy. Additional markers will be used to identify other potential locations within the cell. Future studies will test imaging and trafficking under inflammatory conditions.

This student was financially supported by the Meyerhoff Scholars Program.

VALIDATION OF MUSCLE-SPECIFIC GAL4 DRIVER LINES FOR AGING RESEARCH IN DROSOPHILA MELANOGASTER

Sariah Rimmer^1,3, Elise Delaporte^1,3, Rose Mupende^1,3, Kallaina Basnet¹, Apolline Nurit², Dr. Jeff Leips¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

²Miami University, 501 High St, Oxford, OH 45056

³Meyerhoff Scholars Program, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

FORMALDEHYDE STIMULATES CYTOSOLIC CA²⁺RESPONSE IN OLFACTORY NEURONS INDEPENDENT OF CANONICAL ODOR DETECTION PATHWAY

Favour Nwogu¹, Iskandar Nassar¹, Farhan Augustine¹, Tatsuya Ogura¹, and Weihong Lin¹

¹Department of Biological Sciences, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

This project is supported by the ATIP pilot Grant and START award to WL.

O

ELECTROSPINNING OF GELATIN BASED FIBROUS SCAFFOLDS FOR EXTRACELLULAR MATRIX MIMICRY AND CELL CULTURE

Oluwagbotemi Ogungbile, Yamelak Andargie, Luis Pinzon-Herrera, Ph.D. Jorge Almodovar, Ph.D,

Department of Chemical, Biochemical, and Environmental Engineering, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

The extracellular matrix (ECM) is the natural network of proteins and fibers that holds tissue together and guides its repair, which the body builds effortlessly. Replicating it in the lab is not. Electrospinning offers a promising path: a technique versatile enough to spin fibrous structures that closely mimic the ECM’s architecture, with real implications for regenerative medicine. Where cell attachment and tissue formation can mean the difference between failure and repair. To explore this, we fabricated fibrous scaffolds from Type B gelatin, crosslinking with glutaraldehyde, and seeded with Schwann cells to evaluate biocompatibility. Initial results have shown promise for cell survival, and our focus now is optimizing that process to produce scaffolds that consistently support cell growth. The result is a clearer picture of electrospinning’s potential as a platform technology. Building on this foundation, future work will introduce the metabolite itaconate (ITA) to the fibers to examine cellular response, and will incorporate collagen into the fibers, as we take steps toward a validated electrospun platform for tissue regeneration.

OPTIMIZING T7 RNA TRANSCRIPTIONAL HOMOGENGENITY FOR NUCLEAR MAGNETIC RESONANCE (NMR) APPLICATIONS

Geoffrey Okonkwo¹, Eunice Ewusie¹, Sridhar Alagar¹, Brandon Fonseca¹, Michael Summers, Ph.D¹.

¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County 1000 Hilltop Circle, Baltimore MD 21250

The Human Immunodeficiency Virus (HIV) is a retrovirus that weakens the body’s immune system by primarily targeting CD4+ T cells. As it stands, the most popular method of treatment for HIV are anti-retroviral therapies(ARTs). ARTs help treat the virus by inhibiting key viral proteins that are vital for HIV’s life cycle. Unfortunately, there are many drawbacks that are associated with ARTs that reduce the effectiveness of these medications. Therefore, a deeper understanding of HIV is required, specifically of the interactions supporting the viral life-cycle. We are focused on a region of the genome that is highly conserved that is known to play an important role in supporting the fate of the virus’ RNA, the 5’Leader. One major obstacle in the efforts to study the 5’L is that the RNA polymerase used, “T7,” is prone to producing heterogeneous samples. This heterogeneity is caused by T7 inaccurately creating rna products with too few or too many nucleotides than the desired product. The resulting mixture of RNA species complicates downstream analyses, particularly Nuclear Magnetic Resonance (NMR) spectroscopy, where the additional RNA species generate unwanted signals that interfere with structural characterization. To address this challenge, we evaluated a series of DNA template modifications alongside mutant T7 RNA polymerases. Through this approach, a potent system is established for reducing heterogeneity of RNA during synthesis.

Funding for this research is supported by the Howard Hughes Medical Institute and the National Institute of Allergy and Infectious Diseases (NIAID, 5R01AI150498).

COMPARATIVE STUDY OF GENOME PACKAGING MECHANISMS IN HIV-1 AND MoMuLV: DIMERIZATION, CAP SEQUESTRATION, AND RNA FATE

Aaliyah Owens^1,3; Martha Tumwikirize^2,4; Brea A. Manuel²; Michael F. Summers^1,2.
¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, 1000 Hilltop Cir, Baltimore, MD 21250
²Howard Hughes Medical Institute, 4000 Jones Bridge Rd, Chevy Chase, MD 20815
³Department of Biology, Howard University, 2400 6th St. NW, Washington, DC 20059
⁴Department of Biology, Virginia Military Institute, 401 Letcher Ave, Lexington, VA, 24450

Human Immunodeficiency Virus Type 1 (HIV-1) is a retrovirus that affects nearly 41 million people globally. HIV-1 contains a heterogeneous transcription start site (TSS) forming two distinct pools of RNA that differ in sequence by the presence of either one or three guanosines after the 5’ cap (Cap1G and Cap3G respectively). The Cap3G exists as a monomer, where the 5’cap is exposed, and thus functions as messenger RNA (mRNA). In contrast, the Cap1G exists as a dimer, where the 5’cap is sequestered, and thus functions as genomic RNA (gRNA). Moloney Murine Leukemia Virus (MoMuLV) is a distantly related retrovirus that causes cancers and neurological diseases in rodents and shares similarities in genomic RNA packaging and replication with HIV-1. Unlike HIV-1, MoMuLV contains a unique start site, producing one pool of RNA that serves both mRNA and gRNA function. The goal of this project is to explore the role of cap-sequestration on retroviral genome fate across divergent retroviruses. We hypothesize that MoMuLV’s genome fate is based on dimerization-dependent cap sequestration, and that cap sequestration is essential for gRNA packaging, like HIV-1. Thus far, eIF4E-based electrophoretic mobility shift assays (EMSAs) have shown that the MoMuLV monomer readily binds eIF4E, indicating that the 5’ cap is exposed in the monomeric state. Under dimer-promoting conditions, reduced eIF4E binding was observed, suggesting that dimerization may sequester the cap and limit its accessibility. Additional optimization of RNA capping efficiency, along with viral packaging studies are currently underway to confirm dimerization-dependent cap sequestration and determine its role in directing genome fate. Future work includes nuclear magnetic resonance (NMR) experiments to determine the structural means of cap-sequestration. The results of this work will help us understand whether dimerization-dependent cap sequestration is a conserved mechanism regulating genome fate among retroviruses.

This student was supported by CRNA. Support for this research was provided by the Howard Hughes Medical Institute (HHMI) and NIH/NIAID grant (#5R01AI150498).

P

DEVELOPING A BEHAVIOR TRACKING GUI TO EXAMINE THE MODULATORY ROLE OF JUVENILE HORMONE ON SOCIAL CUES IN D. MELANOGASTER

Daniel Palus, Antonio Marini-Davis, Fernando Vonhoff, PhD

Department of Biological Sciences, University of Maryland, Baltimore County, 1000

Hilltop Circle, Baltimore, MD, 21250

In humans, thyroid hormone (TH) is known to play a vital regulatory role in the development of social behaviors. The role of TH in humans can be modeled in Drosophila melanogaster by studying juvenile hormone (JH), the Drosophila analog to TH. Recent work in our lab observed that JH modulates olfactory reward behaviors to low-concentration ethanol, while other literature has also evidenced JH as a crucial hormone in social development. This led us to hypothesize that ethanol is a social cue rather than an appetitive stimulus. By observing social behaviors under different levels of JH, we seek to provide novel insights into the effect of JH on the development of social behavior. Here, we present a behavior-tracking GUI, building upon well-established software to explore real-time changes in social dynamics, such as proximity, interaction, and exploration, under overcrowding-induced chronic stress (OICS). Because OICS is well studied in Drosophila, we employed it here to assess the system’s reliability and precision. This GUI allows us to measure changes in latent fly sociability resulting from changes in JH in the brain, which will provide valuable insight into TH-driven neuroendocrine mechanisms governing sociability.

IN VITRO SYNTHESIS AND PURIFICATION OF POLIOVIRUS (-) STRAND CLOVERLEAF RNA FOR ITS STRUCTURAL CHARACTERIZATION AND INTERACTION STUDIES

Ryan Paradela¹, Bethel G. Beyene¹, and Deepak Koirala¹

¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, Baltimore, MD 21250

Poliovirus Type 1 (PV1) is a positive-sense (+) single-stranded RNA virus in the Enterovirus genus of the Picornaviridae family. Poliovirus is the etiological agent of severe meningitis or paralysis. Although poliovirus vaccines are available, the possible resurgence of new variants and limited access to vaccines in some parts of the world mean this virus still poses a threat. Because no antiviral therapeutics currently exist, understanding the enteroviral genome structure could reveal potential targets for antiviral therapies.

The PV1 genome consists of a single open reading frame flanked by 5′ and 3′ UTRs. Within the 5′ UTR, domain I, called the 5′ cloverleaf (5′ CL), promotes replication. Our lab recently determined the crystal structures of 5′ CLs, revealing a highly conserved H-type four-way junction that mediates interactions with replication-related proteins. The (-) strand, synthesized from the (+) strand template during replication, serves as the template for (+) strand synthesis and is proposed to form a similar cloverleaf at the (-) 3′ end (3′ CL). This project aims to investigate the structural features of the PV1 3′ CL using X-ray crystallography and to elucidate its interactions with replication-related proteins.

The method requires highly pure and concentrated RNA samples, which are prepared by in vitro transcription from DNA templates that are in turn prepared through PCR amplification. After designing and purchasing DNA templates and primers, we optimized several PCR conditions. Using agarose gel electrophoresis to visualize PCR products under various conditions, we found that adjusting the polymerase extension time improved DNA amplification, yielding high-quality templates for subsequent transcription. While the synthesis and purification of RNA samples using these DNA templates are ongoing, future studies aim to grow the RNA crystals and determine their crystal structures. Once the structure is solved, we will investigate structure-guided interactions of the PV1 3′ CL with replication-linked proteins.

This work was supported in part by NSF CAREER award 2236996 to D.K. and by NIH T32 grant GM158458 to B.G.B.

MODELING NUCLEUS ACCUMBENS MEDIUM SPINY NEURON DYNAMICS USING A BIOPHYSICAL CONDUCTANCE-BASED FRAMEWORK

Nicholas Peercy¹, Kathleen Hoffman², Tara LeGates³ , Branwen She³, Matyas Marek³, Ashley Copenhaver³

¹Department of Mathematics, University of Maryland, College Park, MD 20742

²Department of Mathematics and Statistics, University of Maryland, Baltimore County, 1000

Hilltop Circle, Baltimore, MD, 21250

³Department of Biological Sciences, University of Maryland, Baltimore County, 1000

Hilltop Circle, Baltimore, MD, 21250

Medium spiny neurons (MSNs) are the primary output neurons of the nucleus accumbens (NAc), a brain region involved in reward processing, motivation, and stress-related behavioral states. Dysfunction in the NAc has been connected to many neuropsychiatric diseases, such as anhedonia and depression. Our previous work used a modified Izhikevich model to fit current-clamp data, providing a computational framework for estimating parameters across sex and experimental conditions. However, the Izhikevich model uses abstract parameters that do not readily correspond to biophysically meaningful quantities. In this project, we developed a reduced biophysical conductance-based model of MSN excitability. This classical Hodgkin-Huxley-type model includes fast sodium, delayed-rectifier potassium, inward-rectifier potassium, persistent sodium, fast and slow A-type potassium, M-type potassium, L-type calcium, intracellular calcium, and calcium-activated potassium currents. This framework is designed to reproduce key MSN electrophysiological features, including a hyperpolarized resting potential, delayed first-spike firing, repetitive spiking, and spike-frequency adaptation. MATLAB simulations show that the model can generate delayed first-spike behavior under a current-step protocol that matches the experimental recordings. Model parameters were fit to a representative current-clamp recording using a weighted cost function and multistart Nelder–Mead optimization. The fitted model reproduced the experimental spike count, delayed first-spike timing, mean interspike interval, and termination of firing after stimulus offset. The principal remaining mismatch was an overly hyperpolarized post-spike trough. These results provide proof of concept that a reduced conductance-based framework can reproduce key firing features of a representative MSN trace while retaining mechanistic links between fitted parameters and ionic currents.

AUTOMATING TRACTION FORCE MICROSCOPY: A MACHINE LEARNING PIPELINE FOR HIGH THROUGHPUT PLATELET MECHANOBIOLOGY

Donovan Peyton¹, Trevor Whitfield¹, Molly Y. Mollica, Ph.D.¹

¹University of Maryland, Baltimore County (UMBC) Department of Mechanical Engineering

Platelet contractile forces are clinically relevant biomarkers of bleeding, yet high throughput quantification remains limited by the computational and personnel demands of existing analysis techniques. The Mollica Mechanobiology, Engineering and Cell mechanics in Health (MECH) Lab’s “black dots” traction force microscopy (TFM) technique offers a reference-free approach to measuring single-platelet forces on flexible PDMS substrates, but the original manual MATLAB pipeline required approximately 500 minutes to analyze 500 platelets.

This summer project aimed to fully automate the black dots analysis pipeline to reduce personnel and computer time by 80% while maintaining 95% agreement with manual quantifications. Toward this goal, cell boundary detection, work flow, and dot selection were automated. Further, cells that do not meet noise and error thresholds were automatically removed and ambiguous cases will be flagged for user review.

Optimization efforts have reduced CPU run time by 64.5% and human operator time by 84.8% per dataset while maintaining agreement with manual analyses. For a sample dataset of 10 platelets, the automated code resulted in a mean force of 15.063 nN, only a 2.03% difference from the manual average of 15.376 nN. When comparing these analyses with a paired student t-test, the results were not statistically significantly different.

The completed pipeline will support modeling of traction force as a function of substrate coating, cytoskeletal morphology, spread area, and circularity, with the goal of maintaining 95% accuracy relative to manual analysis. Outcomes will directly expand the lab’s capacity to investigate platelet mechanobiology at a larger population scale, with implications for hemostatic dysfunction in bleeding and thrombotic disease.

This research was funded by the UMBC COEIT Summer Award.

DEVELOPMENT OF A LOW-COST ATMOSPHERIC AMMONIA (NH₃) MEASUREMENT SYSTEM

Audrey Pollack, Christopher Hennigan, Ph.D.; Anvita Pentakota; Dami Aimola

Department of Chemical, Biochemical, and Environmental Engineering, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

Atmospheric ammonia is an important air pollutant that contributes to fine particulate matter formation, ecosystem degradation, and adverse human health effects. Continuous monitoring of atmospheric ammonia remains challenging because existing instrumentation is often expensive and difficult to deploy at scale. To address this limitation, our team is developing a low-cost automated spectrophotometric sensor capable of continuously measuring gaseous NH₃and particulate NH₄⁺. Measuring both phases separately further provides insight into ammonia partitioning which is critical for characterizing aerosol thermodynamics. The system integrates automated sampling, microfluidics, Berthelot colorimetric chemistry, and UV-Vis spectroscopy to quantify atmospheric NH₃ and NH₄⁺concentrations. Our research focuses on optimizing reaction conditions, microfluidic flow systems, and analytical instrumentation while generating calibration curves and evaluating instrument performance using OceanView and MATLAB. This work aims to improve instrument sensitivity, accuracy, and detection limits while advancing accessible, continuous atmospheric chemistry measurements.

MACHINE LEARNING APPROACHES FOR NO₂ FORECASTING: DIRECT MODEL PREDICTIONS VS. FOUNDATION MODEL WITH RESIDUAL CORRECTION

Michael Togbe¹, Aryan Poshtiwala², Ray Chen³, Oscar Tuten⁴, Adeline Barden⁵,

Pavan Raj Ravi⁶, Ruoyan Gong⁶, Kyo Hugo Lee⁷, Nicholas LaHaye⁷, Xiaohua Pan⁷,

Hazem Mahmoud⁷,and Jianwu Wang⁶

¹Department of Computer Science and Electrical Engineering, University of Maryland, Baltimore County

²Clarksburg High School, Montgomery County, Maryland

³Department of Computer Science, University of Maryland, College Park

⁴Department of Computer and Data Science, Goucher College

⁵Department of Education, Troy University

⁶Department of Information Systems, University of Maryland, Baltimore County

⁷P.C. Rossin College of Engineering and Applied Science, Lehigh University

⁸ NASA Jet Propulsion Laboratory

⁹ Spatial Informatics Group, LLC

¹⁰ NASA Goddard Space Flight Center

¹¹ NASA Langley Research Center

Accurate nitrogen dioxide (NO₂) forecasting is critical for air quality management, yet training models from scratch is computationally expensive. Pretrained geo foundation models such as Microsoft’s Aurora can be applied to forecasting tasks without retraining. However, it remains unclear whether using a pretrained foundation model improves NO₂ forecasting compared with direct machine learning approaches. This work compares two approaches for forecasting total column NO₂ (tcNO₂) over California. The first post-processes Aurora forecasts through residual correction using Copernicus Atmosphere Monitoring Service (CAMS) atmospheric as input features and NASA TEMPO satellite observations as ground truth. The second trains machine learning models to predict TEMPO tcNO₂ directly from CAMS atmospheric inputs. Several machine learning models are trained and evaluated under both approaches to determine whether incorporating a pretrained foundation model produces forecasts that more closely match TEMPO observations than direct prediction from CAMS inputs.

This work is supported by the grant “REU Site: Online Interdisciplinary Big Data Analytics in Science and Engineering” from the National Science Foundation (grant no. OAC-2348755). We acknowledge the computational resources in the UMBC High Performance Computing Facility (hpcf.umbc.edu) and the financial contributions from NIH, NSF, CIRC, and UMBC for this work.

R

VALIDATION OF MUSCLE-SPECIFIC GAL4 DRIVER LINES FOR AGING RESEARCH IN DROSOPHILA MELANOGASTER

Sariah Rimmer^1,3, Elise Delaporte^1,3, Rose Mupende^1,3, Kallaina Basnet¹, Apolline Nurit², Dr. Jeff Leips¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

²Miami University, 501 High St, Oxford, OH 45056

³Meyerhoff Scholars Program, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

EXPLORING THE STRUCTURE AND MECHANISMS OF HIV-1 VIRION ASSEMBLY THROUGH GAG/RNA INTERACTIONS

Ariana Rodriguez¹, Alexandra Foster¹, Eric Shreiter-Zavala¹, Samuel Li^1,2,3, Aryan
Srivastava^1,2, Pengfei Ding², Michael Summers^1,2

¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, Baltimore, MD 21250

²Howard Hughes Medical Institute, University of Maryland, Baltimore County, Baltimore, MD 21250

³Department of Biology, College of William and Mary, Williamsburg, VA 32185

Human Immunodeficiency Virus-1 (HIV-1) affects over 40 million people worldwide. While antiretroviral therapy (ART) exists, it is not curative, and HIV’s high mutation rate makes long term treatment difficult, necessitating further study of the HIV-1 replication cycle. The Gag polyprotein is crucial for virion assembly, forming a shell encapsulating dimerized viral genomic RNA (gRNA). Gag notably contains a capsid (CA) for lattice assembly and nucleocapsid (NC) for RNA binding, among other domains. Recognition between Gag and the packaging signal (Psi) on the 5’-leader of gRNA is needed for virion assembly initiation and selective gRNA packaging. Preliminary results from cryogenic electron microscopy (cryo-EM) study reveal a model of the Gag/Psi nucleation complex that suggests the dimerization initiation site (DIS) of the RNA bridges two Gag hexamers together. The low resolution of the cryo-EM map and the lack of RNA regions other than DIS in the model, however, renders accurate interpretation difficult. To determine the atomic structure of this Gag/Psi nucleation complex, we focus on improving particle quality for cryo-EM analyses. The binding and assembly behavior of many Gag variants onto the packaging signal under different conditions are evaluated through electrophoretic mobility shift assays (EMSAs) and negative stain electron microscopy (EM). A Gag construct with a bulky tag added to the N-terminus of CA led to homogeneous particles, which is suitable for cryo-EM single-particle analysis (SPA). We have collected high-quality cryo-EM data using this sample and are currently working on the structure refinement. The three-dimensional structure will provide the detailed Gag/Psi recognition mechanism underlying Gag assembly initiation and selective gRNA packaging. Gaining a better understanding of these processes will facilitate future development of therapeutics targeting Gag-gRNA interactions during genome packaging and virion assembly.

Support for this research was provided by NIH/NIAID #U54-AI1-700660, NIH/NIAID #5R01AI150498, the Howard Hughes Medical Institute (HHMI), and the UMBC Meyerhoff Scholars Program.

INVESTINGATING THE INFLUENCE OF POLYMER COMPOSITION ON THE PROPERTIES ON GEL ELECTROLYTES

Owen Rubin¹, Adim Amanyeiwe¹, Rahuldeb Roy¹, Deepa Madan²

¹Department of Mechanical Engineering, University of Maryland Baltimore County, 1000 Hilltop Circle, Catonsville, MD, 21250

Lithium-Ion batteries are extensively used throughout the world, as they have some of the highest energy capacities and lifespans among other batteries. However, they hold numerous problems, such as overheating that leads to fires, and capacity loss that leads to poor performance. Batteries made with a combination of zinc (Zn) and manganese oxide (MnO₂) have the potential to be an efficient, safer and cheaper alternative. In this study, we investigated the influence of polymer composition on the electrochemical and mechanical properties of gel polymer electrolytes by varying the ratio of poly(vinyl alcohol) (PVA) and polyacrylamide (PAM) in both acidic and alkaline media. We hypothesized that the ideal ratio of PVA: PAM would have an optimal balance between ionic conductivity and mechanical properties. We prepared electrolytes with PVA: PAM ratios of 1:0, 7:3, 1:1, and 3:7, and evaluated the electrochemical properties via electrochemical impedance spectroscopy (EIS), linear scan/sweep voltammetry (LSV), Tafel polarization, chronoamperometry, and overpotential measurements to assess ionic conductivity, electrochemical stability, corrosion behavior, current retention, and polarization behaviors. We found that there is a trade-off between mechanical strength and ionic conductivity with varying polymer composition. Increasing the PAM content enhances the ionic conductivity and reduces the mechanical robustness, whereas increasing the PVA has the opposite effect. In acidic environments, we were able to determine the optimal balance for Zn-MnO₂ battery performance falls between 1:1 and 3:7 ratios. In contrast, the alkaline gel electrolytes show poor electrochemical stability due to the formation of zinc oxide (ZnO) and precipitation of zincate species, promoting electrode degradation and inconsistent data. Further studies focus on optimizing longevity of the gel electrolytes for acidic media. In alkaline medium, more research is required to develop strategies to suppress the parasitic reactions while preserving the role of hydroxide ions.

S

INVESTIGATING THE ROLE OF STING IN PROSTATE INFLAMMATION AND CANCER USING MOUSE MODELS

Deena Saberi¹, Alexandria Howe¹, Charles Bieberich¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

Prostate cancer is the second leading cause of cancer-related death among men in the United States¹, prioritizing the need to better understand the factors that can contribute to its development. One possible factor is thought to be chronic inflammation, although there is still limited evidence to support the hypothesis that it can act as a precursor to prostate cancer over time. Chronic inflammation, such as that observed in chronic prostatitis, has been associated with the development of proliferative inflammatory atrophy (PIA), an early inflammatory lesion, and prostatic intraepithelial neoplasia (PIN), a precursor to prostate cancer. Since the STING-mediated innate immune response plays a key role in inflammation, this study focuses on the cyclic GMP-AMP synthase-stimulator of interferon genes (cGAS-STING) pathway. In this pathway, STING acts as an essential upstream activator of the innate immune response, making it a potential regulator of chronic inflammation that may contribute to the development of precancerous prostate lesions and, ultimately, prostate cancer. To understand the function of STING in chronic inflammation, a STING knockout (KO) mouse model was generated using CRISPR/Cas9 gene editing and crossed with our RIG mouse model. The RIG model is a doxycycline-inducible system that induces prostate-specific inflammation by activating reverse tetracycline transactivator (rtTA) and driving the expression of the pro-inflammatory cytokine interleukin-1 beta (IL-1β) and green fluorescent protein (GFP), enabling inflammation to be induced and monitored. STING KO mice were crossed with RIG mice to determine how the loss of STING alters prostate inflammation levels and its associated pathological changes. After induction via doxycycline, prostate tissues are collected for histological and immunohistochemical analysis to compare inflammation levels and precancerous lesions, between STING KO and control mice. This study will allow a better understanding of how STING signaling can influence innate immune responses and contribute to chronic inflammation. As this relationship is better understood, STING could be targeted using future drug therapies to decrease inflammation in patients with greater cancer risk due to chronic inflammation.

This work was supported by NIH grants 2R01CA200900-06A1 and 1U54CA274370-01.

STRUCTURAL STUDIES OF GAG-RNA INTERACTIONS THAT NUCLEATE HIV-1 PARTICLE ASSEMBLY

Eric Schreiter-Zavala^1,2, Alexandra Foster^1,2, Samuel Li^2,3, Ariana Rodriguez^1,2, Pengfei Ding^1,2, Michael Summers^1,2

¹Department of Chemistry & Biochemistry, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

²Howard Hughes Medical Institute, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

³Department of Biology, William & Mary, 116 Jamestown Road, Williamsburg, VA 23185

HIV-1 is an immune cell-targeting virus that affects millions of people worldwide and can lead to life-threatening Acquired Immunodeficiency Syndrome (AIDS) if not treated properly. There is currently no cure, and its high mutation rate often leads to drug resistance, making the need for novel therapies ever present. HIV-1 is a retrovirus, meaning it carries its genetic information as genomic RNA (gRNA). HIV-1’s multi-domain structural protein Gag is responsible for the selective packaging of the RNA genome from a cell milieu containing a large excess of non-viral RNAs. Two parts of Gag are especially important for this packaging–capsid (CA), which forms hexamers that make a shell to enclose gRNA, and nucleocapsid (NC), which binds to the 5’ untranslated region of gRNA, known as the packaging signal (Ψ). The exact mechanisms behind this selective packaging are still unknown, but it represents a promising target for future antiretroviral therapies. We hypothesize that the unique RNA structure of Ψ and the locations of its ~20 Gag binding sites catalyze initial CA interactions. This Gag/Ψ complex then nucleates robust formation of the Gag hexagonal lattice. We are currently working to solve the 3D structure of this Gag/Ψ nucleation complex via Cryogenic Electron Microscopy (Cryo-EM). Through protein and RNA engineering, we have significantly improved our particle quality, making it more feasible for Cryo-EM single particle analysis. By determining this 3D structure, we can further understand how HIV-1’s RNA genome is selectively packaged, which can aid in the development of RNA-targeted antiretroviral therapies.

Support for this research was provided by NIH/NIAID #U54-AI1-700660, NIH/NIAID #5R01AI150498, and the Howard Hughes Medical Institute (HHMI).

INVESTIGATING THE ROLE OF THE PHD DOMAIN IN THE SET4 PROTEIN AND ITS FUNCTION IN HYPOXIC GENE REGULATION

Nathaniel Shelton^1,2, Phoenix Bryant¹, Winny Sun, Ph.D.¹, Maraki Negesse, Ph.D.¹, Erin Green, Ph.D.^1,3

¹Department of Biological Sciences, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

²Howard Hughes Medical Institute, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

³University of Maryland Greenebaum Comprehensive Cancer Center, University of Maryland, Baltimore, 22 S. Greene Street, Baltimore, MD, 21201

Although protein domains have a prolifera of functions, their complex interaction with gene regulation proves to be insightful in understanding genomic diseases, gene expression, and other biochemical functions. The specific cell condition, hypoxia, places stress upon cells through blocking off oxygen. A main characteristic of cells under hypoxic conditions is a pivot in gene expression often activating or repressing genes. The Set4 protein is a mediator of this response acting as a hypoxic stress responder and epigenetic regulator. The Set4 protein’s role in stress response is to bind to chromatin and regulate genes for survival. The PHD domain found in the Set4 protein has a key function related to chromatin binding, but its impact is not fully understood. Previous experiments demonstrated the SET4-PHDΔ genotype behaved similarly to the set4Δ genotype with both displaying an upregulation of genes involved in cell wall structure. To measure the contribution of the PHD domain on expressed genes, four variations of yeast cells were developed, wild-type, set4Δ, GFP-SET4, and GFP-SET4-PHDΔ. These yeast cell strains are grown on plates before being inoculated into liquid media and placed into hypoxic conditions. RNA extraction was then performed, followed by a RT-qPCR in order to measure relative gene expression. We evaluated previously explored genes in addition to other genes related to the cell wall under hypoxia and uncovered similar results along with new discoveries of certain genes appearing more upregulated in the SET4-PHDΔ genotype rather than the set4Δ genotype. These results illustrate that the PHD domain may have an important function in limiting the hyper-expression of cell wall related genes providing a powerful insight into the purpose of the PHD domain and its role in gene expression.

Support for this research was provided by the grant to UMBC from the Howard Hughes Medical Institute through the Precollege and Undergraduate Science Education Program and funding from the National Institute of Health (R01GM148698).

INVESTIGATING CONSERVATION OF THE HIV-1 CORE ENCAPSIDATION SIGNAL

Aaron Johnson^1,2; Sofia Shevchenko^1,2; Vianney Tanifor^1,2; Zeynep Akan^1,2;Horasa Ji²; Michael Summers^1,2

¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

²Howard Hughes Medical Institute, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

Human immunodeficiency virus type 1 (HIV-1) selectively packages its genomic RNA through a highly structured region known as the Core Encapsidation Signal (CES), the minimal RNA element required for genome recognition and packaging into new viral particles. Unlike the full HIV-1 leader RNA, the CES lacks the TAR, poly(A), and primer binding site (PBS) regions while retaining key structural elements, including the dimer initiation signal (DIS), splice donor (SD)-associated region, Ψ packaging signal, AUG region, and the U5:AUG interaction. Previous studies of the HIV-1 NL4-3 laboratory strain demonstrated that the SD region adopts a continuous structure rather than folding back on itself in the 1G conformation. This project investigates whether the same structural arrangement is conserved in the HIV-1 MAL strain. Using nuclear magnetic resonance (NMR) spectroscopy, we aim to characterize the three-dimensional structure of the MAL CES RNA and compare its folding with that of NL4-3. Determining whether this conserved architecture exists across HIV-1 strains will improve our understanding of the structural mechanisms underlying viral genome packaging and may provide insights into conserved RNA features that could serve as future antiviral targets.

Support for this research was provided by the Howard Hughes Medical Institute and NIAID grant #5R01AI150498.

VISUALIZING AND ANALYZING SPATIOTEMPORAL AEROSOL VARIATIONS DURING THE 2023 CANADIAN WILDFIRES

Sophia Summers¹; Jasper Lewis, Ph.D.²

¹Department of Physics, Colby College, 4000 Mayflower Hill Drive, Waterville, ME 04901

²Department of Physics, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

Wildfires are a major source of atmospheric aerosols, which can have many impacts on air quality, climate, and human health. During the summer of 2023, Canada saw an increase in wildfire events which led to the transport of smoke across much of Eastern North America. Using observations from the Micro Pulse Lidar Network (MPLNET) and the co-located Aerosol Robotic Network (AERONET), this study investigates the spatiotemporal variability of aerosol optical properties at multiple sites affected by the smoke event. To improve accessibility and analysis of MPLNET data, Python programs were developed to retrieve, quality screen, and visualize lidar observations. These tools allow for efficient comparison of aerosol properties across multiple locations and time periods through the generation of maps, curtain plots, time series, average profiles, and correlation analyses. These observations were used to investigate changes in aerosol properties, like aerosol optical depth (AOD) and extinction, to illustrate the difference between aerosol loading and vertical structure across the sites. These results provide insight into the transport and variability of aerosols resulting from regional events such as wildfires and contribute to improved understanding of aerosol distributions relevant to atmospheric modeling and remote sensing applications.

T

INVESTIGATING THE STRUCTURE OF THE HIV-1 CORE ENCAPSIDATION SIGNAL IN NL4-3 AND MAL

Vianney Tanifor^1,2,3; Sofia Shevchenko^1,2; Zeynep Akan^1,2; Aaron Johnson^1,2; Horasa Ji²; Michael Summers^1,2,3

1Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250
²Howard Hughes Medical Institute, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250
³Center for Structural Biology of HIV RNA, Department of Microbiology and Immunology, University of Michigan Medical School, Ann Arbor, MI 48109

Human immunodeficiency virus type 1 (HIV-1) must selectively package two copies of its full-length genomic RNA into assembling viral particles. This process is directed by structures within the 5′ leader, including a minimal packaging-competent region termed the core encapsidation signal (CES). HIV-1 transcripts that begin with a single guanosine (1G) preferentially adopt a packaging-competent state. Structural studies of the NL4-3 laboratory strain indicate that, within this state, the major splice donor region forms an extended, continuous helical arrangement rather than the previously proposed local hairpin. Whether this architecture is conserved in genetically divergent HIV-1 strains remains unknown.

We are using solution nuclear magnetic resonance (NMR) spectroscopy to characterize the secondary and three-dimensional structures of the 1G CES from the HIV-1 MAL strain and compare its folding with NL4-3. This comparison will identify conserved and strain-specific base-pairing interactions and test whether the NL4-3 structure represents a shared packaging architecture. Establishing the degree of CES structural conservation will clarify how HIV-1 recognizes and packages its genomic RNA despite sequence variation and may reveal conserved RNA features that could be targeted to disrupt viral assembly.

Support for this research was provided by the Howard Hughes Medical Institute and the National Institute of Allergy and Infectious Diseases grant 5R01AI150498.

CHARACTERIZING ADIPOSE TISSUE EXTRACELLULAR MATRIX STRUCTURAL CHANGES IN A DIET-INDUCED OBESE MOUSE MODEL

Irene Teye¹, Marveline Akinola, BS.¹, Benjamin T. Cole, BS.¹, Diana M. Elizondo, PhD.¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Obesity is a growing public health challenge characterized by chronic low-grade inflammation and extensive remodeling of adipose tissue. As adipose depots expand, the extracellular matrix (ECM) undergoes profound structural changes that promote fibrosis, restrict adipocyte plasticity, and contribute to metabolic dysfunction. Excessive ECM deposition and altered matrix architecture are thought to impair tissue flexibility, adipose expandability, and nutrient handling, yet the structural changes accompanying diet-induced obesity remain incompletely understood.

The objective of this study was to characterize obesity-induced alterations in adipose tissue ECM architecture using a diet-induced obesity mouse model. Male and female C57BL/6 mice were fed either a control chow diet or a high-fat diet to generate lean and obese cohorts. Epididymal white adipose tissue (eWAT) was collected and processed for scanning electron microscopy (SEM) to visualize the three-dimensional organization of the ECM at high resolution. Direct comparison of lean and obese eWAT enabled evaluation of obesity-associated changes in matrix organization and tissue architecture.

SEM images were quantitatively analyzed to assess collagen fiber density, network organization, pore size, tissue porosity, collagen bundle thickness and alignment, and the spatial relationship between ECM deposition and adipocyte organization. Obese mice exhibited increased ECM accumulation, denser and more disorganized collagen networks, and reduced pore size compared with chow-fed controls, consistent with fibrotic remodeling. These structural changes were accompanied by decreased tissue porosity, suggesting the development of a stiffer ECM that may limit adipose tissue expansion.

Together, these findings demonstrate that diet-induced obesity profoundly alters the three-dimensional architecture of the adipose ECM. Characterizing these obesity-associated structural changes provides complementary insight into the mechanisms of adipose tissue fibrosis and may identify ECM-directed therapeutic strategies to preserve adipose tissue function, improve insulin sensitivity, and reduce obesity-associated metabolic disease.

This work was supported by the NIDDK grant 4R00DK136921-02 and UM FIRST awards to Dr. Elizondo.

BACTERIA TESTING AT YOUR FINGERTIPS: MODULAR SILVER CHLORIDE ELECTRODE

Umar Thangalvadi, Grayson Ruffner, Chengpeng Chen, Ph.D.

Department of Biological Sciences, University of Maryland, Baltimore County, UMBC, 1000

Hilltop Circle, Baltimore, MD 21250

Bacterial infections are crucial health problems around the globe. This study aimed to design a modular silver chloride electrode to detect bacteria with sensitivity, accuracy, and speed. The modular electrode design was achieved using 3D printing software and computer-aided design (CAD). Each electrode gave specific, selective, and linear responses to silver. First, the 3D-printed electrodes were fabricated as detectors. These detectors required a potentiometer, for which an Arduino was developed to couple the electrode for data transduction. Using this sensory system, bacteria could be detected to as low as 5 CFU/mL within 10 minutes. Not only is this sufficient for many diagnoses (urinary tract infection >1000 CFU/mL), the system was cheaper, required less sample volume, and has a lower response time compared to commercial silver electrodes. Overall, the low-cost modular silver electrode design would make testing for bacteria more accessible without sacrificing accuracy, sensitivity, and precision.

MACHINE LEARNING APPROACHES FOR NO₂ FORECASTING: DIRECT MODEL PREDICTIONS VS. FOUNDATION MODEL WITH RESIDUAL CORRECTION

Michael Togbe¹, Aryan Poshtiwala², Ray Chen³, Oscar Tuten⁴, Adeline Barden⁵,

Pavan Raj Ravi⁶, Ruoyan Gong⁶, Kyo Hugo Lee⁷, Nicholas LaHaye⁷, Xiaohua Pan⁷,

Hazem Mahmoud⁷,and Jianwu Wang⁶

¹Department of Computer Science and Electrical Engineering, University of Maryland, Baltimore County

²Clarksburg High School, Montgomery County, Maryland

³Department of Computer Science, University of Maryland, College Park

⁴Department of Computer and Data Science, Goucher College

⁵Department of Education, Troy University

⁶Department of Information Systems, University of Maryland, Baltimore County

⁷P.C. Rossin College of Engineering and Applied Science, Lehigh University

⁸ NASA Jet Propulsion Laboratory

⁹ Spatial Informatics Group, LLC

¹⁰ NASA Goddard Space Flight Center

¹¹ NASA Langley Research Center

This work is supported by the grant “REU Site: Online Interdisciplinary Big Data Analytics in Science and Engineering” from the National Science Foundation (grant no. OAC-2348755). We acknowledge the computational resources in the UMBC High Performance Computing Facility (hpcf.umbc.edu) and the financial contributions from NIH, NSF, CIRC, and UMBC for this work.

PROTEOMIC AND MACHINE LEARNING-BASED SUPPLEMENTATION OF CELL-FREE SYSTEMS FOR BIOLOGICAL PRODUCTION

Judy Tran¹, David Garcia¹

¹Department of Chemical, Biochemical, and Environmental Engineering, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

Cell-free protein synthesis (CFPS) is an important tool in synthetic biology used for protein production, biosensing, and therapeutic development. However, one of its biggest limitations is inconsistent protein yields between cell-free extracts. Small differences in the preparation of these extracts can lead to large changes in protein production. This project focuses on determining and understanding why these differences occur and identify the proteins that can improve the performance of low-yield extracts. A library of cell-free extracts with varying levels of activity will be created by modifying the preparation conditions and introducing variability by changing the temperature, overgrowing, or environmental changes. Protein production is measured using GFP fluorescence. Proteomic analysis will be conducted to compare protein composition between high and low-performing extracts to identify proteins that contribute to the differences. Candidate proteins are expressed using linear expression templates (LETs) and are tested individually and in combinations in low-performing extracts, while using an optimization algorithm to help prioritize which proteins and concentrations to test, allowing for a more efficient and machine-learning approach for improving cell-free extract performance. Identifying and understanding the proteins that contribute to CFPS will help provide a strategy for improving the consistency and reliability of CFPS for future research and industrial applications. Currently, LETs for the selected candidate proteins have been successfully generated by PCR and are being quantified for the preparation of protein expression testing.

FROM PAIN TO POWER: THE ROLE OF BLACK PSYCHOLOGICAL EMPOWERMENT AND LIBERATION IN FOSTERING SOCIOPOLITICAL DEVELOPMENT AMONG BLACK YOUTH IMPACTED BY COMMUNITY VIOLENCE

Aanayah Trotman¹, Fantasy Lozada Ph.D²

¹Department of Psychology, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250

²Department of Psychology, Virginia Commonwealth University, 921 West Franklin Street, Richmond, VA 23284

In the present study, we investigated the role of Black psychological empowerment and liberation in fostering sociopolitical development among Black youth impacted by community violence. This study developed a culturally grounded measure of Black psychological empowerment based on Black youths’ reports of empowerment, racial identity, and sociopolitical leadership informed by a literature review on Black psychological empowerment. Although it is not specifically defined, existing research allows us to bridge the gap toward a unified concept of Black psychological empowerment. Data were derived from the Youth Voices program evaluation as part of the larger Healthy Communities for Youth study (see Lozada et al., 2025). The Youth Voices program included 115 Black youth, ages ranging from 14-18 years old, who were in high school (9th-12th grade) and were from communities impacted by community violence. A convergent triangulation mixed-methods design was utilized to: (1) establish a multidimensional measure of Black psychological empowerment, (2) examine the extent to which Black psychological empowerment predicts critical consciousness and sociopolitical development, and (3) examine youths’ narratives of emotional resilience, positive identity development, and sociopolitical engagement. Quantitative measures used to develop the multidimensional Black psychological empowerment construct included positive racial identity measured using the Multidimensional Inventory of Black Identity-Teen version (MIBI-T), empowerment and motivation measured by the Sociopolitical Control Scale for Youth (Opara et al., 2019; Peterson et al., 2011), and agency and leadership measured by the Boston University Empowerment Scale (Rogers et al., 1997). Critical consciousness was measured by the Critical Consciousness Scale (Diemer et al., 2010), which considers critical reflection. An exploratory factor analysis of a social support questionnaire was conducted using the maximum likelihood extraction method, with the ProMax oblique rotation, resulting in four factors were loaded: Awareness and Motivation for Community Engagement (α = .85), Collective Action Beliefs (α = .84), Positive Racial Identity (α = .82), and Positive Self-Esteem/Self-Concept (α = .82). Correlational analyses revealed that the four factors were positively interrelated (rs = .05-.41, p < .001). Regression analyses indicated that collective action beliefs predicted critical action: when Black youth’s beliefs about coming together to make change had a negative influence on their critical action, β = -.31, SE = 0.11, p = .005. Further, positive racial identity predicted critical reflection of inequalities, such that when Black youth have a positive self-racial identity, they can perceive inequalities that occur more often, β = -0.05, SE = 0.20, p = .818. Finally, Black youths’ narratives demonstrate youth being in the earlier stages of sociopolitical development with limited connections between aspects of Black psychological empowerment.

MULTITHREADING PARALLELIZATION OF GPU-CPU-BASED MODELS OF MORPHOGENESIS

Courtland Tucker, Emma Chaney, Daniel Lobo

Department of Biological Sciences, University of Maryland, Baltimore County

Understanding how tissues maintain stable shapes while continuously growing, shrinking, and regenerating is a central challenge in developmental biology. Plastic organisms such as planarians maintain precise shapes and sizes through coordinated feedback between biochemical morphogen signals and mechanical cell-level interactions that are not completely understood. Here, we present a computational study expanding this understanding by developing and analyzing a high-performance, GPU-accelerated agent-based modeling framework for tissue morphogenesis. Prior work has shown that morphogen gradients and mechanical forces jointly regulate cell growth, mitosis, and apoptosis, forming feedback loops that drive emergent tissue shapes. Using a lattice-free, center-based model, which couples diffusive morphogens with cell mechanics and dynamic size regulation, we show how GPU-parallelized simulations can accelerate large-scale tissue modeling. The model represents each cell as a mechanically interacting agent whose growth rate is controlled by local morphogen concentration and crowding, allowing tissues to expand, reorganize, and reach certain precise and steady shapes. Crucially, we have expanded this framework with multithreading algorithms that can compute both GPU and CPU cells in parallel, allowing the simulation, logging, and visualization to run concurrently. Across simulations, we demonstrate that the GPU simulation preserved biological dynamics while reducing the runtime from multiple hours to under two hours, which enabled stress tests of up to 1,000,000 cells. These results support the feasibility of high-resolution, large-scale studies of mechano-chemical feedback in tissue morphogenesis. Overall, the simulations demonstrate how morphogen patterning, mechanical stress, and cell-level behaviors interact to produce stable tissue shapes. This work lays the foundation for future extensions involving dynamic memory allocation and the study of self-organized Turing patterning in growing tissues.

This work was supported by the National Institute of General Medical Sciences of the National Institutes of Health under award number R35GM137953. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Computations used the UMBC High Performance Computing Facility (HPCF).

GALLERIA MELLONELLA LARVAE (WAXWORMS) AS A MODEL FOR STUDYING KLEBSIELLA PNEUMONIAE PATHOGENESIS

Gulnur Tuluoglu, Aakriti Adhikari, Khandra T. Sears

Department of Natural and Physical Sciences, Baltimore City Community College, 2901 Liberty Heights Avenue, Baltimore, MD, 21215

The larval stage (waxworms) of Galleria mellonella, or the greater wax moth, is gaining increasing use as a low cost animal model in biomedical research. Larvae can be used to assess the virulence of bacterial pathogens and innate immune responses to infection. Klebsiella pneumoniae is a Gram-negative rod-shaped bacteria normally found in the human intestinal tract. It causes a variety of infections including pneumonia, urinary tract infections, and hospital acquired infections. Antibiotic resistance in K. pneumoniae strains is also an increasing concern and drives the need for development of preventative measures. We are developing a model of K. pneumoniae infection in fifth instar G. mellonella larvae to assess virulence of clinical strains. Larvae measuring approximately 2-3 cm were used for mortality assays. Groups of 10–13 larvae were injected with 10l of K. pneumoniae (10 to 7.9 × 10⁷ CFU) into the hemocoel through the last proleg. Control groups included non-injected and sham (PBS only) injected larvae. Larvae were monitored for melanization and death for five days post-infection (p.i.). Untreated larvae remained alive for the duration of the experiment while 0% mortality was observed in sham injected larvae 24 hours p.i. Importantly, we observed dose-dependent increases in melanization and mortality in this animal model of K. pneumoniae infection. More than 30% of larvae receiving the highest inocula died within 24 hours p.i., and nearly all remaining larvae died by day 5. Fewer than 20% of larvae receiving the lowest inocula died within 24 hours p.i.; surviving larvae continued development into cocoons. Future goals are to use this model to assess the virulence of clinical isolates and interventions that may limit infection.

This research was supported by Baltimore City Community College.

COMPARATIVE STUDY OF GENOME PACKAGING MECHANISMS IN HIV-1 AND MoMuLV: DIMERIZATION, CAP SEQUESTRATION, AND RNA FATE

Martha Tumwikirize^2,4; Aaliyah Owens^1,3; Brea A. Manuel²; Michael F. Summers^1,2.
¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, 1000 Hilltop Cir, Baltimore, MD 21250
²Howard Hughes Medical Institute, 4000 Jones Bridge Rd, Chevy Chase, MD 20815
³Department of Biology, Howard University, 2400 6th St. NW, Washington, DC 20059
⁴Department of Biology, Virginia Military Institute, 401 Letcher Ave, Lexington, VA 24450

Support for this research was provided by the Howard Hughes Medical Institute (HHMI) and NIH/NIAID grant (#5R01AI150498).

W

ANALYZING TUMOR HOXB13 EXPRESSION IN NEW MODELS OF PROSTATE CANCER

Casey Wehr^1,2; Lily Handwerger, B.S.¹; Alexander Chin, M.S.¹; Charles J. Bieberich, Ph.D.¹

¹Department of Biological Sciences, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

²Howard Hughes Medical Institute, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

Prostate cancer (PCa) is the most commonly diagnosed malignancy and second leading cause of cancer deaths in American men. Nearly all cases of lethal PCa have gain of proto-oncogene MYC and loss of tumor suppressor phosphatase and tensin homolog (PTEN). To study prostate carcinogenesis and metastasis, the Bieberich laboratory developed the genetically engineered BMPC mouse model which uses the Hoxb13 promoter to drive prostate-specific human MYC overexpression and deletion of mouse PTEN, which synergize to promote highly aggressive carcinogenesis. Breeding challenges and lengthy tumor latency in the BMPC model limit our ability to study tumor progression, so we derived several primary mouse cancer cell lines from tumors in end-stage BMPC mice. We hypothesized that our cell lines differ in growth and structure due in part to variable HOXB13 protein expression. To test this hypothesis, we investigated differences in the expression of the HOXB13 protein, mRNA, and DNA. We did not detect HOXB13 levels via immunohistochemistry in implanted tumors derived from one cell line. We followed this with Western Blot, which confirmed the decreased amount of protein. Then, quantitative reverse transcription PCR of Hoxb13 cDNA demonstrated a significantly reduced mRNA expression which was confirmed via RNA-Sequencing. From there, we confirmed that there were no changes or mutations in the DNA for the Hoxb13 gene itself via PCR, cloning, and sanger sequencing. We concluded that Hoxb13 is not lost but has very low expression in one of our BMPC tumor-derived cell lines, which suggests that HOXB13 is not required to sustain progression of PCa.

This work was supported by NIH grants 2R01CA200900-06A1 and 1U54CA274370-01.

DFT STUDIES OF THE 2D QUANTUM MATERIAL GALLIUM SELENIDE

Sarah Will¹; Yussuf Fasasi¹; Zeev Rosenzweig¹; Joseph W. Bennett^1

1Department of Biological Sciences, University of Maryland, Baltimore County, UMBC, 1000 Hilltop Circle, Baltimore, MD 21250

The materials needed for sensing and quantum technology need to exhibit high surface area and well defined electronic and optical properties. Two-dimensional (2D) layered chalcogenides, like GaSe, are favored for their ability to be easily exfoliated and modified. GaSe has already shown sensitivity to changes in surface chemistry and distinct optical properties useful in quantum information science. These properties depend on the crystal structure and impurities of GaSe so we use density functional theory (DFT) calculations to model the structural and electronic effects of vacancies and defects on single layer GaSe. DFT is an efficient and computationally inexpensive way to probe materials at the atomistic level and here we use it to compare total energies of modified GaSe surfaces.

Support for this research was provided by Department of Chemistry and Biochemistry

INVESTIGATING THE SEX-SPECIFIC ROLE OF CHRONIC STRESS IN INDUCING DEPRESSIVE-LIKE BEHAVIORS

Cherub Willmott¹, Nick Anderson¹, Tara LeGates^1,2

¹Department of Biological Sciences, University of Maryland, Baltimore County, UMBC, Baltimore, MD

²Department of Pharmacology and Physiology, University of Maryland School of Medicine, Baltimore, MD

Men and women differ in their susceptibility to stress-related disorders, with women exhibiting greater vulnerability to disorders such as depression. However, the sex-specific mechanisms through which this is mediated remains unclear. Rodents are a common model used to mimic depressive-like behaviors. These models show that plasticity is a key factor in the exhibition of stress-related behavioral deficits. Males and females use different plasticity mechanisms, alongside this, chronic stress can impair long-term synaptic plasticity which may affect susceptibility to depression. Therefore, we focused on the CaV1.2, GluN1 (NR1), GluN2A (N2A), and GluN2B (N2B) proteins which are key components in the regulation of synaptic plasticity. We used mice models to determine the effects of chronic stress on behavior and the sex-specific difference in expression of synaptic proteins. We conducted a Chronic Variable Stress (CVS) paradigm where mice were exposed to physiological stressors daily for 17 days. Sucrose preference and Novelty Suppressed Feeding (NSF) was utilized in accordance with the chronic stress period to test for anhedonia and reward-seeking latency, which are hallmark symptoms of depression. At the conclusion of our behavioral experiments, we isolated the synaptically relevant proteins from the nucleus accumbens (NAc), the central hub of the reward system in the brain that is sensitive to stress. Sex-specific mechanisms of long-term plasticity have been identified, where NMDA receptors are utilized in males and L-type calcium channels in females. We used Western blotting to measure CaV1.2, NR1, N2A, and N2B expression as these proteins play a key role in synaptic strengthening. We found a modest decline in N2A in female mice following chronic stress alongside a significant increase in food reward seeking latency in stressed females. This finding may suggest that sex-specific mechanisms act upon the effects of chronic stress, with N2A possibly playing a role in female synaptic function.

This work was supported by Sarah Snider, Noah Johansson, and Deeya Mistry who aided in conducting the Chronic Variable Stress Paradigm.

This work was also supported by the Thomas F. Roth Undergraduate Research Award and the National Science Foundation (Grant No. IOS2402645 [to TAL])

Y

ANALYSIS OF GENETIC COMPENSATION IN THE GREEN ALGA CHLAMDOMONAS REINHARDTII

Jacob Yang, Robin Bridgman, Dr. Stephen Miller

Department of Biological Sciences, University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

Green algae hold great promise as sustainable fuel producers, so understanding their biology is important for energy security. Under stress, such as phosphate limitation, algae dramatically increase synthesis of triacylglycerols (TAGs), lipids that are readily converted into biofuel. Recent studies showed that when certain genes in the TAG pathway in Chlamydomonas reinhardtii contain premature termination codon (PTC) mutations, related genes were upregulated to compensate, a biological phenomenon known as genetic compensation. Understanding the mechanisms driving this compensatory response will be crucial for optimizing C. reinhardtii to enhance biofuel production. This study investigates whether additional genetic compensation occurs in a mutant when a compensating TAG gene is mutated, and whether nonsense mediated decay (NMD) factor UPF1 is required for compensation, since NMD has been shown to be involved in genetic compensation in other organisms. Here we analyzed the C. reinhardtii double mutants dgtt1/pdat and dgtt1/upf1, as PDAT expression was previously found to increase in a dgtt1 PTC mutant, leading to increased TAG production. To examine expression of TAG genes related to DGTT1 and PDAT in the double mutants, cells were cultivated in phosphorus-deficient media to induce TAG gene expression. Then we used RT qPCR to measure transcript abundance for TAG-related genes DGTT1, DGTT2, DGTT3, and PGD1. We found that the upf1 mutation eliminated compensation in the dgtt1 background, and that DGTT2 and PDG1 transcript accumulation was significantly upregulated in the pdat/dgtt1 double mutant. These results indicate that NMD is required for TAG gene compensation, and that additional compensation can be triggered when a compensating gene is mutated. Our findings related to TAG gene expression in response to homologous TAG gene PTC mutations should shine light on algal TAG gene regulation that can be used to improve biofuel production.

I would like to thank my graduate mentor, Robin Bridgman, and Dr. Stephen Miller for their guidance and feedback throughout this project.

PROGRAMMABLE BIOPLASTICS: HARNESSING ENZYME-SUBSTRATE REACTIONS FOR NEXT-GENERATION SUSTAINABLE MATERIALS

Kyah Young¹, David Garcia¹

¹Department of Chemical, Biochemical, and Environmental Engineering,University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD, 21250

The growing environmental impact of petroleum-based plastics has created an urgent need for more sustainable material alternatives. Alginate-based bioplastics have emerged as a promising biodegradable substitute because they are derived from renewable resources, can be produced using relatively simple and reusable laboratory equipment, and their formulations can be modified to produce materials with a wide range of physical properties. In addition, agar-based formulations such as those used in this project can be remelted and reused, reducing material waste during development. However, current formulations remain brittle and lack the mechanical strength required for many practical applications. Building upon previous work, which established a baseline alginate bioplastic formulation, this project aims to optimize the material composition before the future incorporation of melanin to strengthen the bioplastic matrix. To accomplish this, a high-throughput screening strategy is being developed to systematically evaluate multiple formulations, improve physical properties and reproducibility, and establish a workflow that can be efficiently scaled for future enzyme-substrate screening. Preliminary screening has shown that changes in material composition produce noticeable differences in drying time, shrinkage, and transparency while also revealing reproducibility challenges associated with the high viscosity of the materials, prompting the development of diluted stock solutions to improve consistency. These observations have guided the development of a more reproducible preparation workflow. These improvements will be applied to a high-throughput 96-well screening platform. Identifying an optimal formulation will provide a reliable foundation for future melanin reinforcement and support the development of stronger, more sustainable bioplastics for material applications.

Support for this research was provided by the COEIT Interdisciplinary Grant

Z

ELUCIDATING THE RESIDUE-LEVEL MECHANISM OF ARGINYLTRANSFERASE-1 (ATE1)-CATALYZED PROTEIN ARGINYLATION

Caleb Zeleke¹, Rajat Kumar Jha¹, Aaron T. Smith¹

¹Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, 1000

Hilltop Circle, Baltimore, MD, 21250

Arginylation is a critical eukaryotic post-translational modification that is catalyzed by the enzyme arginyltransferase-1 (ATE1). The ATE1-catalyzed addition of the amino acid Arg to the N-terminus of a mature, translated protein is part of the N-degron pathway that links the degradation of a protein to the identity of its N-terminal amino acid. Although the process of arginylation has been studied at the cellular level, less is known about the mechanism of ATE1 action and its regulation. We have previously shown that the N-terminal regulatory domain of ATE1 contains four conserved Cys residues that coordinate an [Fe-S] cluster to regulate enzyme function. However, how this cluster controls arginylation remains unclear. Additionally, the N-terminal regulatory domain also contains two conserved, vicinal Arg residues that we hypothesize to be essential for both substrate recognition and structural integrity. To probe the role of these residues, we have generated, expressed, and purified Cys and Arg variants of Saccharomyces cerevisiae ATE1 (ScATE1). We have successfully crystallized these variants of ATE1, we have solved the structure of the Cys variant of ScATE1, and we are currently testing the crystals of the Arg variant of ScATE1 for diffraction. To correlate the importance of these residues to function, a luminescence-based ATE1 activity assay has been recently optimized and is currently being used to assess the activity of these protein variants. Once completed, this work will reveal the importance of these residues to ATE1 structure and function, ultimately helping to elucidate the mechanism of this essential post-translational modification.

This investigation was sponsored by the U-RISE Program at the University of Maryland, Baltimore County (UMBC), which is supported by the National Institute Of General Medical Sciences of the National Institutes of Health under Award Number T34GM136497. This work was also supported by NIH grant R35GM133497.

EVALUATION OF MODEL PRECIPITATION FOR POINT-BASED LOCATIONS: HOW MUCH IS IT GOING TO RAIN TONIGHT?

Alex Zielinski¹, Ali Tokay²³, Charles N. Helms³⁴, Stephen D. Nicholls³⁵

¹Department of Climate, Meteorology, and Atmospheric Sciences, University of Illinois at Urbana-Champaign, Urbana, IL 61801

²Department of Geography and Environmental Systems, University of Maryland Baltimore County (UMBC), Baltimore, MD 21250

³NASA Goddard Space Flight Center, Greenbelt, MD 20771

⁴University of Maryland, College Park, MD 20742

⁵Rutgers University, New Brunswick, NJ 08901

Gauge-adjusted and radar-based precipitation products such as MRMS and Stage IV are heavily used to verify model precipitation products, as they create a map of precipitation across the US. Their accuracy at a given location primarily depends on the data availability and quality. A recent study found MRMS and Stage IV are comparable to each other, missing about 10% of rain events and 18-44% errors in rainfall totals. NASA’s Global Precipitation Measurement Ground Validation (GPM-GV) program has been deploying Platforms for In-Situ Estimation Rainfall Systems (PIERS) at various locations across the US since late 2023. This study uses nine (9) PIERS+ sites, which include a PARSIVEL disdrometer and dual tipping bucket gauges, for 2024. The sites are primarily in the Mid-Atlantic, with two sites in Texas and one in Colorado. The PARSIVEL disdrometer defined the events, event duration, and mean and maximum rainfall, while the tipping bucket gauge was the reference for rainfall amount. Unlike previous validation studies where a fixed time scale was selected, this study investigates the event rainfall. This study evaluates the performance of four models: the High Resolution Rapid Refresh (HRRR) model, the North American Mesoscale (NAM) model, the National Severe Storms Laboratory Model for Prediction Across Scales (NSSL-MPAS) model, and the National Severe Storms Laboratory Weather Research and Forecast (NSSL) model. These models all provide hourly forecasts but vary in spatial and temporal resolution. PIERS+ is not part of any operational algorithm and therefore provides independent evaluation. This study follows a farmer’s approach, meaning that it does not investigate the model production stage. Preliminary results indicate the HRRR underestimated rainfall by about 25%, with an average error of 53%, and missed 25% of the events. The NAM underestimated rainfall by about 10%, with an average error of 72%, and missed 40% of the events.

Search UMBC

Poster Sessions:

Abstracts Alphabetical by Last Names

A

COMPLEX REFRACTIVE INDEX OF VOLCANIC ASH RETRIEVED FROM FOURIER TRANSFORM INFRARED SPECTROSCOPY

CHARACTERIZING THE ROLE OF PLEXIN A AND SEMAPHORIN-1A IN DROSOPHILA MELANOGASTER BORDER CELL MIGRATION

GALLERIA MELLONELLA LARVAE (WAXWORMS) AS A MODEL FOR STUDYING ACINETOBACTER BAUMANNII PATHOGENESIS

DECIPHERING THE ADIPOSE TISSUE EXTRACELLULAR MATRIX COMPOSITION ALTERATIONS DRIVEN BY DIET-INDUCED OBESITY

INVESTIGATING THE THERAPEUTIC UTILITY OF METFORMIN IN OVARIAN CANCER

INVESTIGATING CONSERVATION OF THE HIV-1 CORE ENCAPSIDATION SIGNAL

THE IMPACT OF NEBULOSA KNOCKDOWN IN THE FAT BODY ON AGE-DEPENDENT IMMUNOCOMPETENCE IN DROSOPHILA MELANOGASTER

ENHANCING T7 RNA POLYMERASE TRANSCRIPTIONAL HOMOGENEITY FOR STRUCTURAL STUDY USING TEMPLATE MODIFICATIONS

INFLUENCE OF THE CERVICOVAGINAL MICROBIOME ON FIBROBLAST ACTIVATION IN A CERVICAL CANCER MICROPHYSIOLOGICAL SYSTEM

EFFECTS OF ZINC EXPOSURE ON NEURODEGENERATION IN DROSOPHILA MELANOGASTER

INVESTIGATING THE ROLE OF RNAI SNR1 ON CLIMBING SPEED AND ENDURANCE IN DROSOPHILA MELANOGASTER

PARTICLE SIZE DISTRIBUTION ANALYSIS OF GARDEN SOILS AT PRINCE GEORGE’S COMMUNITY COLLEGE

INVESTIGATING THE INFLUENCE OF POLYMER COMPOSITION ON THE PROPERTIES OF GEL ELECTROLYTES

THE EFFECTS OF PARTICLE SHAPE, SIZE, AND COMPOSITION ON LIGHT SCATTERING IN THE ATMOSPHERE

EXPERIMENTAL ASSESSMENT OF LATENT HEAT ENHANCEMENT IN MEPCM-DIELECTRIC SLURRIES FOR MICROCHANNEL ELECTRONIC COOLING

THE MICROBIOME-GUT-BRAIN AXIS IN ATTENTION DEFICIT HYPERACTIVITY DISORDER: ANALYZING THE IMPACT OF SPECIFIC NUTRITIONAL MODULATOR (CAPSAICIN).

DIURNAL EVOLUTION OF CLOUD FRACTION AND SURFACE METEOROLOGICAL VARIABLES ACROSS CONVECTIVE REGIMES OVER THE AMAZON RAINFOREST

B

VALIDATION OF MUSCLE-SPECIFIC GAL4 DRIVER LINES FOR AGING RESEARCH IN DROSOPHILA MELANOGASTER

CASE STUDY ON SMOKE EVOLUTION AND AEROSOL OPTICAL PROPERTIES OVER THE SOUTHWESTERN UNITED STATES

DESIGNING STRUCTURAL MEMBERS USING BIO-INSPIRED DESIGN PRIMITIVES

CELLS BREATHE TOO: RE-ENGINEERING THE T-FLASK FOR IMPROVED DIFFUSION AND OXYGEN TRANSFER

FAILURE-AWARE EPISODIC MEMORY FOR PENETRATION-TESTING AGENTS

CORRELATION OF EASTERN BLUEBIRD (SIALIA SIALIS) AND TREE SWALLOW (TACHYCINETA BICOLOR) FLEDGING SUCCESS IN PAIRED VERSUS UNPAIRED NEST BOXES

DESIGNING STRUCTURAL AND HYDRAULIC MEMBERS USING BIO-INSPIRED DESIGN PRIMITIVES

DETERMINING THE EFFECT OF SET4’S PHD DOMAIN ON GENE REGULATION

MASS SPECTROMETRY IN QUANTIFYING SALIVARY BIOMARKERS OF ORAL CANCER

INVESTIGATING DIFFERENCES IN RNA SPLICING BETWEEN MAL AND NL4-3 HIV-1 STRAINS

AUTOMATED MONITORING SENSORS FOR PROCESS ANALYTICAL TECHNOLOGY APPLICATIONS IN CELL AND GENE THERAPY

C

INVESTIGATING DIFFERENCES IN RNA SPLICING BETWEEN MAL AND NL4-3 HIV-1 STRAINS

THE IMPACT OF NEBULOSA KNOCKDOWN IN THE FAT BODY ON AGE-DEPENDENT IMMUNOCOMPETENCE IN DROSOPHILA MELANOGASTER

UNLOCKING HEALTHIER CELL CULTURES: HOW BREATHABLE SHAKE FLASKS ENHANCE MAMMALIAN CELL CULTURES

BEYOND THE MAP: AI-MEDIATED COUNTERMAPPING AND COMMUNITY NARRATIVES IN BALTIMORE

D

INTEGRATED SENSORS AND CONTROLS TO ACCURATELY GAUGE UAV POSITION

THE IMPACT OF NEBULOSA KNOCKDOWN IN THE FAT BODY ON AGE-DEPENDENT IMMUNOCOMPETENCE IN DROSOPHILA MELANOGASTER

CORRELATION OF EASTERN BLUEBIRD (SIALIA SIALIS) AND TREE SWALLOW (TACHYCINETA BICOLOR) FLEDGING SUCCESS IN PAIRED VERSUS UNPAIRED NEST BOXES

GEANT4 SIMULATION OF Y-90 IMAGING: A DEEP LEARNING FRAMEWORK FOR 3D IMAGE ENHANCEMENT

THE IMPACT OF NEBULOSA KNOCKDOWN IN THE FAT BODY ON AGE-DEPENDENT IMMUNOCOMPETENCE IN DROSOPHILA MELANOGASTER

E

OPTIMIZING T7 TRANSCRIPTIONAL HOMOGENEITY VIA TEMPLATE MODIFICATIONS AND MUTANT ENZYMES TO APPLY SEGMENTAL LABELING WITH MUTANT ENZYME TGK FOR STRUCTURAL STUDIES

F

STRUCTURAL AND FUNCTIONAL ANALYSIS OF THE KH-DOMAIN PROTEIN FLK IN REGULATING ARABIDOPSIS DEFENCE AND DEVELOPMENT

STRUCTURAL CHARACTERIZATION OF RNA DIMERIZATION SITE DOWNSTREAM OF HIV-1 5′-LEADER

THE ROLE OF GPR155 IN MECHANISTIC TARGET OF RAPAMYCIN COMPLEX 1 SIGNALING IN MICE

EVALUATING AIRHARP2 AND HARP2 MULTI-ANGLE MEASUREMENTS DURING NASA’S PACE-PAX CAMPAIGN

G

INVESTIGATING CERAMIDES AND THEIR METABOLITES IN THE BRAIN REGIONS USING MASS SPECTROMETRY IMAGING

H

INVESTIGATING DIFFERENCES IN RNA SPLICING BETWEEN MAL AND NL4-3 HIV-1 STRAINS

EVALUATING ANTIMICROBIAL ACTIVITY IN SOIL SAMPLES: A COMPARATIVE ANALYSIS OF MICROBIAL RESISTANCE

INVESTIGATING THE EFFECTS OF COPPER EXPOSURE ON APP GENE FUNCTION AND ALZHEIMER’S DISEASE RELATED NEURODEGENERATION IN DROSOPHILA MELANOGASTER

DETERMINING MECHANISMS THAT REGULATE USP15 LEVELS IN OVARIAN CANCER CELLS

SUCROSE DENSITY GRADIENT CENTRIFUGATION VERSUS ANION EXCHANGE FOR EXOSOME SAMPLE PURIFICATION

I

DEVELOPMENT OF DIRECT INK WRITING TECHNIQUES OF A FE3O4-SILICONE BOUND MAGNETIC SLURRY USING A FE3O4 VOLUME FRACTION OF 20%

INVESTIGATING THE ROLE OF PERIOD ON CLIMBING SPEED AND ENDURANCE IN DROSOPHILA MELANOGASTER

J

VECTOR-FREE SYNTHETIC CIRCULAR SUPERCOILED DNA (SCSDNA) ENABLES RAPID CELL-FREE PRODUCTION FOR THERAPEUTIC DEVELOPMENT WORKFLOWS

INVESTIGATING SEX-DEPENDENT BEHAVIORAL AND MOLECULAR EFFECTS OF CHRONIC ETHANOL EXPOSURE IN MICE AT THE NUCLEUS ACCUMBENS-HIPPOCAMPUS PATHWAY

INVESTIGATING CONSERVATION OF THE HIV-1 CORE ENCAPSIDATION SIGNAL

K

EXPERIMENTAL PROMPT GAMMA SCATTER CLASSIFICATION AND REORIENTATION FOR PROTON BEAM IMAGE RECONSTRUCTION

INVESTIGATING THE ROLE OF RNAI SNR1 ON CLIMBING SPEED AND ENDURANCE IN DROSOPHILA MELANOGASTER

DEVELOPMENT OF A VAGINA-ON-A-CHIP MODEL WITH A PHYSIOLOGICALLY RELEVANT OXYGEN GRADIENT TO STUDY BACTERIAL VAGINOSIS

INVESTIGATING THE EFFECTS OF MATERNAL CARE ON DNMT3AB EXPRESSION ACROSS DEVELOPMENT STAGES IN AN AFRICAN CICHLID

DFT STUDIES OF GaS, A 2D QUANTUM MATERIAL

L

DFT STUDIES OF THE 2D QUANTUM MATERIAL InSe

OPTIMIZING ASSEMBLY OF THE HIV-1 GAG-RNA NUCLEATION COMPLEX

EFFECT OF ANCHOR KNOCKDOWN ON INSULIN PRODUCTION IN THE INSULIN PRODUCING CELLS OF DROSOPHILA MELANOGASTER

SUBMERGED CULTURE VISCOSITY (RHEOMETRY) OF ASPERGILLUS NIDULANS KINASE DELETION STRAINS

M

UTILIZATION OF WORLD AND HEAD-CENTRIC REST-FRAMES IN A HIGH-SPEED VR ENVIRONMENT

INVESTIGATING THE ROLE OF MYELOID-DERIVED TRANSGLUTAMINASES IN DIET-INDUCED METABOLIC DYSFUNCTION

**CHARACTERIZING THE ROLE OF PLEXIN A AND SEMAPHORIN-1A IN DROSOPHILA MELANOGASTER BORDER CELL MIGRATION**

**GALLERIA MELLONELLA LARVAE (WAXWORMS) AS A MODEL FOR STUDYING ACINETOBACTER BAUMANNII PATHOGENESIS**

**THE IMPACT OF NEBULOSA KNOCKDOWN IN THE FAT BODY ON AGE-DEPENDENT IMMUNOCOMPETENCE IN DROSOPHILA MELANOGASTER**

**THE IMPACT OF NEBULOSA KNOCKDOWN IN THE FAT BODY ON AGE-DEPENDENT IMMUNOCOMPETENCE IN DROSOPHILA MELANOGASTER**

**THE IMPACT OF NEBULOSA KNOCKDOWN IN THE FAT BODY ON AGE-DEPENDENT IMMUNOCOMPETENCE IN DROSOPHILA MELANOGASTER**

**THE IMPACT OF NEBULOSA KNOCKDOWN IN THE FAT BODY ON AGE-DEPENDENT IMMUNOCOMPETENCE IN DROSOPHILA MELANOGASTER**

DEVELOPMENT OF DIRECT INK WRITING TECHNIQUES OF A FE₃O₄-SILICONE BOUND MAGNETIC SLURRY USING A FE₃O₄ VOLUME FRACTION OF 20%

**SUBMERGED CULTURE VISCOSITY (RHEOMETRY) OF ASPERGILLUS NIDULANS KINASE DELETION STRAINS**

DEVELOPMENT OF DIRECT INK WRITING TECHNIQUES OF A FE₃O₄-SILICONE BOUND MAGNETIC SLURRY USING A FE₃O₄ VOLUME FRACTION OF 20%

FORMALDEHYDE STIMULATES CYTOSOLIC CA²⁺RESPONSE IN OLFACTORY NEURONS INDEPENDENT OF CANONICAL ODOR DETECTION PATHWAY

DEVELOPMENT OF A LOW-COST ATMOSPHERIC AMMONIA (NH₃) MEASUREMENT SYSTEM

MACHINE LEARNING APPROACHES FOR NO₂ FORECASTING: DIRECT MODEL PREDICTIONS VS. FOUNDATION MODEL WITH RESIDUAL CORRECTION

MACHINE LEARNING APPROACHES FOR NO₂ FORECASTING: DIRECT MODEL PREDICTIONS VS. FOUNDATION MODEL WITH RESIDUAL CORRECTION