Author(s): Joie Dooloukas, Rachel Towner
Department(s): Biological Sciences
Shade tolerance of invasive Callery pear and Amur honeysuckle: Implications for forest understory invasion
Abstract: Callery (Bradford) pear (Pyrus calleryana) and Amur honeysuckle (Lonicera maackii) are invasive woody plants that become established in open or disturbed areas. Whether they persist depends on their shade tolerance, which determines whether they can survive in the forest understory. Honeysuckle has some shade tolerance, but pear’s tolerance is currently unknown. Shade tolerance is determined from curves of photosynthesis rates against light levels. Stomatal conductance can also limit photosynthesis rates. Pear and honeysuckle were compared to shade-intolerant red cedar and shade-tolerant white snakeroot. Pear and honeysuckle had maximum photosynthesis rates that were similar to red cedar and much higher than snakeroot. Conductance rates of pear and honeysuckle were similar. Honeysuckle is known to acclimate to low light levels. Whether pear exhibits similar plasticity is currently unknown and is an avenue for future research. However, in the absence of photosynthetic plasticity, pear is not expected to invade forest understories.
Author(s): Sei Lillard, Mackenzie Feltner, Brooke Wetzel, Taylor Easybuck, Elliot Glos, Christine Perdan
Department(s): Biological Sciences
Can Exercise Mitigate the Neurotoxic Effects of Neonatal Benzo[a]pyrene Exposure?
Abstract: Air pollution is a global problem and often highest in urban areas with high traffic, industry and fossil fuel burning power plants. This means pregnant women and their babies are likely to be exposed. Studies in New York City found high levels of polycyclic aromatic hydrocarbons in umbilical cord blood and deficits in IQ and behavior when exposed children reached school age. We use a mouse model and the pollutant benzo[a]pyrene (BaP) to see if genetic differences increase susceptibility to developmental neurotoxicity. We are also using 1h of daily exercise on running wheels to see if either maternal or offspring exercise can improve brain function. To study learning and memory, we used the Morris water maze once exposed offspring reach young adulthood. We also used snapshot ethograms to verify how much each mouse run. Therefore, we will be able to determine if there are sex, genotype, treatment or exercise effects.
Author(s): Renn Spigelmyer, Sei Lillard, Brooke Wetzel, Mackenzie Feltner and Christine Perdan Curran
Department(s): Biological Sciences
The effects of early life social isolation and genotype on cognitive functions and the stress response in adult mice
Abstract: Social isolation stress presents in a variety of scenarios, and in some cases, it can have persistent effects on health and behavior. This project examines the effects of early life social isolation and genetic differences in Cyp1b1 on cognitive function and the stress response. Cyp1b1(+/+) wild type and Cyp1b1(+/-) heterozygous mice were randomly assigned to paired or isolated housing at postnatal day 30 (P30). Behavioral testing began at P60. To look at the stress response and activation of the autonomic nervous system, we recorded blood pressure and corticosterone levels in plasma. Our behavioral tests include the Y-maze for spatial learning and memory, a forced swim test to examine coping, and two tests of anxiety-like behavior (Zero Maze and Marble Burying). In our future studies, we will also include Cyp1b1(-/-) knockout mice.
Author(s): Brooke Wetzel, Mackenzie Feltner, Sei Lillard, Leah Bishop, Taylor Easybuck, Deaysha Fox, Emma Lallai, Elliot Glos, Renn Spigelmyer and Christine Perdan Curran
Department(s): Biological Sciences
The Impact of Prenatal Benzo(a)pyrene Exposure on Offspring Motor Function in Cyp1b1 Knockout and Wild-type Mice
Abstract: The widespread carcinogen and neurotoxicant benzo(a)pyrene (BaP) is a polycyclic aromatic hydrocarbon produced during incomplete combustion. It is found in grilled food, cigarette smoke, vehicle exhaust, or any type of fire; therefore, exposure can occur through both ingestion and inhalation. To determine if genetic differences affect susceptibility to neonatal BaP exposure, we compared two genotypes of mice. Cyp1b1(-/-) knockout mice were modified to have a nonfunctional CYP1B1 enzyme which can metabolize BaP. These mice were compared with wild type Cyp1b1(+/+) mice. Pregnant mice were treated from mid-gestation until postnatal day 25 when pups were weaned, so all exposure came through the mother. In addition, we used 1h of daily running wheel exercise as an intervention to determine if maternal or offspring exercise could mitigate the neurotoxic effects of BaP. We used the Rotarod test to examine motor function and motor memory in control and exposed offspring during early adulthood.
Author(s): Christina Gogzheyan, Dawson Hollingsworth, Soumya Sagar Dey, Sudipta Panja, Pravin Yeapuri, Susmita Sil, R L Mosley, Howard E. Gendelman
Department(s): Biological Sciences
Effect of the Insecticide Fipronil on Oviposition Choice for Aedes Mosquitoes
Abstract: Fipronil is a broad-spectrum insecticide which blocks transmission of nerve cells of insects, ultimately leading to death. This experiment set out to discover if female Aedes mosquitoes preferred ovipositing in fipronil-infused containers. Aedes mosquitoes lay their eggs on the circumference of containers above the waterline. Five replicates of three 5-gallon buckets were filled with 18 liters of grass infusion and one of the following treatments: a grass infusion control, pure fipronil, or Fipronil-Plus-C. To collect the eggs, seed-germination papers were clipped to the inside of the buckets, completely covering the circumference. Egg papers were collected daily for 8 weeks. Aedes eggs were hand-counted under a dissecting microscope. There was no significant difference in the number of eggs laid in any of the buckets, showing that Aedes mosquitoes do not have a preference when ovipositing suggesting that the use of fipronil will not affect Aedes oviposition behavior.
Author(s): Nadia Cline, Kel Daniel, and Dr. Allison Parker
Department(s): Biological Sciences
Effect of the Insecticide Fipronil on Oviposition Choice for Culex Mosquitoes
Abstract: This study examines the effect of the insecticide, fipronil, on oviposition (egg-laying) choice of Culex female mosquitoes. Culex mosquitoes are commonly found in the United States, and are vectors for disease-causing pathogens such as West Nile virus. Female Culex mosquitoes lay 100-300 eggs on the surface of the water in man-made containers, which stick together to form egg rafts. Egg rafts were collected daily from five sets of three five-gallon buckets, one control and two fipronil mixtures, over the span of 8 weeks at NKU’s Research and Education Field Station. Significantly more egg rafts were laid in both the pure fipronil and Fipronil + C compared to the grass infusion. However, female Culex mosquitoes did not show significant preference between the two types of fipronil. Female Culex mosquitoes prefer to oviposit in buckets with insecticide versus buckets without.
Author(s): Linnea Gault, Isabella Payne, and Dr. Allison Parker
Department(s): Biological Sciences
Evaluation of Mulberry Infusion as an ‘Attract & Kill’ Trap for
Culex Mosquitoes
Abstract: The goal of this project is to determine if invasive mulberry (Morus rubra) fruit & leaf infusions are attractive to ovipositing (egg-laying) mosquitoes, but lethal to the resulting mosquito larvae. Three infusions were created from grass, mulberry leaves and mulberry berries set in three buckets throughout 5 sites at the NKU’s Research and Education Field Station. Culex mosquito egg rafts were collected daily for 7 weeks. To determine if the infusions are toxic to Culex larvae, 20 larvae were placed into containers with one of the three whole infusions. All larvae were monitored daily to see if any larvae had died or pupated. Results show that there’s no significant preference between our grass and leaf infusions, however there is a significant preference for mulberry leavfs or grass infusions over mulberry berry infusion. The toxicology tests are currently ongoing.
Author(s): Ethan Waddle, Allison Parker
Department(s): Biological Sciences
Evaluation of Tire Traps for Mosquito Control in Urban and Rural Settings
Abstract: Mosquito Species Distribution in Discarded Tires in Northern Kentucky from 2021 to 2025
This study investigates the use of discarded tires as rainwater collection containers to attract mosquitoes. Discarded tires, often found in landfills and natural areas, provide an ideal habitat for mosquitoes to lay eggs. Our research aims to identify mosquito species distribution in discarded vehicle tires. Mosquito sampling was conducted weekly at five field sites in Boone County, Kentucky from discarded tires containing water and organic materials. Larvae and pupae were collected from tires annually from May to August 2021-2025. All collected mosquitoes were identified to species. For 2025, there were a total of 3897 mosquitoes collected across eight species. Discarded tires are significant to the surveillance of mosquito species in the Northern Kentucky region.
Author(s): Shianah Johns and Benjamin Richards
Department(s): Biological Sciences
Pck1 and Pck2 gene expression patterns in the digestive and respiratory organ buds of Xenopus laevis embryos
Abstract: Within the cells of our body, there are a multitude of genes performing unique roles to keep us functioning. I hypothesized that these genes might also be expressed during embryonic development in the liver bud. I used an experiment called in situ hybridization to see whether pck1 and pck2 genes are present during the early development of the gut tube and respiratory organs. I found that pck1 and pck2 are expressed in the embryonic liver bud, as well as in additional digestive organs like the intestine, at different developmental stages. These results have helped us track pck gene activity and show that these genes have complex and unique roles during organogenesis.
Author(s): Autumn Jouett, Leah Grunert, Hailey Ausbrooks, Meredith Beiting, Dr. Emily Shifley
Department(s): Biological Sciences
Investigating the role of the Retinoic Acid pathways in controlling gene expression during pharyngeal development
Abstract: The development of the pharynx uses protein signaling molecules to help cells communicate and differentiate. If these signals are disrupted, it can cause birth defects. One signaling pathway is the Retinoic Acid pathway (RA). We hypothesized that the RA pathway was important for pharyngeal development in Xenopus laevis during early embryonic stages. We used in-situ hybridization to see if blocking the RA pathway caused reductions in pharyngeal gene expression during different stages of development. Results demonstrated that early or long inhibitions of the RA pathway did disrupt certain gene expression patterns, illustrating the importance of this pathway for pharyngeal development.
Author(s): Leah Grunert, Autum, Hailey, Meredith Beiting
Department(s): Biological Sciences
Investigating differential expression of paralogous genes SAM1 and SAM2 in Saccharomyces cerevisiae
Abstract: In Saccharomyces cerevisiae, the paralogous genes SAM1 and SAM2 encode AdoMet synthase, which regulates AdoMet levels in cells. Although both SAM1 and SAM2 encode the same enzyme, deletions of these genes differentially impact genome stability and AdoMet levels, suggesting functional differences. This project seeks to better understand how these genes are differentially expressed. To explore differences in expression, we have created and started testing reporter constructs, containing SAM1 or SAM2 promoter regions linked to LacZ, to identify differences in promoter activity and expression. Additionally, we have created various truncated SAM1 and SAM2 promoters with the aim of identifying promoter regions essential for gene expression control. Ultimately these studies will allow us to further elucidate differences between these two genes and how they are controlled and utilized in the cell.
Author(s): Yara Abdou, Daisy Grove, PhD, and Erin D. Strome, PhD
Department(s): Biological Sciences
Deletion of purine synthesis genes have possible link to genome instability in Saccharomyces Cerevisiae
Abstract: A precursor to cancer development is genome instability, which allows mutations to occur rapidly. Several genes play key components in building purines from either salvage or de novo biosynthesis pathways. The goal of this project is to investigate the removal of these genes to test for resulting increases or decreases in instability within the genome. We were interested in these pathways because our lab studies the impacts of availability of AdoMet, the main methyl donor in cells, and purine levels are linked to AdoMet availability. Using Saccharomyces cerevisiae as our model, we aimed to delete a set of genes from these pathways. We successfully made full knockouts of the genes ADO1 and ADK1 and were able to preliminarily test for impacts on genome stability. Creation and testing of other mutants is still underway.
Author(s): Dominic Reinhart, Daisy Grove, Erin D. Strome
Department(s): Biological Sciences
Exploring The Effect On Genome Stability Due To Deletion of Genes Of The Folate Cycle
Abstract: Methylation reactions are common in cells. SAM1 and SAM2 both code for the same enzyme, that enzyme being S-AdenosylMethionine synthetase, which makes S-AdenosylMethionine, also known as AdoMet. AdoMet donates a methyl group in hundreds of reactions that happen in cells. Previous research has shown that deleting SAM1 increases genome stability, but deleting SAM2 increases genome instability. In our lab, we are working to narrow down specific reactions/pathways impacted by the sam1 and/or sam2 loss that lead to genome stability effects. For my summer project, I was investigating nine genes with roles in the folate cycle: SHM1, MET7, MTD1, SHM2, ADE3, MET12, MET13, DFR1, and MIS1. The folate cycle is a series of conversions of one carbon units, based on folate/vitamin B9, that have vital roles in DNA and RNA synthesis. We aimed to delete each gene and then test to see if it causes a change in genome stability.
Author(s): Nhyira Sarpong, Daisy Grove, Erin D. Strome
Department(s): Biological Sciences
Investigating the Roles of Pyrimidine Synthesis and Salvage Pathway Genes in Maintaining Genomic Stability in Saccharomyces cerevisiae
Abstract: Genomic instability is an early step in cancer development. Evidence from many model systems have identified genes with effects on stability. In this study, eight genes of interest, DUT1, RNR3, CDC21, YNK1, URA6, RNR4, FUR1, and CDC8 found in the Saccharomyces cerevisiae model system were targeted for examination and deletion. All of these genes function in biosynthesis and/or salvage of pyrimidine nucleotides. In S. cerevisiae, pyrimidine biosynthesis and salvage are regulated to maintain nucleotide balance. Disruption of these pathways can lead to replication stress and genomic instability, offering insight into similar mechanisms that might drive cancer progression in humans. Through experimental research, these genes of interest were targeted for deletion, with the goal to then test if their loss impacts genome stability. Further research aimed at understanding the importance of each gene seeks to provide deeper insight into the potential relevance of homologs in cancer development.
Author(s): Hannah Childs, Daisy Grove, Erin D. Strome
Department(s): Biological Sciences
Testing the impact Of S-Adenosylmethionine-Dependent Methyltransferases on Genome Stability
Abstract: SAM1 and SAM2 are genes that encode the protein that synthesizes S-AdenosylMethionine (AdoMet). AdoMet serves as the main methyl donor in all cells and is generated through the methyl cycle. Our lab previously discovered that the deletion of either SAM1 or SAM2 affects the concentration of AdoMet and alters genome stability, a hallmark of cancer. Interestingly, the deletion of sam1 and sam2 have opposite impacts on these phenotypes, despite encoding the same enzyme. The mechanism by which altered AdoMet affects genome integrity has not been fully elucidated. This project aims to investigate the impact of AdoMet-Dependent MethylTransferases (ADMTs) on genome instability. With over 80 ADMTs in Saccharomyces cerevisiae, we begin our study focused on 21 ADMTs whose expression is altered due to loss of SAM1 or SAM2. We will individually knockout the genes encoding ADMTs and then analyse genome instability rates to determine the impact of each gene.
Author(s): Tuyet Do, Daisy Grove, Erin D. Strome
Department(s): Biological Sciences
Ovarian Zinc Transporter Expression and Squirrelfish Reproduction
Abstract: Zinc is an essential micronutrient required for a variety of functions. Female squirrelfish (Holocentrus adscensionis) hyperaccumulate zinc in their liver and ovaries during reproduction, reaching levels up to 500 times higher than that typically observed in vertebrates. Prior to spawning, zinc is redistributed from the liver to the ovaries, where it supports development and embryo viability. This transport is likely due to ovarian uptake of the zinc-binding protein vitellogenin and the impact of the ZIP and ZnT families of cellular zinc transporters. This project tests the hypothesis that ovarian zinc transporter gene expression varies with reproductive stage. Specifically, we aim to characterize the timing and extent of this expression relative to reproductive status. Samples were collected during two reproductive stages, and gene expression was assessed using quantitative PCR. Understanding how zinc transporters contribute to reproductive zinc dynamics in squirrelfish may reveal novel insights into trace metal regulation during vertebrate reproduction.
Author(s): Madison Spangler, Dr. David Thompson
Department(s): Biological Sciences
Epithelial Zinc Transport Associated with Reproduction in Squirrelfish
Abstract: Epithelial Zinc Transport Associated with Reproduction in Squirrelfish
Blake McAndrew, Dr. David Thompson
Studies in squirrelfish (Holocentridae) indicate that females exhibit heightened hepatic zinc levels during reproductive periods. This zinc is then shuttled to the ovary and the developing oocyte prior to spawning, where high zinc levels are thought to be needed to produce viable eggs. Cellular zinc transport is primarily governed by the actions of the ZIP and ZNT1 families of membrane transporters. The hypothesis of this project is that, in order to accomplish zinc hyperaccumulation from the environment, female squirrelfish are differentially expressing cellular zinc transporters compared to male squirrelfish in transport epithelia such as the gill. In this study cDNA was constructed from RNA isolated from male and female squirrelfish gills and subjected to quantitative PCR analysis to explore differences in gene expression between the sexes.
Author(s): Blake McAndrew, Dr. David Thompson
Department(s): Biological Sciences
Helminth treatment effects on hippocampal dependent context object discrimination memory and microglia morphology
Abstract: Commensalist parasites may provide anti-inflammatory signals that diminish disease-causing inflammation. Early life inflammation has a significant effect on behavior and brain cell morphology, especially when combined with inflammation in adulthood (Williamson et al., 2016). This study assesses the interactions between helminth (commensalist worm) treatment, early-life and adult inflammation, and sex on a) hippocampal dependent memory using context object discrimination (COD) testing, and b) microglia density within the hippocampus. Maternal rats received either worms or saline. Pups then received a treatment of either PBS or E. coli on postnatal day 4. Adult treatment was administered during day 1 of COD by an injection of either saline or lipopolysaccharide (LPS). We measured a sex-dependent difference in COD behaviors and are analyzing the brain’s immune cells, microglia. This study will allow us to further understand immune stressors, their effects on the brain and behavior, and possible interventions.
Author(s): Alexis Smith, Logan Hein, and Lauren L. Williamson, Ph.D.
Department(s): Biological Sciences
Effects of Early-life Acetaminophen On hippocampus dependent learning and neuroinflammation
Abstract: This study investigates the effects of early life acetaminophen exposure on contextual fear learning in neonatal rodents. Acetaminophen is a commonly used fever reducer and analgesic for babies and children. We hypothesize that it affects cognitive function if given in early life and may correlate with the increased incidence of autism spectrum disorder. We treated pups with either acetaminophen (150mg/kg, 3X daily) or saline from postnatal day (P)4 to P10. We assessed cognitive function with the contextual fear conditioning (CFC) test in juveniles, P24 to P27. Previously, we assessed early-life acetaminophen exposure and its effects on juvenile cognition in the context-object discrimination task. We will show comparisons of the two hippocampus-dependent behavioral tasks and some cellular and molecular outcomes from early acetaminophen exposure. We analyzed changes in microglial morphology, within the hippocampus region specifically in the CA1, DG, and CA3 regions to understand neuroimmune development following acetaminophen exposure.
Author(s): Chloe Routt, Alfredo Alvarez, Miranda Smith, and Lauren L. Williamson, PhD.
Department(s): Biological Sciences
Assessment of fungi diversity and associated environment in forested wetlands
Abstract: Fungi, both as microorganisms and macro fungi, often recognized as mushrooms, play a crucial role in nutrient cycling, decomposition of decaying materials, and as a food source in ecosystems. Variety of fungi in the forested wetlands in northern Kentucky were observed; however, it is still not clear about the diversity and association with the environmental conditions. Surveys on mushrooms along with slimes and molds were conducted weekly in fall 2024 and spring and summer 2025 at Northern Kentucky University’s Research and Education Field Center (REFS) to assess fungal diversity and associated environmental conditions. So far 216 occurrences of fungi were recorded and 82 genera of them were identified. Most fungi surveyed were present on decaying wood, with occurrences on the hiking trails, and occasionally trees. Survey results also suggested varied fungal diversity and seasonality between sites with or without invasive plants, which provide valuable information for studying wetland fungi.
Author(s): Alex Adams, Lily Zerhusen, and Yingying Xie
Department(s): Biological Sciences
Pond-breeding amphibian phenology and the association with pond water conditions
Abstract: Pond-breeding amphibians have important roles in freshwater ecosystems, which are threatened by environmental changes. This project investigates the reproduction timing and amphibian development and their association with ephemeral water conditions in forested wetlands. Occurrence of amphibian life stages and water conditions were monitored across eight ponds at NKU REFS in spring and summer 2025. Early-season breeding amphibians were observed (wood frogs, Jefferson salamanders, and Spring Peepers). Activity increased with rising temperature but declined sharply in late June as ephemeral ponds dried. Our results highlight the vulnerability of amphibian reproduction to environmental changes, which may disrupt timing and reduce breeding success.
Author(s): Griffin Jones, Will Rogers, Samuel Chapman, Dr. Yingying Xie
Department(s): Biological Sciences
The temporal synchrony of female and male flowering time of Northern Spicebush and the trait comparison
Abstract: Climate change affects the timing of seasonal life events of plants, such as flowering time. Previous studies suggested earlier male flowering time than females for wind- pollinated dioecious plants. However, it is still unknown how the female and male flowering time synchronize for insect-pollinated dioecious plants. In this study we focus on a native shrub in the United States, Northern spicebush (Lindera benzoin). Female and male individuals were randomly selected at NKU REFS. We observed the flowering phenology twice a week in spring 2024 and 2025 and found that males had earlier flowering time than females. Interestingly, we found individuals producing both male and female flowers on the same branch (i.e., monoecious), which has been rarely documented. We also measured morphological and physiological traits for all individual plants to compare among female, male and monoecious individuals. This study improves our understanding in dioecious plants flowering behaviours and their morphology and physiology.
Author(s): Jayme Burks, Lily Zerhusen, Yingying Xie
Department(s): Biological Sciences
Invasive plants gain early timing advantages over native species captured by time-lapse cameras
Abstract: Invasive plant species gain a competitive edge over natives through their distinct timing of seasonal activities, explained by the priority effect: a proposed hypothesis in which invasive species gain more light and nutrients through early leaf emergence. However, it is still not clear what timing advantages facilitate plant invasion on Nantucket Island, Massachusetts. This study was conducted in Nantucket Island to compare the growth patterns of native and invasive species across nine field sites in 2023 and 2024. Time-lapse cameras were set up to take hourly images of the plants to measure the change in plant greenness across the seasons. Invasive species consistently emerged sooner and remained active longer, resulting in an extended growing season compared to native species. This research highlights the use of cameras in invasive species monitoring and will inform conservation management.
Author(s): Raymond Than, Jamey Herbst, Yingying Xie
Department(s): Chemistry & Biochemistry
An investigation of toxic substances in UV and non-UV tattoo inks
Abstract: UV tattoo ink is an increasingly used form of ink. This study investigated the amounts of heavy metals that were found in the UV colors of pink, orange, green, white, and yellow. The experimental tattoo inks were diluted and filtered, then analyzed visually. Copper, Zinc, and phosphate were run through an ICP-OES and HPLC to find a curve to compare the inks. The results showed that UV inks were far more polar and much easily filtered out compared to normal ink. The UV inks also formed an unusual residue. These findings suggest the UV inks exhibit unique properties unlike the normal inks. There should be more research done to describe concentration of heavy metals and the chemical composition.
Author(s): Ella Betz, Macey Ruth, and Charlisa Daniels
Department(s): Chemistry & Biochemistry
Determination of the effect on translation of 2’-O-methylation at positions 32 and 34 in eukaryotic tRNA
Abstract: Post-transcriptional modification of tRNA is vital for protein translation. Mutations in the yeast methyltransferase TRM7 cause intellectual disability and slow cell growth. Trm7 interacts with Trm732 and Trm734 to perform 2’-O-methylation at C32 and G34, respectively, on tRNAPhe. Loss of both modifications impairs yeast growth; however, lack of only one does not. Since both UUU and UUC codons encode Phe, it is unclear how individual modifications, and the specific codons affect translation. Additionally, we are assessing the effects of UGG and UUA codons, decoded by other tRNAs with Nm32 and Nm34 modifications, on GFP expression. Using RFP and GFP via the RNA-ID reporter system in yeast enables analysis of how tRNA modifications affect translation. Codons were added to GFP to test translation in trm732∆, trm734∆, and wildtype strains by comparing GFP/RFP ratios. This research clarifies how Trm7-dependent tRNA modifications influence codon-specific translation in yeast and potentially in other eukaryotes.
Author(s): Danika Maki, Dipshika Rai, Holly M. Funk, Anh Bui, Linh T Le, and Michael P. Guy
Department(s): Chemistry & Biochemistry
Determining Proteins Responsible for tRNA Binding
Abstract: The tRNA anticodon loop must undergo several post-transcriptional modifications for proper protein translation to occur. In yeast, the protein Trm7 forms a complex with Trm732 and Trm734 to methylate nucleotides C32 and G34, respectively. Mutations causing a lack of these two methylations result in slow growth in yeast. Whereas in humans, these same mutations in the homolog FTSJ1 lead to intellectual disabilities. Both proteins in each complex are needed for the methylations to occur, but the exact roles of Trm732 and Trm734 in the methylation reaction. The goal of this research is to determine if the main function of Trm732 and Trm734 is binding the protein complex to tRNA. Our results show that by themselves Trm732 and Trm734 are unable to bind to tRNAPhe. We are also testing Trm732 and Trm734 variants in complex with Trm7 for their tRNA binding ability.
Author(s): Julia Verhoff, Ashton Davey, Anabel Lillie, Alisha Detmer, Alex Nystrom, Holly Funk, and Michael P. Guy
Department(s): Chemistry & Biochemistry
Determining the Role of Yeast tRNA Body Modifications in Protein Translation
Abstract: Post-transcriptional modifications are critical for tRNA function and, therefore, protein translation. Nearly all modifications are conserved between yeast and humans. While yeast lacking a single tRNA body modification is usually healthy, humans lacking a single tRNA body modification have issues ranging from lung cancer to intellectual disabilities. Most body modifications appear to protect tRNAs from degradation by quality control pathways. Because modification enzymes act on many different tRNAs, it is unclear which specific tRNA drives disease, although, in yeast, only one tRNA is often critical. We are investigating the roles of tRNA body modifications on individual tRNAs in yeast mutants using a tRNA engineered to read through a stop codon in GFP. If a modification is important for the function of a given tRNA, GFP expression will be reduced in mutants lacking the modification. Identification of important tRNAs in yeast may give clues to which tRNAs drive disease in humans.
Author(s): Alisha Detmer, Drew Scott, Julia Verhoff, Maira Faisal, Michael P. Guy
Department(s): Chemistry & Biochemistry
Identifying Important Residues for Trm732 Function via Random Mutagenesis and Next Generation Sequencing
Abstract: Post-transcriptional tRNA modifications are crucial for efficient protein translation. In yeast, the Trm7 methyltransferase, with Trm732, modifies tRNAPhe at position 32. Defects in human TRM7 cause cognitive impairment; though Trm732’s mechanism as an auxiliary protein remains unclear, two key motifs have been identified in the DUF2428 domain. We seek the essential residues within the entirety of Trm732. To find them, randomly mutated Trm732 variants will be expressed in a sick yeast strain lacking Trm732. DNA from the colonies will be sequenced before and after restrictive growth conditions to track mutations quantitatively. Mutations that decrease in frequency or disappear indicate Trm732 functional importance and will be identified for future testing of their impact. As a proof of principle, we have optimized conditions of the aforementioned experimental framework using known mutants. We now await the ongoing generation of a Trm732 mutagenic library to help discover the functionally significant residues.
Author(s): Maira Faisal, Johannes Smal, Natalie N. Creech, Holly M. Funk, and Michael P. Guy
Department(s): Chemistry & Biochemistry
Photodegradation of 6PPD and 4HDPA
Abstract: 6PPD is a chemical additive to tires which reduces ozonation and extends tire lifespans, and which enters the environment as tires wear down. It has been found in air and waterways, and its quinone derivative is toxic to certain species of fish. Also, 6PPD quickly transforms into 4-HDPA and other products in water. All of these compounds undergo UV photodegradation. Their chemical properties were analyzed using UV-Vis spectroscopy, UV exposure, and LCMS. Experiments were conducted in methanol, ethanol, acetonitrile, and aqueous solutions at pH’s 2, 7 and 12. Both 6PPD and 4HPDA produce a variety of solvent-dependent products. The rate of the compounds breakdown has been observed to correlate with specific product production. Improvements in understanding solvent dependent reaction pathways have been made, while questions remain for potential products and behavior in different pH solutions. Identifying these products is necessary to evaluate their potential environmental toxicity.
Author(s): Taylor Adams, Methmi Kasturiratna, Emily Wymer and Patrick M. Hare
Department(s): Chemistry & Biochemistry
Time-resolved fluorescence spectroscopy of estrogens
Abstract: The natural estrogens estrone and 17-beta-estradiol share a phenol ring that gives rise to a reasonably strong absorption at ~280 nm and fluorescence at ~304 nm. Despite the shared structure, the fluorescence quantum yield of estrone is smaller by an order of magnitude, and it has a biexponential fluorescence lifetime that is dominated by a sub-nanosecond component. Estrone also possesses an additional fluorescence band at ~410 nm from its carbonyl. The 13-methyl epimer of estrone has fluorescence properties that are much more similar to 17-beta-estradiol than estrone, including a higher fluorescence quantum yield and a stronger nanosecond lifetime component. Steady-state and time-resolved fluorescence spectroscopy in several solvents, supported by computational chemistry, was used to map and understand the photophysics that give rise to these differences. These results provide context for the very different photochemistries of these potential environmental pollutants.
Author(s): Olivia Neal, Micah J. Cole, Saige Bradley, Tessa Sangermano, Jason Grothaus, and Patrick M. HareDepartment(s): Chemistry & Biochemistry
Synthesis of two Novel Enediyne Amino Acids
Abstract: Enediynes are best known for their potent and selective antitumor activity. Enediynes undergo Bergman cyclization to generate highly reactive diradicals capable of inducing double-stranded DNA breaks. The closely related arenediynes not only have potential biological application, but are attractive starting materials for dehydrobenzoannulenes. Both arenediynes and dehydrobenzoannulenes are promising candidates for the manufacture of conductive polymers, molecular electronics, and organic batteries. In this study, we report our progress on the synthesis and characterization of two novel arenediyne amino acids, 3,4-diethynylphenylalanine (1) and its corresponding 1,2,3,4-tetrahydroisoquinoline variant (2). Both 1 and 2 are derived from commercially available L-DOPA. The tetrahydroisoquinoline version of L-DOPA was prepared by a classic Pictet-Spengler reaction L-DOPA with formalin under acidic conditions. Each of the amino acids was then protected as its Boc methyl ester and the free phenols converted to triflate esters. The alkynes were installed using a Pd-catalyzed Sonogashira reaction. These enediyne amino acids will serve as building blocks for the construction of both aromatic and antiaromatic dehydrobenzoannulenxe amino acids, which can then be used to synthesize novel peptide-based polymers.
Author(s): Devin Flaugher, Dane Larson, Tyler Warner, Lilly Baumann, Jewels Flege, Sam Thompson, and KC RussellDepartment(s): Chemistry & Biochemistry
Characterizing the Impact of Affinity Tag Location on the Function of Chorismate Synthase
Abstract: Chorismate synthase (AroC), the last enzyme in the shikimate pathway, is responsible for converting 5-enolpyruvyl shikimate-3-phosphate (EPSP) to chorismate. Bifunctional chorismate synthase catalyzes both the conversion of EPSP to chorismate and possesses an additional catalytic activity that uses NADPH for the reduction of oxidized FMN to FMNH2. Previous attempts to crystallize the AroC protein from Neurospora crassa resulted in small crystals from a heterogenous mixture of protein assembly states. Further characterization demonstrated that the protein assembled primarily as a dimer. A modified construct was created using site directed mutagenesis to relocate the His6x affinity tag from the N- to the C-terminus. The construct with a C-terminal tag assembled predominantly as a tetramer. Further, an enzymatic assay to test the function of chorismate synthase demonstrated that only the tetramer was functional while the dimer was not. Future studies to characterize FMN binding in the bifunctional enzyme will utilize the tetrameric construct.
Author(s): Jamie Richey, Cate Reis, Jessica Forbeck, and Catherine SheltonDepartment(s): Chemistry & Biochemistry
Computed-Structure Models of Enzymes Involved in Chorismate Synthesis and Utilization
Abstract: AlphaFold3 predicts protein structures from an amino acid sequence. It is used to generate computed structure models of unknown structures, and to generate improved structure models over those solved from moderate or poor experimental data. In this work, the AlphaFold3 server interface was used for both applications. An experimental structure of isochorismate synthase (MenF) from Yersinia pestis had insufficient data to resolve all the secondary structure elements. AlphaFold3 was used to predict a more complete structure. An additional isochorismate synthase, PchA from Pseudomonas aeruginosa, which has not been solved experimentally, was predicted by AlphaFold3 to have a structure quite similar to MenF. The computed structure model for chorismate synthase (CS) from Neurospora crassa was similar to the experimental structure of chorismate synthase from Saccharomyces cerevisiae. However, in both the experimental model and the computed structure model, the orientation of important ligand binding residues could not be located with certainty.
Author(s): Joseph Land, Gianna Galicia, and Catherine SheltonDepartment(s): CINSAM
Recruitment to STEM Disciplines and higher Education of Participants in CINSAM Camps and Academies
Abstract: This project investigates how participation in CINSAM’s STEM camps and high school academies relates to students’ choices of universities and fields of study, particularly recruitment into Northern Kentucky University and STEM disciplines. Data spanning eighteen years of camps was collected from CINSAM’s office. Python scripts were used to clean and aggregate this data. A detailed report was obtained from NKU’s Institutional Research (IR) office, which drew on data from the National Student Clearinghouse. This report contained data on institutions attended, programs enrolled in, and degrees awarded. KNIME, a low code data analytics platform, was used to join and analyze the data sets to identify trends in university attendance and academic disciplines. With our findings, we aim to assess the current impact of CINSAM’s early STEM engagement programs.
Author(s): Chris King, Jordan McGlade, Charlisa Daniels, Madhura KulkarniDepartment(s): Educational Leadership & Advanced Studies
Inspiring Futures: Developing Workforce Skills and STEM Career Exploration Curriculum for Teens
Abstract: This summer research project, under the mentorship of Dr. Kimberly Clayton-Code at the College of Education, focused on the development of a dynamic, interactive online curriculum aimed at preparing high school students for future STEM careers and workforce readiness. The project was conducted in hybrid mode, combining virtual research with select in-person collaborations and interviews.
Our primary goal was to design curriculum content that fosters essential 21st-century skills—critical thinking, communication, collaboration, adaptability, and problem-solving—while promoting STEM career exploration through authentic storytelling. A major focus was on amplifying the voices and experiences of women in STEM, especially those from underrepresented backgrounds, to break down stereotypes and inspire the next generation of learners.
Key responsibilities included conducting a comprehensive literature review on current STEM workforce trends, identifying and interviewing women professionals in STEM fields, and transforming their career journeys into accessible, inspiring profiles. These profiles are now featured on the Young Women Lead website, where they serve as powerful role models for students nationwide. Additionally, I contributed to designing interactive curriculum components aligned with national educational standards, using platforms like Canva to create visually engaging resources.
This project not only advanced my research and curriculum development skills but also deepened my passion for STEM advocacy and educational equity. The final curriculum and digital stories will be implemented in high school clubs across the country in the upcoming academic year, making a meaningful impact on how students envision and prepare for STEM careers.
Author(s): Sudikshya Rajbanshi and Professor Kimberly Clayton-Code
Department(s): Mathematics & Statistics
The US Navy’s RATTLER codebreaking machine
Abstract: On 1 December 1942, US Navy codebreakers and British codebreakers began intercepting Imperial Japanese Navy messages that were enciphered on a new machine, which the Allies called JADE. The intercepts consisted of high command logistic messages between major shore stations. JADE enciphered messages by cascading katakana plaintext through five telephone stepping switches and a plugboard. By late 1943, Navy codebreakers requested a machine to determine the settings of the five switches given an intercept and a little bit of known plaintext and the corresponding ciphertext. The machine, called RATTLER, was designed by engineers at the Naval Computing Machine Laboratory, located at the National Cash Register in Dayton, Ohio. RATTLER was delivered and began operation on 8 May 1944. This project is an exploration of RATTLER, including its components and its operation.
Author(s): Luke Clay and Dr. Chris ChristensenDepartment(s): Mathematics & Statistics
Finding Order in Chaos: The Construction and Properties of the Graph R(5)
Abstract: The Ramsey number R(5) has been narrowed to between 43 and 46 inclusively. The existence of a Ramsey graph of order 45 is conjectured to be found among all potentially self-complementary degree sequences. These have been computed, an algorithm to search for this graph has been extended and possibilities for R(5) have been explored. It is conjectured that the Theorem on Friends and Strangers for n=5 is 46.
Author(s): Blake Settle, Clove NataleDepartment(s): Physics, Geology, & Engineering Technology
Chasing Stars: Commissioning the NKU Spectrograph
Abstract: Stars are massive celestial bodies that emit light that can be captured and analyzed to determine a star’s chemical makeup, temperature, surface gravity (density), and motion. One of the principal tools for characterizing a star is a spectrograph. Once light from a star is captured by the spectrograph, the light is split into its spectral wavelengths (spectrum). By measuring the widths, depths, and locations of absorption features (lines) in this spectrum, one can characterize the star.
This project’s objective is to commission the NKU spectrograph for this purpose. Setting up, calibrating, and aligning the Alpy-600 optical spectrograph, as well as the Atik 314L+, and Atik 460EX CCD cameras is the project's focus. Once properly calibrated, the equipment was attached to the 14-inch telescope located at NKU’s Observatory on top of Founder’s Hall. Then commissioning spectra were obtained and the autoguiding feature tested by locating and mapping easily identified bright stars.
Author(s): Adisyn Donk, Elyse Kuzma, Dr. De LeeDepartment(s): Physics, Geology, & Engineering Technology
Design of a 300 lbf N2O/IPA Liquid Rocket Engine (LRE)
Abstract: The preliminary design of a small-scale bi-propellant liquid-fueled rocket engine (LRE) utilizing nitrous oxide (N2O) and isopropyl alcohol (CH3CHOHCH3), as well as the preliminary design of the feed system from which propellants are delivered to the engine, is described. This research intends to provide a further basis for the use of nitrous oxide as an alternative to liquid oxygen as an oxidizer for amateur liquid rocket engine projects, which provide much-needed experience for students who are to enter the space launch industry. In order to complete the design, a mixture of personal research, past-built N2O engines described in literature, and expert advice was utilized. Moving forward, the engine and accompanying test stand must be manufactured using various machines, such as CNC lathes, CNC mills, grinders, drill presses, and tube benders. Pending manufacturing, a series of tests shall commence, including cold flow tests and engine hotfire tests.
Author(s): Jordan Niemann, Dr. Dirk GrupeDepartment(s): Physics, Geology, & Engineering Technology
The Literature of AGN And Theory Around Variability
Abstract: This study focused on investigating the theories and literature behind AGN variability. Multiple contributors had theories that developed AGN theory.
Author(s): Skie PleshingerDepartment(s): Physics, Geology, & Engineering Technology
Ultraviolet Flux Variability in Active Galactic Nuclei
Abstract: This study investigates ultraviolet (UV) flux variability in Active Galactic Nuclei (AGN) to better understand its behavior and underlying physical mechanisms. UV data was obtained from the Neil Gehrels Swift Observatory’s UVOT instrument and analyzed using HEASoft software. Observations were recorded across six UV filters and plotted by UVOT magnitude (Vega system) versus Modified Julian Date (MJD). The analysis revealed clear variability patterns, with periodic flux increases and decreases that suggest instabilities within the accretion disk play a key role. Future work will incorporate X-ray observations from the same sources to identify patterns and correlations, deepening our understanding of AGN variability.
Author(s): Cameron Easton, Dr. Dirk GrupeDepartment(s): Physics, Geology, & Engineering Technology
X-Ray Analysis of Active Galactic Nuclei
Abstract: Active Galactic Nuclei (AGN) are high luminosity objects at the center of galaxies. A type of supermassive black hole (SMBH), these objects emit significant high energy electromagnetic radiation. In this project we will observe how this high energy radiation in the X-ray spectra.
Author(s): Catherine HeileDepartment(s): Physics, Geology, & Engineering Technology
A Practical, Low-cost, Internet of Things (IoTs) - based Intelligent Air Quality and Environmental Monitoring System
Abstract: Over the past two decades, the expansion of large-scale Animal Feeding Operations (AFOs) in the United States has raised significant environmental concerns due to the emissions of harmful gases, greenhouse gases, and particulate matters. In response to the need for affordable, real-time air quality monitoring, this project serves as an initial step toward developing a user-friendly IoT (Internet of Things) sensing device. This phase focuses on building a compact unit capable of measuring and displaying indoor atmospheric conditions in livestock environments. The device integrates a low-cost Wi-Fi-enabled System-on-Chip (SOC), a BME280 sensor for temperature, humidity, and pressure, and an SSD1306 OLED display. Developed using Visual Studio Code, the system continuously reads environmental data and displays results in real time. This successful implementation demonstrates the feasibility of using accessible, low-cost components for environmental sensing and lays the foundation for a field-deployable system to monitor emissions without reliance on expensive research-grade equipment.
Author(s): Michael Brooks and Dr. Gang Sun
Department(s): Physics, Geology, & Engineering Technology
Development of a 3D-Printed Quadruped Robot Dog for Education and Research
Abstract: This project focuses on the construction and adaptation of a fully 3D-printed quadruped robot dog, based on the open-source Spot Micro design, to be utilized for educational and research purposes. The robot will integrate components of a mechatronic system, including sensors, actuators, a control system, and a user interface, offering students hands-on experience in robotics design, programming, and system integration.
The project will involve two main phases: design and simulation, followed by development and optimization. The initial phase includes equipment procurement, CAD-based mechanical design, and simulation of motion algorithms to ensure stability and functionality. The second phase focuses on assembly, integrating advanced features such as sensor feedback for obstacle avoidance and improved gait and balance, alongside the development of user-friendly instructional materials.
The project’s outcomes include a functional quadruped robot dog, enhanced student proficiency in robotics, and a foundation for future research and development in advanced mechatronic systems.
Author(s): Abby Russell, Prashis Poudel, Vu Tran, and Mahdi YazdanpourDepartment(s): Psychological Science
Modeling the Impact of Pediatric Antipsychotic Use on Later Opioid Reinforcement
Abstract: Many children are prescribed antipsychotic drugs, yet the effects of these medications on brain development remain unclear. A substantial number of children receive antipsychotics for attention-deficit/hyperactivity disorder (ADHD), a condition associated with an elevated risk for developing substance use disorders. Using a rat model, we investigated whether early-life administration of antipsychotics alters conditioned responses to oxycodone—a widely prescribed opioid with high abuse liability—in young adulthood. This study focused on risperidone, a commonly used antipsychotic in pediatric populations, and its effects on oxycodone-conditioned place preference (CPP). Female and male Long-Evans rats received subcutaneous injections of risperidone (0.75 mg/kg) or vehicle twice daily from postnatal day (PND) 14 to 28. CPP testing began at PND 73. It involved a place preference pretest, 12 days of place conditioning with saline or one of two doses of oxycodone (0.2 or 0.6 mg/kg), and a place preference posttest. Data analysis is currently underway, and findings will be presented in our poster presentation. These experiments could provide critical insights into how early-life exposure to antipsychotics may influence later sensitivity to drugs of abuse.
This work was supported by Summer Research Fellowships from the Welti Fund to ARC and KS, and by the National Institute of Health (grant numbers P20GM103436, R16GM149498).
Author(s): Ana L. Rey Caldera, Kaitlyn. J. Strange, & Tyler DownnenDepartment(s): Psychological Science
Prenatal Fentanyl Exposure Alters White Matter Markers in the Prefrontal Cortex of Male Mice
Abstract: As the opioid epidemic rages, the amount pregnant women with opioid-use disorder continues to climb. Fentanyl is a synthetic opioid traditionally utilized for pain relief due to its rapid analgesic effects. Though potency is what makes fentanyl a powerful prescription drug, its low cost of production has encouraged increases in illicit use. Imaging studies done on human brains suggests that prenatal opioid exposure may be associated with lessening integrity and magnitude of the brain’s white matter, but there is a need to conduct more controlled studies that examine fentanyl specifically. Using brains from male mouse offspring exposed prenatally to fentanyl, we used immunohistochemistry to tag myelin, oligodendrocyte cell bodies, and a cholesterol transporter (ABCA1)—all components of white matter. We discovered that prenatal fentanyl exposure increased the overlap of myelin fibers with both oligodendrocytes and ABCA1. This may indicate issues with myelin transport, warranting the need for continued research.
Author(s): Maria Taylor, Martin Powers, Grace Bochenek, Anna G. Makela, and Brittany L SmithDepartment(s): School of Computing & Analytics
A Lightweight Machine-Learning Pipeline for Markerless Remote Eye- and Gaze-Tracking
Abstract: Accurate eye- and gaze-tracking is fundamental for human–computer interaction, accessibility, and cognitive research, yet existing solutions often rely on specialized hardware or heavyweight deep networks unsuited to low-power devices. We present an end-to-end machine-learning pipeline that delivers real-time remote gaze estimation on commodity laptop and tablet cameras.
Author(s): Tyler Egloff, Elaina Hall, Sophia Kapszukiewicz, Mackenzie Glaser, and Nicholas CaporussoDepartment(s): School of Computing & Analytics
Brand Name Resolution for Bidding Behavior Analysis
Abstract: Data sourced from live online auction platforms often suffers from inconsistencies in brand name entries, obstructing accurate data analysis and product filtering. This research addresses the challenge of brand name resolution in a dataset of over 400,000 auction listings by developing a semi-automated pipeline for canonicalizing and validating brand names. The pipeline integrates rule-based normalization, string similarity metrics (e.g., Jaro-Winkler), and targeted human review to map diverse surface brand names (e.g., “Cocacola,” “Coca Cola,” “COCA COLA INC.”) to a unified canonical form. The resulting dataset includes 97,000 canonicalized and 60,000 automatically-validated brand names, offering a robust benchmark resource for researchers and developers. This work demonstrates a scalable approach to entity resolution that improves the reliability of brand-related analysis, particularly for studying bidding behavior and product trends in real-world auction environments.
Author(s): Sarita Gautam, Stephen Kroeger, Ayush Bhandari, Alina Campan, and Nicholas CaporussoDepartment(s): School of Computing & Analytics
Evaluating Output Novelty in Iterative Prompting in Educational Content Generation
Abstract: Large Language Models (LLMs) have been explored as a valuable tool for creating course outlines. To this end, iterative prompting can be utilized to correct and expand the output. However, iterative prompting often results in LLMs producing redundant content that requires extensive manual review. In this context, distinguishing between genuinely novel content and rephrased existing information creates a trade-off between LLM capabilities and validation efforts. To address this, this paper proposes a novelty metric that combines lexical similarity measures with semantic analysis to automatically classify generated content as novel or repeated, thus simplifying human review. We evaluated this approach using five widely used LLMs (ChatGPT 4o, Claude 3.7 Sonnet, Gemini 2.5 Flash, DeepSeek v3, and DeepSeek r1) to create a Java programming course outline over multiple iterations. Our methodology automated the classification of 90.52% of the generated content, significantly reducing the number of items requiring manual review. Moreover, our findings show that the proposed metric can also be utilized for ranking the performance of models in the context of iterative prompting.
Author(s): Vickey Ghimire, Jaljala Shreshta Llama, Bijay DhunganaDepartment(s): School of Computing & Analytics
Preserving Requirements-Driven Development in the Age of AI: an Empirical Evaluation of the Impact of Software Specifications in LLM-Assisted Coding
Abstract: This paper presents an empirical study evaluating the impact of requirements-driven development in Large Language Model (LLM)-assisted software engineering. As individuals, particularly among non-expert users, increasingly use LLMs to develop software, concerns arise regarding code quality, explainability, and alignment with established software engineering (SE) standards. We assess the performance of seven popular LLMs across coding tasks of varying complexity, comparing default prompt-based generation with a structured, requirements-first approach. Using ISO/IEC 25010 quality metrics, our findings demonstrate that incorporating explicit software requirements before code generation significantly enhances the quality of output across both functional and non-functional dimensions. Models guided by formal specifications consistently produced more robust, maintainable, and explainable artifacts.
Author(s): Jeffrey Perdue, Connor Helms, and Nicholas CaporussoDepartment(s): School of Computing & Analytics
Accelerating Access: A Practical Approach to Edge Computing in AWS
Abstract: This research explores the use of edge computing in AWS to improve data retrieval efficiency. Using the AWS CLI, an EC2 instance was configured as an edge node running an Apache server referencing a static website hosted in an S3 bucket. AWS Lambda was also integrated to serve a dynamic database component. A Python program was developed to simulate 100 GET requests to both the EC2-hosted edge application and the S3 bucket directly. Results showed that routing requests through the EC2 instance significantly improved performance when accessing both static and dynamic resources. This demonstrates the potential of edge computing in reducing latency and increasing responsiveness for hybrid content delivery architectures in cloud environments.
Author(s): Terran Meek Department(s): School of Computing & Analytics
Enhancing Performance and Security through Smart Edge Computing
Abstract: Cloud and edge computing are key technologies for delivering fast and secure web applications. This project explores how edge computing using AWS CloudFront can improve performance compared to traditional cloud delivery through Amazon S3. A simple static website built with HTML and JavaScript was hosted on S3 and distributed via CloudFront, and its performance was then evaluated using WebPageTest.org from multiple global locations to assess latency and loading times. To enhance security, AWS WAF features were implemented including CAPTCHA, geo-blocking and a challenge-response rule. These configurations were tested using a VPN to simulate different regions and verify WAF behavior. Overall, the results show that edge computing improves both speed and security by lowering user-perceived latency and intercepting potentially harmful traffic before it reaches the application.
Author(s): Wei Hao, Jenni Tonche SotoDepartment(s): School of Computing & Analytics
LLMs for Business Planning: Evaluating Novelty and Relevance in Iterative Prompting
Abstract: Large Language Models (LLMs) are increasingly employed for structured content generation across diverse domains, including business strategy. When generating content, iterative prompting can refine and expand LLM-generated outputs to fit specific goals better. However, repeated iterations often yield redundant or rephrased responses, requiring significant manual effort to identify genuinely novel and contextually relevant output, which is particularly important for creating material such as marketing plans, where outputs must be tailored to a specific business. To address this challenge, this paper proposes a framework that can assess both the novelty of each iteration and the relevance of generated plans to a given business profile. We evaluate this approach across multiple LLMs and business scenarios, enabling automatic classification of generated content and reducing the need for manual review. In addition, the framework provides a basis for benchmarking the performance of generative models in iterative marketing plan development.
Author(s): Linh Nguyen, Prajita Tandukar, Vickey Ghimire, Nazmus Sadat, Nicholas CaporussoDepartment(s): School of Computing & Analytics
Investigate Application of AI in Various Games: DFS in Sokoban game
Abstract: This project investigates the application of artificial intelligence (AI) in game development through the design and implementation of a Sokoban puzzle solver. We develop a modular game environment using Python and Pygame, then integrate AI algorithms—search-based approaches, such as breadth first search and depth first search—to autonomously solve puzzles. By testing different hyperparameters, we evaluate their performance in terms of efficiency, adaptability, and scalability within the Sokoban domain. The resulting framework serves as a testbed for AI experimentation, demonstrating how machine learning and traditional pathfinding techniques can enhance gameplay automation. Our findings provide insights into the challenges of AI-driven puzzle solving and lay the groundwork for future research in adaptive game AI.
Author(s): Arthur Zapata, Dr. Yangyang Tao, and Dr. Junxiu ZhouDepartment(s): School of Computing & Analytics
Smart agent for Chrome Dino Game
Abstract: This research focuses on training an AI agent to play the Chrome Dino game using Convolutional Neural Networks (CNNs). The project applies deep learning techniques to enable the model to interpret game visuals and respond to obstacles in real time. By learning from game frames, the CNN recognizes patterns and improves the agent’s ability to jump or duck effectively, aiming for higher scores. The study examines how well the model performs in terms of accuracy, adaptability, and decision-making during gameplay. Overall, this work explores the potential of CNNs in automating visual-based games and making fast, efficient decisions in constantly changing environments.
Author(s): Ashok Gaire, Yangyang TaoDepartment(s): School of Computing & Analytics
Analyzing Urban Heat Islands in Columbus, Ohio Using ArcGIS Pro
Abstract: The Urban Heat Island (UHI) effect results from impervious surfaces that trap and reemit solar energy. This study maps UHI in Columbus, Ohio using ArcGIS Pro to support future urban planning projects. Kriging interpolation was applied to point temperature data collected by bike traversals to model surface temperature in the study area; UHI was calculated from the surface temperature model using the Raster Calculator. The UHI map shows densely built areas in downtown Columbus experience a stronger UHI effect than less urbanized areas with ample green space. The Zonal Statistics tool was used to compute the UHI in each land cover type. The result shows higher UHI effect from buildings correlates to their reemission of heat in the morning; better cooling effect from green spaces relates to their lower UHI result. This study demonstrates an approach to mapping UHI and highlights how land use and vegetation influence urban thermal patterns.
Author(s): Cecelia Harner, Leticia Muñoz, Chris BowmanDepartment(s): University of Nebraska Medical Center
Bovine Colostrum Extracellular Vesicles as Parkinson Disease Therapeutics
Abstract: Parkinson's disease (PD) is caused by inflammatory microglial processes that contribute to dopaminergic neuron death and alpha-synuclein production. Bovine colostrum-derived extracellular vesicles (EVs) reduce pro-inflammatory factors in infected macrophages and cross the blood-brain barrier. We hypothesized that EV treatment prior to MPTP exposure could modulate microglial activation and prevent dopaminergic neuron death. EVs were isolated from donor mice using ultracentrifugation and validated through nanoparticle tracking and surface marker analysis. IVIS imaging confirmed EV brain trafficking within 24 hours of injection. Mice were treated with PBS, MPTP only, or MPTP+EVs. Flow cytometry revealed increased regulatory T cell populations in blood and spleen of MPTP+EV groups at days 2 and 7. Transcriptomic analysis showed downregulated inflammatory pathways and upregulated immune regulatory pathways in the MPTP+EV group. Immunohistochemistry demonstrated significant reductions in microglial activation but no significant increase in dopaminergic neuron survival. These findings suggest that EVs effectively modulate neuroinflammation in PD models.
