Digital Commons Network^™

Galora: A Lightweight Graph-Aware Llm Framework For Node Classification On Text-Attributed Graphs, Mayur Choudhary

Master's Projects

With the exponential rise of language models (LMs) and their potential to understand semantic relationships, large LMs are being used across a wide range of applications. Text-attributed graphs (TAGs) are one notable example where LLMs can be combined with Graph Neural Networks (GNNs) to enhance node classification results. TAGs associate textual content with each node and are commonly seen in various domains such as social networks, citation graphs, recommendation systems, etc. Effectively modeling TAGs would enable deeper insights into different aspects of the graph and improve decision-making in relevant domains. We present GaLoRA, a parameter-efficient framework to integrate structural information …

Extraction Of A Knowledge Graph Of Biomedical Relationships, Brian Tran

Master's Projects

Rapid release in biomedical literature poses a challenge in linking information. This thesis aims to extract data from expanding datasets to identify and form meaningful relationships between biomedical entities. Large language models (LLMs) enable us to learn at a rapid pace. Creation of LLms from scratch are impractical. This thesis aims to collect a small dataset, containing biomedical papers, and use it to train large language models (LLMs) to extract entities from the text and learn the relationships between these entities. The experiment will be divided into two stages and utilize EU-ADR and ChemProt dataset. Starting with named entity recognition …

Performance Comparison Of Machine Learning Across Metal, Cuda, And Neuromorphic Frameworks, Ryan Saini

Master's Projects

Machine learning’s computational demands necessitate optimal performance and utilization. This research compares Apple Silicon M3 Pro with MPS, NVIDIA RTX 3070 GPU with CUDA, and neuromorphic computing for machine learning methods. We provide a cross-platform and cross-architecture performance analysis of machine learning methods to identify optimal configurations for training and inference scenarios. On traditional neural networks, Apple Silicon with MPS delivers superior energy efficiency at the cost of longer processing times for training and inference. NVIDIA with CUDA offers faster computation in training and inference at higher energy costs. Convolutional spiking neural networks perform competitively on event-based data, particularly on …

Malware Opcode Embedding And Quality Assessment Of Generative Sample Embeddings, Atishay Jain

Master's Projects

Malware is software used to damage and disrupt computer systems with the intent to cause damage to the victim. Malware detection and classification into malware families is a crucial problem for cybersecurity researchers. One of the major bottlenecks in improving these systems is the shortage of good quality labeled malware data, especially for malware families with scarce samples. Researchers have utilized generative models to generate malware data to address this issue. Malware embeddings encode patterns within a malware file, which can be used to detect and classify malware. Recently, encouraging results have been obtained in generating malware embeddings using generative …

Clustering Organ Cell Types, Venkata Satya Swathi Mattaparthi

Master's Projects

The Human Cell Atlas (HCA) created a reference map of all human cells. My project uses the Tabula Sapiens dataset, developed under HCA and based on single-cell RNA sequencing data, to explore cell type and tissue diversity. I performed experiments using the Elbow method and a formula based on dataset observations to determine the number of clusters, then applied k-means clustering on two representative subsets of the All Cells dataset. Clusters were selected for analysis using Shannon’s Diversity Index and Pielou’s Evenness. A novel algorithm based on the cell differentiation tree was used to validate the biological coherence of the …

Framework For Identity Privacy Through Gender Based Skeletonization, Harrison Hwang

Master's Projects

The protection of one’s privacy and sensitive information is becoming increasingly difficult in the modern age full of surveillance and data collection. Through the use of image based object detection machine learning models trained for human and facial recognition, people can be identified and tracked to a terrifyingly accurate degree. On the other hand, the information present in surveillance media can play a key role in security and law enforcement. This presents a problem of how to preserve key information without compromising the privacy of any individuals present in the video. In this research project, Computer Vision techniques and a …

Large Language Models For Bacterial Genomic Analysis, Manvendra Chavan

Master's Projects

Identification of bacterial gene sequences with agricultural applications has the potential to transform agricultural biotechnology. These genes can be used in environmentally friendly pest control strategies. One such use case is identifying genes with potential insecticidal properties. With an increasing number of genomic information and decreasing numbers of available annotated sequences, finding new insecticidal genes has become more challenging.The traditional methods relying on sequence alignment and annotated databases are not effective in detecting functionally relevant genes lacking close homology to known cases. This project investigates the data-driven classification of genes by sequence modeling. This research is focused on learning DNA …

Advanced Knowledge Extraction With Biomedical Data Using Llms, Akshat Krishna

Master's Projects

The rapid growth of biomedical research has led to an overwhelming volume of unstructured textual data in the scientific literature. This has necessitated the development of an automated approach for knowledge extraction and integration. In

this project, we present a comprehensive pipeline for constructing a unified biomed- ical knowledge graph by combining two well-known datasets: CHEMPROT [1],

which captures chemical–protein interactions, and EU-ADR [2], which annotates drug–gene–disease relationships. In order to identify important biomedical entities and interactions from CHEMPROT dataset, we perform Named Entity Recognition (NER) and relation Extraction (RE) using state-of-the-art biomedical models like BioBERT [3], BioGPT [4] and …

Temporal Patterns Of The Introduced Sponge Hymeniacidon Perlevis (Montagu, 1814) In The Elkhorn Slough, California, Usa, Jackson T. Hoeke, Kerstin Wasson, Amanda S. Kahn

Faculty Research, Scholarly, and Creative Activity

Hymeniacidon perlevis is a cosmopolitan sponge with a seasonal life cycle. We investigated seasonal and interannual dynamics of H. perlevis in Elkhorn Slough estuary, where it is an introduced species, and explored correlations between sponge cover and environmental conditions. We used sponge cover to estimate the potential effects of H. perlevis on its environment, and how those could vary across its seasonal life cycle. We found that recruitment is currently restricted to the upper estuary and while it varies annually, the frequency and density of sponge recruits have generally increased from 2007 to 2023. A seasonal life cycle was confirmed …

Coral Vision – Crustose Coralline Algae Detection With Computer Vision, Ryan Tseng

Master's Projects

Crustose coralline algae (CCA) are a group of red algae that are vital contributors to the health of coral reef ecosystems. Monitoring CCA abundance can serve as an indicator for coral reef health and improve reef conservation efforts. Autonomous Reef Monitoring Structures (ARMS) are artificial structures that can be deployed into coral reef ecosystems and retrieved to gather ecological data without harming reef structures. Traditional methods of calculating CCA abundance require manual analysis and are labor-intensive. Recent developments in computer vision and deep learning technology have provided an avenue to fully automate this task. This research aims to train a …

Characterizing Somatic Variants In Nanopore Data With Machine Learning, Shwethal Sayeeram Trikannad

Master's Projects

Oxford Nanopore Technology (ONT) is a popular long-read sequencer in genomics. However, its high base-calling error rate produces several sequencing artifacts. Detection of somatic variants in ONT sequenced tumor-normal samples remains challenging due to low frequencies. In this study, machine learning was applied to a dataset created by benchmarking ClairS output against HCC1395 and colo829 truth sets to classify variants and artifacts. Relevant features were engineered from sequence context and variant site characteristics to model artifact profiles. HistGradientBoostingClassifier achieved 0.876950 accuracy, outperforming all other models. Variant quality was the top predictor with an aggregate accuracy of over 85%. This work …

Automated Sponge Segmentation With Mask R-Cnn On Arms Plate Images For Reef Monitoring, Mrunali Abhijit Thokadiwala

Master's Projects

Climate change-driven ocean warming and acidification are disrupting the ecological balance of coral reefs. Notably, these oceanic conditions are undermining coral health and accelerating their decline which is favoring some sponge species in outcompeting them for dominance. Although functional, these altered reefs destabilize the reef architecture, hinder nutrient cycling, and support fewer marine species. Thus, monitoring the growth and abundance of various sponges and understanding their roles at different stages of ecological succession in coral reefs is vital. Autonomous reef monitoring structures (ARMS) are often used for this purpose, but manual taxonomic analysis using their images is time-consuming, inconsistent and …

Evorgcn: Harnessing Esm-2 Evolutionary Embeddings With Relational Gcns For High-Fidelity Protein-Protein Interaction Prediction, Mohit Kunder

Master's Projects

Accurately predicting protein-protein interactions (PPIs) is essential for understanding cellular function and advancing biomedical discovery. We model PPIs as graphs, where nodes represent proteins and edges denote interactions. Using interaction data from the STRING database, we use two samples of it, namely the benchmark datasets—SH27K and SH148K—filtered by confidence score and annotated by interaction mode (multiple relations). In this project, we present EvoRGCN, a graph-based machine learning framework for PPI prediction that integrates both sequence-level (ESM-2 embeddings) and network-level information. We incorporate various Graph Neural Network architectures, including Graph Convolutional Networks, Graph Attention Networks, and Relational Graph Convolutional Networks. Our …

Stream Response To A Dam Removal: Mill Creek, Davenport, Ca, Jessica J. Williamson

Master's Theses

Dams have many effects on stream morphology and ecology, including a loss of sediment supply downstream, reduced access to the entire stream for aquatic fauna, and a flooded area upstream of the dam. Dams are a hurdle for migrating organisms such as anadromous fish, which is a group of concern. An over 100-year-old dam on Mill Creek, a stream draining a coastal watershed in central California, was removed in the fall of 2021. The site is in the Santa Cruz Mountains, where endangered salmonids such as steelhead, Oncorhynchus mykiss, have historically lived. To study the change in the stream from …

Separation And Surface Examination Of Spacecraft Cabin Particulates, Grace A. Belancik

Master's Theses

Particulate matter in the atmosphere is a known detriment to human health, but several factors affect any particulate’s particular toxicity including particle size, surface area, and chemical composition. Extensive studies have been conducted to examine particulate hazards on Earth, but in crewed spacecraft, like the International Space Station (ISS), the particulate environment is wholly unique due to both the controlled environment and lack of gravity. Atmospheric sampling studies are underway, but another convenient source of airborne particulates for analysis is the contents of the vacuum bags used by the astronauts to clean the air vents. However, no method currently exists …

Science, Physiology, And Nutrition For The Nonscientist, Judi S. Morrill

Open Educational Resources

A wonderful blend of physiology, nutrition, biochemistry, genetics, biology, evolution, chemistry--what we all need to know as informed citizens. A basic knowledge of the life sciences and how our bodies work--to promote our own good health, especially as we're bombarded with misleading advertisements, soundbites, and the like. DNA fingerprinting, calorie requirements, dietary advice, genetic engineering (including gene editing with CRISPR cas9)--all in an easy-to understand book.

Realities Of Nutrition, Judi S. Morrill

Open Educational Resources

What Is Nutrition Reality? Is it in the advertising claims that a food is natural or supplies instant energy? Is it in the myriad of dietary supplements? Is it in the diet plans for instant weight loss and glowing health? The only way to find out is to learn the basic principles of nutrition science, so you can be your own nutrition expert. Here is the true introduction to nutrition that you will read with pleasure and real understanding. It will free you from dependence on popular sources of information - often misinformation - so you can distinguish illusion from …

Evaluation Of A Community-Based Socialization Program For Independently Living Older Adults, Patricia A. Shane, Anne T. Ferguson, Erin L. Woodhead

Faculty Research, Scholarly, and Creative Activity

Social isolation among older adults is a major public health concern because of the adverse impact it has on health and well-being. Reducing social isolation can improve health and decrease the use of costly healthcare services. However, there is little consensus on the critical ingredients of programming to reduce social isolation. Bay Area Older Adults® (BAO) is a nonprofit organization whose community-based Healthy Living Program (HLP) focuses on building and nourishing older adults’ social networks through transitions such as retirement, divorce and loss of loved ones. The current study was a program evaluation of the HLP using a survey to …

Exploration Of The Effects Of Tai Chi Practice On Lower Limb Corticomuscular Coherence During Balance-Demanding Virtual Reality Conditions In Older Adults, Yang Hu, Elizabeth T. Hsiao-Wecksler, Manuel E. Hernandez

Faculty Research, Scholarly, and Creative Activity

Tai Chi practice has been widely adopted to improve balance and prevent falls in older adults. However, the neural mechanisms underlying the benefits of Tai Chi are difficult to evaluate during traditional balance assessments. The goal of this study was to evaluate the effects of Tai Chi and healthy aging on corticomuscular coherence (CMC) while standing in virtual balance-demanding environments. We recorded neural, muscular, and behavioral data in an immersive virtual reality environment while implementing sensory and mechanical perturbations to introduce high postural threats. Through the acquisition of electroencephalography and electromyography signals, we examined β and γ CMC changes in …

Physiological Effects Of Research Handling On The Northern Elephant Seal (Mirounga Angustirostris), Lauren A. Cooley, Allyson G. Hindle, Cassondra L. Williams, Paul J. Ponganis, Shawn M. Hannah, Holger Klinck, Markus Horning, Daniel P. Costa, Rachel R. Holser, Daniel E. Crocker, Birgitte I. Mcdonald

Faculty Research, Scholarly, and Creative Activity

Wildlife researchers must balance the need to safely capture and handle their study animals to sample tissues, collect morphological measurements, and attach dataloggers while ensuring their results are not confounded by stress artifacts caused by handling. To determine the physiological effects of research activities including chemical immobilization, transport, instrumentation with biologgers, and overnight holding on a model marine mammal species, we collected hormone, blood chemistry, hematology, and heart rate data from 19 juvenile northern elephant seals (Mirounga angustirostris) throughout a translocation experiment. Across our six sampling timepoints, cortisol and aldosterone data revealed a moderate hormonal stress response to handling accompanied …

Words And Stardom: Modernist Poetic Responses To Asta Nielsen, Karl E. Toepfer

Open Educational Resources

This essay discusses how four modernist poets portrayed the Danish movie star Asta Nilesen (1881-1972). German Hans Schiebelhuth (1895-1944) describes the "demonic" power of Nielsen's image to awaken dark aspects of his adolscent personality. For the Flemish poet Paul van Ostaijen (1896-1928), Nielsen's cinematic image provokes a radical shattering of language into extravagant "word-images." Danish writer Rudolf Broby-Johansen (1900-1987) linked Nielsen's image to a radical (Marxist) critique of oppressive sexual politics within a capitalist society. German Franz Behrens (1895-1977) composed a lengthy, expressionist ode to Nielsen from the perspective of one who actually worked as a screenwriter in the German …

Conceptual Critical Success Factors Model On Infrastructure Sustainability Rating System For California Construction Projects, Joseph J. Kim, Jose A. Arroyo-Turcios

Mineta Transportation Institute

This report addresses the sensitivity and reliability of the sustainability rating systems by comparing each category’s verified scores with its respective submitted scores to evaluate how points are awarded for infrastructure projects, examine which categories present the most challenges for verification, and identify the best category for verification in the sustainability rating system. The authors conducted three analyses using credit score data obtained from fourteen actual civil infrastructure projects certified under Envision. First, the Natural World category had the highest average score from the submitted and verified data. Pairwise comparisons using t-tests indicate that the mean value of one category …

The Structure And Function Of A Coastal State Aquaculture Plan, C. R. Fong, C. M. Gonzales, M. Rennick, H. J. Lahr, L. D. Gardner, B. S. Halpern, H. E. Froehlich

Faculty Research, Scholarly, and Creative Activity

Aquaculture contributes approximately half of all global aquatic food production and will be critical in addressing future seafood demand, including the United States (U.S.). Many states, territories, and countries are developing and updating aquaculture specific policy in an effort to plan for, and better manage the expansion of aquaculture in their territorial waters. The U.S. has considerable potential to sustainably expand aquaculture production, which has been a focus in both federal and state policy. State aquaculture plans have been used for over 40 years to help plan and coordinate aquaculture development, with a growing emphasis on sustainability. Here, we located …

Deep Subseafloor Sediments In Guaymas Basin Harbor Cosmopolitan Microbiota And Traces Of Hydrothermal Populations, Paraskevi Mara, David Beaudoin, Ivano Aiello, Yuki Morono, David Geller-Mcgrath, Virginia P. Edgcomb, Andreas Teske

Faculty Research, Scholarly, and Creative Activity

Environmental factors shape subsurface microbial ecosystems, and well-characterized sites are ideal for determining how environmental parameters shape sediment communities. Sediments from eight geologically and thermally distinct sites were drilled during International Ocean Discovery Program Expedition 385 in Guaymas Basin, an expedition focused on the hydrothermal deep biosphere. Using high-throughput 16S ribosomal nucleic acid sequencing, cell counts, phylogenetics, metatranscriptomics, and mineralogical/elemental X-ray spectroscopy, we examine linkages and feedbacks between mineral composition, temperature, geochemistry, and microbial populations. We show subsurface life is dominated by heterotrophic, cosmopolitan prokaryotes that thrive within a range of sediments and temperature conditions across Guaymas Basin. Hydrothermally-affiliated lineages …

Private Landowner Interest In Prescribed Fire In California: Findings From Workshops In The Sierra Nevada, Kate M. Wilkin, Amanda M. Stasiewicz, Susan D. Kocher

Faculty Research, Scholarly, and Creative Activity

Background: Globally, prescribed fire political interest and practice has been rekindled following recent devastating wildfire seasons. This phenomenon was especially acute in areas with dual wildfire and forest health crises, like California. Previous research has investigated prescribed fire on public lands or on private lands in other regions, but little is known about prescribed fire practice or interest on private lands in California. Therefore, we sought to understand private land managers’ perceptions of prescribed fire compared to other land management techniques, treatment pathways, motivations, and barriers to complete these treatments in California. Results: Before workshops on prescribed fire for private …

A Multiscale Approach To Understanding The Shared Blue-Orange Flower Color Polymorphism In Two Lysimachia Species, Mercedes Sánchez-Cabrera, Eduardo Narbona, Montserrat Arista, Pedro L. Ortiz, Francisco J. Jiménez-López, Amelia Fuller, Benjamin Carter, Justen B. Whittall

Faculty Research, Scholarly, and Creative Activity

Background: Polymorphisms are common in nature, but they are rarely shared among closely related species. Polymorphisms could originate through convergence, ancestral polymorphism, or introgression. Although shared neutral genomic variation across species is commonplace, few examples of shared functional traits exist. The blue-orange petal color polymorphisms in two closely related species, Lysimachia monelli and L. arvensis were investigated with UV-vis reflectance spectra, flavonoid biochemistry, and transcriptome comparisons followed by climate niche analysis. Results: Similar color morphs between species have nearly identical reflectance spectra, flavonoid biochemistry, and ABP gene expression patterns. Transcriptome comparisons reveal two orange-specific genes directly involved in both blue-orange …

Food Insecurity Knowledge And Training Among College Students In Health Majors, Virginia B. Gray, Cara Cuite, Megan Patton-López, Rickelle Richards, Mateja Savoie-Roskos, Stephanie Machado, Emily Heying, Matthew Landry, Susan Chen, Rebecca L. Hagedorn-Hatfield, Georgianna Mann, Zubaida Qamar, Kendra Oonorasak, Victoria A. Zigmont

Faculty Research, Scholarly, and Creative Activity

Objective: To describe current food insecurity (FI)-related training among nutrition/dietetics, public health, and social work students. Methods: A cross-sectional online survey was used among students (n = 306) enrolled in health-related programs at 12 US universities. Participants reported FI-related course-based and extracurricular experiences and rated confidence to address FI on a scale of 1–3. Open-ended questions investigated perceived definitions of FI and impactful course activities. Descriptive statistics and thematic analysis were used for data analysis. Results: Participants’ FI definitions were multifaceted. Most (80.6%) reported FI being covered in at least 1 course. The overall mean confidence to address FI was …

Piloting A Novel Computational Framework For Identifying Prosthesis-Specific Contributions To Gait Deviations, Jacques Ezechiel N'Guessan, Muhammad Hassaan Ahmed, Matthew Leineweber, Sachin Goyal

Faculty Research, Scholarly, and Creative Activity

This paper introduces a novel computational framework for evaluating above-knee prostheses, addressing a major challenge in gait deviation studies: distinguishing between prosthesis-specific and patient-specific contributions to gait deviations. This innovative approach utilizes three separate computational models to quantify the changes in gait dynamics necessary to achieve a set of ideal gait kinematics across different prosthesis designs. The pilot study presented here employs a simple two-dimensional swing-phase model to conceptually demonstrate how the outcomes of this three-model framework can assess the extent to which prosthesis design impacts a user's ability to replicate the dynamics of able-bodied gait. Furthermore, this framework offers …

Beyond Boundaries: Governance Considerations For Climate-Driven Habitat Shifts Of Highly Migratory Marine Species Across Jurisdictions, Bianca S. Santos, Elliott L. Hazen, Heather Welch, Nerea Lezama-Ochoa, Barbara A. Block, Daniel P. Costa, Scott A. Shaffer, Larry B. Crowder

Faculty Research, Scholarly, and Creative Activity

The mobile nature of migratory marine animals across jurisdictional boundaries can challenge the management of biodiversity, particularly under global environmental change. While projections of climate-driven habitat change can reveal whether marine species are predicted to gain or lose habitat in the future, geopolitical boundaries and differing governance regimes may influence animals’ abilities to thrive in new areas. Broad geographic movements and diverse governance approaches elicit the need for strong international collaboration to holistically manage and conserve these shared migratory species. In this study, we use data from the Tagging of Pacific Predators program to demonstrate the feasibility of using climate-driven …

Creatine And Depression In Rodents: A Meta-Analysis, Ruchika S. Tirupachur

Master's Theses

Traditional treatments for major depression are based on the monoamine deficiency hypothesis, and they fail to achieve remission in a substantial portion of patients. Newer evidence suggests the pathophysiology of depression is a more complex story, highlighting a need for research in alternative therapies. Mounting evidence shows a linkage between depression and the supplement creatine monohydrate (CRMH). Research in rodent models suggests that CRMH may both alleviate depressive symptoms and enhance the efficacy of standard antidepressants like SSRIs. The following study describes findings from a combined fixed- and random-effects meta-analysis based on four empirical studies investigating the effect of CRMH …