We are excited to continue the Summer REU Site under the NSF Grant titled “Advancing high-performance computing (HPC) opportunities in undergraduate research at UW-Eau Claire to meet challenges of multidisciplinary computational science” in 2025.

The REU site will support research for 10 undergraduate students each summer engaging them in a collaborative research experience with the central theme of designing computationally intensive algorithms and simulation models using HPC. The research will be conducted using UW-Eau Claire's Blugold High-Performance Computing facility.

Details of the research opportunities are given below.

Biology

Nora Mitchell

Project: Understanding Evolutionary Phenomena Using Bioinformatics

The Mitchell Research Group investigates rapid biological radiations to understand evolutionary drivers that underpin biodiversity. Focusing on the adaptive radiation of the South African plant genus Protea, the group examines how shifts in pollinators have driven novel morphological, chemical, and functional traits. Using phylogenomic tools, they analyze hundreds of genes to unravel evolutionary relationships within Protea and gain insights into the mechanisms of adaptation and diversity. Students in the lab work with biological samples and utilize bioinformatics pipelines to process sequence data from Anchored Phylogenomics. These pipelines include both traditional methods and emerging approaches leveraging machine learning, such as DeepVariant.

Computer Science

Rahul Gomes

Project: Designing Optimized Deep Learning Algorithm for Image Classification

Research in the Gomes Lab focuses on optimizing deep learning algorithms to enable faster image classification and segmentation with fewer training parameters and training labels. Using deep learning, students are developing software tools that have a direct impact on health sciences. They are exploring optimization techniques that could find application in imaging devices like CT scanners capable of rapid disease diagnosis as well as detection of surgical implants.

Jim Seliya

Project: Investigating Artificial Intelligence in Developing Novel Solutions for Modeling and Analysis of the Aphasia Language Disorder.

The Seliya Research Group focuses on investigating and developing effective solutions for various data-centric problems across different domains. The research group proactively collaborates on research projects with undergraduate students and faculty in the computing sciences as well as those in other disciplines. Students work on developing novel solutions for both basic and applied research problems. A common thread among the various projects is the involvement of one or more of the Artificial Intelligence associated areas, including Machine Learning, Deep Learning, and Natural Language Processing (and LLMs). Some select domains wherein students have developed and assessed effective AI-based solutions include healthcare, security, natural language, music analysis, speech fluency disorders, computing pedagogy, fraud detection, performance in sports, biometrics, and natural disasters.

Chemistry and Biochemistry

Sudeep Bhattacharyay

Project: Developing Inhibitor Screening Techniques Using Deep Learning and High-Throughput Docking

The Bhattacharyay Research Group in the Department of Chemistry and Biochemistry is exploring to develop drug molecule screening method that employs deep learning and artificial intelligence-based algorithms in addition to computational quantum chemistry and high throughput docking of small molecules. Computational methods are being used to identify small-molecule inhibitors of Quinone Reductase 2 (QR2), a cytosolic enzyme linked to oxidative stress and upregulated in Alzheimer’s patients. By targeting QR2, the group aims to advance therapeutic strategies for neurodegenerative diseases. Their research leverages Harris Hawk Optimization-based deep learning to efficiently screen potential inhibitors from large cheminformatics databases, classifying molecules based on binding free energy at QR2's active site through high-throughput docking.

Stephen Drucker

Project: Testing and Refinement of Computational Methods for Characterizing Electronically Excited

The Drucker Lab in the Department of Chemistry and Biochemistry explores the science of molecules that interact with electromagnetic radiation. Using quantum chemical calculations as well as the experimental jet-cooled laser spectroscopy technique, the researchers try to understand how the structure and bonding of the molecules under study have changed as a consequence of the laser excitation.

Sanchita Hati

Project: Molecular Dynamics Simulations to Investigate Macromolecular Condensates Formation in the Aqueous Solution of Synthetic Crowders of Variable Sizes and Shapes

The Hati Research Group investigates the formation of biomolecular condensates, membraneless organelles critical for processes like gene expression, DNA damage response, and signal transduction. To uncover the molecular mechanisms driving condensate formation, the group uses synthetic polymers such as PEG (polyethylene glycol) of varying sizes, ranging from 100 Da to 20,000 Da, to study their ability to form condensates and induce biomolecular assembly. This research provides molecular-level insights into condensate formation, aiding the development of protein-based therapeutics and industrial enzymes in intracellular-like environments.

Geography and Anthropology

Papia Rozario

Project: Geospatial Deep Learning Framework in Precision Agriculture

The Rozario Lab in the Department of Geography and Anthropology applies Data-Driven AI to precision agriculture, land-use, and land-cover research. The lab also explores forestry applications, using hyperspectral and multispectral imagery for biomass and carbon mapping. Research integrates UAV-derived high-resolution imagery, digital elevation models (DEMs), and soil sensor data to develop AI pipelines for analyzing land-use changes. Deep learning techniques, including convolutional neural networks (CNNs), are employed to estimate aboveground biomass (AGB) and nitrogen levels. Advanced feature selection methods further enhance predictive accuracy, supporting sustainable agricultural and forestry management.

Materials Sciences and Biomedical Engineering

Matt Jewell

Project: Application of Deep Learning Algorithms for Superconducting Wire Assessment

The Jewell Research Group in the Department of Materials Science and Biomedical Engineering employs deep learning to advance the study of superconducting materials, focusing on the Bi₂Sr₂CaCu2O₈ (Bi-2212) system. Using semantic segmentation models, the group classifies superconducting filaments as either conjoined or individual, a distinction critical to optimizing the material's performance as a magnet conductor in next-generation proton-proton particle accelerators and muon colliders. These improvements could reduce accelerator size and cost while increasing particle interaction frequencies. The group's research explores the relationship between filament bridging, critical current density, and hysteretic losses, essential for practical magnet applications.

Ying Ma

Project: Machine Learning Enabled Design of Amorphous Materials for Electrochemical Energy Storage

The Ma Research Group in the Department of Materials Science and Biomedical Engineering focuses on advancing lithium-ion battery (LIB) technology to address challenges such as limited range, slow charging, and safety concerns in electric vehicles. The group specializes in studying novel amorphous materials for electrodes and electrolytes, such as Li₃OCl, Li₃OBr, and SiO₂​. Their research leverages machine learning force fields (MLFFs) to model electrochemical behaviors at a fraction of the computational cost of traditional ab initio molecular dynamics (AIMD) simulations. Undergraduate researchers (UGRs) in the lab develop MLFFs for chosen materials and study key properties, including amorphous structure stability, lithium-ion diffusivity, and electrolyte-electrode interface interactions. This approach enables deeper insights into materials' potential for enhancing battery performance while reducing computational demands.