Elias DeVoe

I am a Data Scientist and software developer with a background in the biomedical domain. In May of 2024 I completed a Master's degree in Biomedical Data Science at UW-Madison where I worked with Sushmita Roy's group on one project to understand how deep models predict 3D genomic structures using HiC data, and another to discover clusters of genetic variants that allow patients with metastatic breast cancer to have extremely long survival times despite their advanced disease. Before beginning graduate studies, I worked with the Mellowes Center and Research Data Warehouse at the Medical College of Wisconsin, where I developed several applications to streamline translational research.

Roy Group

Graduate Research Assistant

Leveraged explainability techniques to understand deep model predictions of 3D genome structure in HiC datasets.

Identified gene networks associated with long-term survival of metastatic breast cancer.

Several other projects, including deep learning approaches to scRNA-seq and HMM-based ortholog finding

Present

September 2022

Medical College of Wisconsin

Bioinformatics Software Developer

Developed software applications to support clinical research and practice, including:

A suite of tools to explore phenotypic and genetic data, including EMR search powered by deep learning and unification of variants across the proteome to facilitate hypothesis generation.

Tools to streamline identification of at-risk patients by genetic counselors 3. Gather, verify, and analyze the data supporting each application

Leveraged the EMR in a variety of research projects developing the precision medicine center.

Translated between many domains of expertise (data science, genomics, software)

August 2022

December 2018

Theseus LLC

Full Stack React/Django Developer

Developed a data-driven learning application to facilitate learning and knowledge retention in adult learners, and provide a metric on source quality and personal expertise in any category

Built front-end, REST API, and content recommendation engine

Managed tight communication with client including goals, progress, and setbacks

November 2018

June 2016

Roden Biogeochemistry Group

Undergraduate Researcher

Studied interactions between microbial communities and geological systems

Built data organization pipeline, automated BLAST workflow

May 2016

March 2015

Changes in Daily Apparent Diffusion Coefficient on Fully Quantitative Magnetic Resonance Imaging Correlate With Established Genomic Pathways of Radiation Sensitivity and Reveal Novel Biologic Associations

William A. Hall, Angela J. Mathison, Elias DeVoe, Michael Tschannen, Jaime Wendt-Andrae, Michael Straza, Musaddiq Awan, Lindsay L. Puckett, Colleen A.F. Lawton, Christopher Schultz, Raul Urrutia, Sarah Kerns, Javier F. Torres-Roca, X. Allen Li, Beth Erickson, Marja T. Nevalainen, Michael T. Zimmermann, Eric Paulson

International Journal of Radiation Oncology, Biology, Physics

Changes in quantitative magnetic resonance imaging (qMRI) are frequently observed during radiation therapy (RT). We hypothesized that qMRI changes may correlate with DNA damage response (DDR) capacity within human tumors. Therefore, we designed the current study to correlate qMRI changes from daily RT treatment with underlying tumor transcriptomic profiles. We found that multiple genomic pathways including DNA repair, peroxisome, late estrogen receptor responses, KRAS signaling, and UV response were significantly associated with qMRI feature changes (P < .001), indicating that common tumor biology that may correlate with qMRI changes during the course of treatment. Such data provide hypothesis-generating novel mechanistic insight into the biologic meaning of qMRI changes during treatment and enable optimal selection of imaging bio-markers for biologically MR-guided RT.

P²T²: Protein Panoramic annoTation Tool for the interpretation of protein coding genetic variants

Elias DeVoe, Gavin R. Oliver, Roman Zenka, Patrick R. Blackburn, Margot A. Cousin, Nicole J. Boczek, Jean-Pierre A. Kocher, Raul Urrutia, Eric W. Klee, and Michael T. Zimmermann

JAMIA Open

Our tool assists research efforts to interpret genomic variation by aggregating diverse, relevant, and proteome-wide information into a unified interactive web-based interface. Additionally, we provide a REST API enabling automated data queries, or repurposing data for other studies. The unified protein-centric interface presented in P2 T2 will help researchers interpret novel variants identified through next-generation sequencing.

Benchmarking deep and shallow methods for Hi-C count prediction within and across species

RSG-DREAM 2023 (UCLA)

Wisconsin Institute for Discovery 2023 Symposium (UW - Madison)

Canine Classifier

See which of the ImageNet dog classes is closest to yours. On mobile, be sure to enable browser access to your camera.

Reductiscope

The system proven (by one very successful data point) to lose weight with self-competition.