Summary of Projects, 2025
Research Overview Generation with LLMs
Multimodal Foundation Models in Medicine: Promise, Pitfalls, and Performance
Defining “Normal”: Personalized Interpretation of Routine Blood Labs
Mixed Lineages and Variant Landscapes in TB Sampling
A Machine Learning Model to Predict Phage Infection Susceptibility for M. abscessus
Tracking Micro-variants: Within-Host Evolution of Mycobacterium Tuberculosis During Treatment
SEMPER: SystEm-Scale Multimodal ProtEin Representations
Pilot Study to Identify Undiagnosed Patients Utilizing Anomaly Detection Methods on Medical Claims Data
Proteomic Insights into Stroke Subtypes, Cardiovascular Disease, and Pharmacotherapy in the UK Biobank
TargetLinker: Harnessing our Body’s “Dark Matter”
Genetic and Modifiable Risk Factors of Alzheimer’s Disease
Integrative Single-cell Machine Learning Reveals Transcriptional Programs and Cell-type Vulnerabilities in Alzheimer’s Disease
Metabite: Advancing Data-Driven Prediction of Cardiovascular Disease from Metabolic Profiles
BM-why? Investigating Biologically Relevant Indicators of Obesity
Determining the Prevalence of Insertion Sequence Mediated Heterodimerization Events
Beyond the Slice: A New Framework for Cell Interaction in 3D Spatial Omics
IsoFormer: Virtualizing the Splicing Machinery of the Cell
The Foundation for AI-Powered Genomics Exploration: A Dataset of Genomics Questions and Visualization Responses
Identifying a Gene Signature for TP53 in LFS Patients
Transfer Learning of Single-Cell Foundation Models on Organoid Data
Summary of Projects, 2024
Compositional Reasoning Beyond Generation: From Pattern Recognition to Agentic Generalist Medical AI
Preceptron: Towards Automated Assessment of Large Language Models in Medicine
Filling in the Gaps: Elucidating Cell Type to Biomarker Relationships in Single Cell Data
Effective Use of Mobile Health Map Data to Visualize and Inform Data-Driven Population Health Management
Self-and Semi-supervised Learning for Kidney Function Prediction
Beyond Data: Harnessing Computer Vision in the Operating Room
Fine-grained Alignment of Abnormalities in Medical Imaging and Reports
Towards Celltype-Specific PPIs
A Knowledge-Grounded Foundation Model for Precision Medicine
Large Genetic Variants Associated with Extrapulmonary Tuberculosis
Comparing Genetic Diversity in Direct Sputum and Culture Samples of Mycobacterium Tuberculosis
Mitigating Race-Based Medicine Outputs in LLMs
Large-scale Genomic Instability Across Non-Cancerous Human Tissues with scATAC-seg
AI in Pathology: Autonomous Microscopy for Brain Tumor Diagnosis
Uncertainty Estimation in Radiology Reports
Biomedical Agents for Personalized Drug Repurposing
Summary of Projects, 2023
The Human BioMolecular Atlas Program (HuBMAP)
HuBMAP Data Integration in Mixed Reality: Building an Immersive High-Resolution Human Atlas
Estimating dosage sensitivity using gene expression in copy-number-defined tumor subclones
Multimodal Pharmacology AI for Drug Interactions and Safety
Contextual Deep Learning on Knowledge Graphs for Precision Medicine in Neurological Disorders
Protein Function Prediction with Graph Neural Networks
Benchmarking Vision-Text Biomedical Models with CLIPContrast
AI in Pathology: Autonomous Microscopy for Brain Tumur Diagnosis
Beyond Data: Harnessing Computer Vision in the Operating Room
Unraveling CD8+ T cell development throughout childhood via multi-omics
Reducing the dimensions of the environment
Visualizing chronic disease burden and mobile health van presence across the United States
Identifying areas of highest need with a mobile clinics dashboard
Using KidneyDB to Examine the Inclusion of Race in Kidney Function Estimates
Cardiokid: R-based platform to generalize NHANES analysis
Fine-tuning LLMs to be copilots in high-quality epidemiological research
Characterizing Promoter Variation in Mycobacterium Tuberculosis to Predict Drug Resistance
Approaches for Generating Homoplasy Scores in SARS-COV-2 Without a Phylogenetic Tree
Summary of Projects, 2022
Developing a Method to Semi-Automate Harmonization of Clinical Variables Across Multiple Large-Scale Cohorts
Comparing the Predictive Value of Clinical Variables for Type 2 Diabetes Onset Based on Social Determinants of Health in the Framingham and Jackson Heart Studies
Multi-Cohort Variable Harmonization: Semantic and Variable Distribution Learning
Edge Bundling for Enhanced Genomics Visualizations
LINCS Compound Dataset And Causal-PPI
Assessing BMI and other Adiposity Measurements in Type 2 Diabetes Risk Prediction & the Variation in Differing Community Based Cohorts
Detection of Proper Endotracheal Tube Placement via Computer Vision
Developing Vitessce, a Single-Cell Sequencing Tool, to Investigate Immune Checkpoint Blockades
Validation of GenTB Resistance Prediction with User-Submitted Mycobacterium tuberculosis Sequencing Data
Predicting the Onset of Diabetes Using Cardiac History and Social Determinants of Health
Mycobacterium Tuberculosis GWAS to Associate Genotype to Extrapulmonary TB
Prevalence and Selection of Pediatric Suicidality Cases in the Emergency Department Before and During the COVID-19 Pandemic
Genetic and Geospatial Shared Environment Associations with Individual-Level Exposures Expand Understanding of Cardiometabolic Health Etiology
Randomized Trials of AI in Clinical Practice
Finding Aberrant Gene Expression and Splice Junctions from RNA Sequencing Data in Undiagnosed Patients
Power Calculations for Genetic Variation Studies Focused on Severe COVID-19
PYPE: A Python Pipeline for PheWAS Execution, Visualization, and Interpretation
National Projections for Clinical Applications of New Kidney Function Equations in Asian Americans, 2011-2018
Automatic Detection of Endotracheal Tube Placement in Chest X-Rays
Using Multimodal Modeling for Novel Drug-Drug Interaction Prediction
Summary of Projects, 2021
Quantifying the Constraint on Missense and LoF Variants in the Human Genome Using a Fine-Scale Mutation Map
Designing a Mobile Platform for Personal Medical Data Exploration
Clinical Analysis and Evaluation of Gene-Phenotype Relationships in the Undiagnosed Disease Network
Liver Disease and Environmental Exposures: an Exposome-Wide Association Study
Somatic Mutation Calling in Slide-SEQ Prostate Cancer Samples
Marker Gene Panel Selection for Imaging-Based, Spatially-Resolved Transcriptions
Structural RNA Velocity
Role of Tandem Repeat Expansions in Ewing Sarcoma
Detecting Microsatellite Instability in Cancer Genomes
Translational Data Science Center for a Learning Healthcare System
Pediatric Reference Ranges: Applying Direct and Indirect Methods to NHANES Data
Race-Free Equations for Spirometry
What Do We Do When Machine Learning Models and Clinicians Disagree?
Ultra-Rare Epilepsy Genetics in the UDN
Co-Expression of Neighboring Genes Using Next Generation Sequencing
Network Diffusion Based Approach for Disease Gene Discovery
GraphXAI: Benchmarking Explainablity Methods for Graph Neural Networks
Characterizing Drug Mechanisms with Cell Type Specific Protein Networks
Segmenting Glioblastoma Slides Using Graphical Neural Networks
Genomic Data Visualization and the Implementation of Themes into Gosling
GenTB
Intragenome Recombination in M. tuberculosis
Investigating Viral Genetics of SARS-CoV-2 Patients in Bahrain
Summary of Projects, 2020
Detecting Mutational Signatures in Cancer Cells
Analyzing Relationship Between Social Determinants of Health and Multidrug-Resistant
Tuberculosis and Developing Data Visualizations and Resistance Prediction Pipeline on the genTB
Website
Predicting Bronchopulmonary Dysplasia in NICU Infants with Claims Data
Towards Deconvolution of Spatial Transcriptomics Data
Evolution of Mycobacterium Abscessus within Human Infection
Summary of Projects, 2019
Unsupervised Data Augmentation for Semisupervised Learning of Patient Outcomes
Analyzing the Source of Overdispersion in Allelic Imbalance Estimates
R Shiny App on HIV Positivity in Sub-Saharan Africa
Developing Infrastructure for Cell Set Operations and Organization in Single-Cell Visualization
Discerning Simple Phenotype from EEG: Patient Biometrics and Demographic Tasks using CNN Methodologies
'An Experimental Framework for Examining Hyperparameter Choices in Crowdsourced Environments
Of Mice and Men: Assessing Schizophrenia Drug Success in the Context of Human-Mouse Brain Homology
Prediction of Chronological and Biological Age from Laboratory Data
Investigating Biological Similarities among Rare Diseases