Research Program
Our research weaves five interconnected lineages -- from AI-driven drug discovery to translational data infrastructure -- into a unified effort to make artificial intelligence truly useful in medicine.
Systems Pharmacology & Drug Discovery
Computational identification of drug targets, AI-driven therapeutic development, and drug repurposing using network pharmacology, neuro-symbolic AI, and large language models. From early connectivity map approaches to modern LLM agent swarms for hypothesis-driven drug discovery.
Key Methods
- Network pharmacology
- Drug repurposing (PETS, C2Maps)
- Neuro-symbolic AI (QSAR)
- LLM agent swarms
- Antibody-antigen binding prediction
- ADMET & hERG cardiotoxicity prediction
Lab Software
Representative Publications
- LLM Agent Swarm for Hypothesis-Driven Drug Discovery (2025)
- LlamaAffinity: A Predictive Antibody-Antigen Binding Model Integrating Antibody Sequences with Llama3 Backbone Architecture (2025)
- NeSyDPP4-QSAR: Discovering DPP-4 Inhibitors for Diabetes Treatment with a Neuro-symbolic AI Approach (2025)
- Integrative multi-scale network simulation for precision drug repurposing with PETS (2025)
- An NLP-based Technique to Extract Meaningful Features from Drug SMILES (2024)
- C2Maps: A Network Pharmacology Database with Comprehensive Disease-Gene-Drug Connectivity Relationships (2012)
Biomedical Knowledge Networks & Network Biology
Construction of large-scale knowledge graphs, pathway/gene-set repositories, protein interaction databases, and network biology methods. From the foundational HAPPI protein interactome and PAGER gene-set repository to modern network-based biomarker discovery and drug target prioritization.
Key Methods
- Gene set / pathway enrichment (PAGER, PAGED)
- Protein interaction networks (HAPPI, WIPER)
- Biomedical entity expansion (BEERE)
- Ontology-guided analysis (GOALS)
- Network topology & reordering
- Semantic web & data mining
Lab Software
Representative Publications
- GOALS: Gene Ontology Analysis with Layered Shells for Enhanced Functional Insight and Visualization (2025)
- AI-Driven Network Biology Identifies SRC as a Therapeutic Target in Metastatic Pancreatic Adenocarcinoma (2025)
- Toden-E: Topology-Based and Density-Based Ensembled Clustering for the Development of Super-PAG in Functional Genomics (2025)
- WIPER: Weighted in-Path Edge Ranking for Biomolecular Association Networks (2019)
- PAGER: Constructing PAGs and new PAG-PAG Relationships for Network Biology (2015)
- HAPPI: an Online Database of Comprehensive Human Annotated and Predicted Protein Interactions (2009)
Multi-omics, Visualization & Interpretable AI
Pioneering visual analytics for high-dimensional biological data. From the original GeneTerrain platform to Mondrian-inspired differential pathway analysis and single-cell spatial embedding tools (PGC, SpatialRSP).
Key Methods
- Terrain-based expression visualization
- Single-cell spatial embeddings (PGC, SpatialRSP)
- Mondrian abstraction for pathway analysis
- Network layout algorithms (DEMA, GraphWaGu)
- Multi-omics integration
- Kinome profiling visualization
Lab Software
Representative Publications
- Temporal GeneTerrain: Advancing Precision Medicine Through Dynamic Gene Expression Visualization (2025)
- KinoViz: A User-Friendly Web Application for High-Throughput Kinome Profiling Analysis and Visualization in Cancer Research (2025)
- Mondrian Abstraction and Language Model Embeddings for Differential Pathway Analysis (2024)
- Polar Gini Curve: a Quantitative Technique to Discover Gene Expression Spatial Patterns from Single-cell Data (2022)
- DEMA: a Distance-bounded Energy-field Minimization Algorithm to Model and Layout Bio-molecular Networks (2022)
- GeneTerrain: Visual Exploration of Differential Gene Expression Profiles Organized in Native Biomolecular Interaction Networks (2010)
Digital Twins & Precision Medicine
Building patient-level computational models that fuse genomics, phenomics, and clinical data into personalized digital twins for cancer simulation, treatment optimization, and precision drug screening.
Key Methods
- Multi-scale digital twin simulation (MLPA)
- Sample-level statistical enrichment (SEAS)
- Clinotype-phenotype-genotype linking
- Cancer systems pharmacology
- Biomarker discovery & prioritization
- Personalized drug response prediction
Lab Software
Representative Publications
- MLPA: A Multi-scale Digital Twin Framework for Personalized Cancer Simulation and Treatment Optimization (2024)
- Statistical Enrichment Analysis of Samples (SEAS): a General-purpose Tool to Annotate Metadata Neighborhoods of Biomedical Samples (2021)
- Linking Clinotypes to Phenotypes and Genotypes from Laboratory Test Results in Comprehensive Physical Examinations (2021)
- Network Medicine: Finding the Links to Personalized Therapy (2013)
- Unraveling Human Complexity and Disease with Systems Biology and Personalized Medicine (2010)
Translational Infrastructure, Data Ecosystems & Team Science
Designing and deploying the data platforms, AI-ready data ecosystems, and multi-institutional collaboration frameworks that underpin biomedical AI. Includes leadership of the NIH-funded CONNECT consortium and Bridge2AI program, the U-BRITE translational platform, COVID-19 data resources, and the UAB CCTS bioinformatics core.
Key Methods
- NIH Common Fund data ecosystems (CONNECT, Bridge2AI, CFDE)
- Translational informatics platforms (U-BRITE)
- COVID-19 data integration (PAGER-CoV, N3C)
- AI-ready cell architecture maps (Cell Maps for AI)
- Talent knowledge graphs & team science analytics
- Privacy-preserving health informatics
Lab Software
Representative Publications
- Cell Maps for Artificial Intelligence: AI-Ready Maps of Human Cell Architecture from Disease-Relevant Cell Lines (2024)
- PAGER-scFGA: Unveiling Cell Functions and Molecular Mechanisms in Cell Trajectories through Single-Cell Functional Genomics Analysis (2024)
- PAGER-CoV: A Comprehensive Collection of Pathways, Annotated-gene-lists, and Gene Signatures for Coronavirus Disease Functional Genomic Studies (2021)
- Empowering Team Science Across the Translational Spectrum with the UAB Biomedical Research Infrastructure Technology Enhancement (U-BRITE) (2020)
- PAGER Web APP: An Interactive, Online Gene Set and Network Interpretation Tool (2022)
Research Funding
Over $100M in cumulative research funding from 60+ grants since 2004
Current Grants
CONNECT: Collaborative Network for Nurturing Ecosystems of Common Fund Team Science
Contact MPI · PI: Jake Y. Chen · 2024-2029
ReCARDO: Using Real-World Data to Derive Common Data Elements for AD/ADRD Research
Co-I & UAB Site PI · PI: Guo-Qiang Zhang · 2024-2029
Building an Interpretable Genomic Translator Using Maps of Cell Architecture
MPI · PI: Trey Ideker · 2022-2026
UAB-UCSD O'Brien Center for Acute Kidney Injury Research
Co-I · PI: Anupam Agarwal · 2023-2028
Computing for AI-enabled Systems Pharmacology and Drug Discovery
PI · PI: Jake Y. Chen · 2025-2026
UAB Center for Clinical and Translational Science (CCTS)
Co-I · PI: Orlando Gutierrez · 2024-2031
Selected Past Funding
Prior funded research includes major NIH-supported programs in cancer systems biology (U01CA223976, R01CA258248), immunology (R01HL150078, R01AI134023, R01AR073850), diabetes and metabolic diseases (R21DK129968), translational science infrastructure (U54TR001005, UL1TR001417, 3UL1TR003096), as well as Department of Defense grants in neurodevelopmental research and cardiac regeneration through the NHLBI Progenitor Cell Translational Consortium.
Research Collaborators
Multi-institutional partnerships advancing biomedical AI
Cross-Lineage Synergies
The power of the AI.MED lab lies in the intersections: knowledge networks feed drug discovery models, digital twins consume multi-omics visualizations, and every insight travels through our translational data infrastructure on its way to patient care. We design our research program so that progress in one lineage accelerates all the others.