Machine Learning for Therapeutic Intervention & Biological Dynamics
PhD Candidate · Computer Science, University of Minnesota
Predoctoral Fellow · National Library of Medicine, NIH
I develop machine-learning methods to understand how biological systems respond to interventions and translate those representations into better therapeutic predictions.
Academic Website · Google Scholar · Research Statement · Email
| Research direction | Focus |
|---|---|
| 01 · Perturbation modeling | Learn how drugs change cellular states and how these changes generalize across conditions. |
| 02 · Therapeutic response prediction | Transfer molecular representations from experimental systems to drug-response prediction. |
| 03 · Evidence-aware ML | Connect computational predictions with biological networks and literature to support interpretation. |
|
From single-cell perturbations to patient drug response Learns drug- and dose-conditioned latent transitions from single-cell populations, then transfers these signals to patient-level treatment-response prediction. |
Interpretable drug response prediction with graph learning Models drug–cell–gene relationships to predict sensitivity and investigate gene-level contributions. |
|
Multi-agent reasoning over biomedical evidence Combines machine-learning scores, knowledge-graph signals, and literature retrieval to assess drug–target interactions with evidence-linked reasoning. |
A provenance-aware oncology knowledge graph Explore relationships among therapeutics, targets, diseases, and clinical studies while tracing the evidence behind each connection. |
| Repository | Description |
|---|---|
| awesome-computational-biology | Curated datasets, databases, software, and papers for computational biology. |
| research_ideas | Research questions, hypotheses, and experimental directions. |