PhD Candidate in Computer Science at the University of Minnesota
Pre-doctoral Fellow at NLM / NCI, National Institutes of Health

Representation learning and structured machine learning for therapeutic response modeling,
computational drug discovery, and precision medicine.

🌐 Website · 📄 CV · 🎓 Google Scholar · 🧬 ORCID · 💼 LinkedIn

My research develops machine-learning methods for modeling therapeutic response across molecular, cellular, and patient contexts. I am particularly interested in predictive representations, structured biological knowledge, and robust modeling under intervention.

• Representation learning — learning predictive molecular, cellular, and patient representations that preserve treatment-relevant biological context.
• Perturbation / response modeling — modeling treatment-conditioned changes in biological state and generalizing response predictions across drugs, cells, and patient contexts.
• Structured / causal modeling — using biological structure and causal principles to improve interpretability, robustness, and generalization under intervention.

My PhD work follows a progression from interpreting drug response, to reasoning across biomedical evidence, to learning treatment-conditioned representations for response prediction.

Attention-guided gene assessment of drug response using a drug–cell–gene heterogeneous network.

Reliable multi-agent aggregation for biomedical evidence synthesis and computational drug discovery.

Ongoing thesis research on predictive representations for therapeutic perturbation and patient-response modeling.

More projects and publications are available on my academic website.

I maintain a small set of open-source repositories for organizing research, evaluating ideas, and reusing scientific tools:

• research_ideas — capture, compare, and evaluate research questions before committing substantial research time.
• research_log — preserve research decisions, lessons, negative results, and open questions.
• research_toolbox — reusable utilities and templates for computational biology and scientific workflows.
• awesome-computational-biology — curated databases, software, datasets, papers, and resources in computational biology.

Together, these repositories support a lightweight workflow from existing resources → research ideas → projects and experiments → lessons and reusable tools.

• University of Minnesota — PhD Candidate, Computer Science and Engineering
• National Library of Medicine / National Cancer Institute, NIH — Pre-doctoral Fellow
• Location: Bethesda, Maryland