Decoding the language of genomes. Bridging machine learning and molecular biology to discover how cells make decisions under stress.
Associate Professor
Dept. of Computational Biology, Stanford University
47 publications across Nature, Science, Cell, and top-tier ML venues. 1,240+ total citations.
We present ChromFormer, a transformer architecture trained on 2.3M single-cell ATAC-seq profiles that predicts chromatin accessibility with 94% accuracy across 42 cell types.
Using integrated RNA-seq, ATAC-seq, and proteomics, we characterize 847 stress-responsive regulatory elements in neural progenitors, revealing a conserved transcriptional circuit.
ProteGraph achieves state-of-the-art PPI inference across 18 organisms by encoding evolutionary conservation as edge attributes in a heterogeneous graph architecture.
CausalTracer identifies 23 causal regulatory drivers of cell fate bifurcation in hematopoiesis, validated by CRISPR perturbation screens in 3 independent cohorts.
FedGenome enables genome-wide association studies across 14 institutions without sharing raw patient data, achieving 97% concordance with centralized analysis.
A systematic evaluation of 19 sequence-to-function models across 8 genomic prediction tasks, establishing the RegBench framework and identifying key architectural trade-offs.
From decoding protein networks as an undergrad in Toronto to leading a 12-person lab at Stanford — my research has always been driven by one question: how do cells know what to become?
B.Sc. Bioinformatics
Graduated Summa Cum Laude
M.Sc. Computational Biology
Thesis: Graph-based methods for regulatory network inference
Ph.D. Computational Biology
Advisor: Prof. Michael Chen · NSF Graduate Fellowship
Postdoctoral Fellow
Focus: Single-cell genomics and ML foundations
Associate Professor
Dept. of Computational Biology · Vasquez Lab
Core Competencies
Five interconnected research streams bridging computational methods with molecular biology.
Transformer and GNN architectures trained on large-scale genomic datasets to predict regulatory element activity.
Pre-training DNA language models on 3B+ base pairs to learn generalizable representations of regulatory grammar.
Integrating genomic, transcriptomic, proteomic, and epigenomic layers to build holistic models of cellular state and fate.
Trajectory inference, cell type annotation, and perturbation modeling at single-cell resolution across 2M+ profiled cells.
Federated learning, differential privacy, and algorithmic fairness for equitable genomic medicine.
Education
Ph.D. Computational Biology
NSF Graduate Research Fellowship
M.Sc. Computational Biology
GPA 4.0/4.0
B.Sc. Bioinformatics (Hons.)
Summa Cum Laude · Gold Medal
Awards & Grants
NSF CAREER Award
Causal AI for single-cell biology
NIH R01 Grant (PI)
Multi-omics integration in neurodegeneration
Chan Zuckerberg Initiative
Collaborative science grant
Service & Leadership
Associate Editor
Area Chair
Vasquez Lab Director
12 graduate students & postdocs
12-page comprehensive CV including all publications, grants, teaching experience, invited talks, and service roles.
Updated June 2026 · 12 pages
Recent Invited Talks
Foundation models for genomic regulation
NeurIPS 2025 · Vancouver, BC
Causal inference in single-cell trajectories
ISMB 2025 · Montreal, QC
ChromFormer: lessons learned
Broad Institute Seminar · Cambridge, MA
Open to collaboration proposals, seminar invitations, media inquiries, and prospective graduate students.
Office
Clark Center S241, Stanford, CA 94305