Aakash Patel
I build machine-learning methods for biological data: cell sentences for single-cell genomics, ligand–receptor priors for spatial transcriptomics, and agentic systems for fMRI analysis.
Ph.D. student in Computer Science, Yale University (advised by David van Dijk)
About
I am a Ph.D. student in Computer Science at Yale, advised by Dr. David van Dijk. My research develops machine-learning methods for biological data, including representing single-cell gene expression as natural language (Cell2Sentence), modeling cell–cell interactions in spatial transcriptomics with ligand–receptor priors, and vision–language models and agentic systems for interpreting fMRI recordings.
I am interested in interpretable and multimodal models, few-shot evaluation, and workflows that integrate prior biological knowledge. Before graduate school I worked as a software engineer and completed a dual Masters at the University of Michigan in Computer Science and Mathematics.
-
presentPh.D. in Computer Science, Yale University
-
priorM.S. Computer Science & M.S. Mathematics, University of MichiganSoftware engineer at EPIC
Publications
-
The AI Neuroscientist: An Interactive Agentic Interface for Neuroimaging Analysis
BrainBodyFM and AgenticLS workshop at NeurIPS 2026.
-
Induced T1 Degradation Selectively Destabilizes Surface-Mapped fMRI Representations
BrainBodyFM workshop at NeurIPS 2026.
-
Benchmarking foundation models for immunotherapy response prediction in renal cell carcinoma: Data, not architecture, is the current bottleneck.
Abstract accepted to SITC 2026.
-
Scaling Large Language Models for Next-Generation Single-Cell Analysis
-
Intelligence at the Edge of Chaos
-
Neural Posterior Estimation for Cataloging Astronomical Images with Spatially Varying Backgrounds and Point Spread Functions
News
-
Aug 2026Accepted to the inaugural cohort of Yale's Peter Salovey and Marta Moret Data Science Fellows Program.
-
Apr 2026Presented Knowledge-Grounded Reinforcement Learning for High-Fidelity Single-Cell Cluster Captioning at the AI at Yale Symposium.
-
Mar 2026Presented Predicting renal cell carcinoma outcomes from clinical trial data using multimodal genomic foundation models at the Yale AI in Cancer Workshop and the Yale Medical AI Symposium.
-
Feb 2026Organized the Toward an AI Neuroscientist workshop at the Wu Tsai Institute.
-
Jan 2026Invited talk on BrainLM at the Columbia Eating Disorders Research Seminar. slides
-
Sep 2025Invited talk on NPE for Cataloging Astronomical Images at the Yale Data Science × Astrophysics Seminar. slides