About me

I am a PhD candidate at the Center for Computational Molecular Biology at Brown University, advised by Dr. Lorin Crawford and Dr. Jeffrey Bailey. My work sits at the intersection of machine learning, Bayesian statistics, and translational biology.

My research focuses on a single through-line: how do we build AI models that extract reliable, interpretable signal from the messy, heterogeneous biological data that actually exists in the real world? That means genomic surveillance data with sparse sampling, multi-omics cohorts with missing modalities, and clinical datasets with confounded outcomes.

I work closely with domain experts such as biologists, clinicians, and public health scientists to ensure that models don’t just perform well in silico, but directly improve how decisions are made in biotechnology, pharmaceutical research, and public health settings. My work has been cited in the WHO World Malaria Report 2025.

In July 2026, I was awarded the Blavatnik Family Graduate Fellowship at Brown University, which selects recipients for outstanding academic achievement and demonstrated potential to conduct original research that advances scientific knowledge and understanding in the basic and clinical life sciences. I was also a Frontera Computational Science Fellow 2025-2026, an NSF-funded Fellowship at the Texas Advanced Computing Center, University of Texas, Austin.

I am currently a Computational Biology Intern at Takeda Pharmaceutical (Jun–Aug 2026). In the spring of 2026, I was a Visiting PhD Researcher at Oxford University and Imperial College London working with Dr. Seth Flaxman and Dr. Robert Verity. During the summer of 2025, I was a Research Intern at Microsoft Research New England.

Research Interests

• Interpretable ML
• Generative models
• Biomarker discovery
• Drug discovery
• Multi-omics integration
• Clinical decision-making
• Translational AI
• Precision medicine
• Infectious disease

Mentors

I have over seven years of research training, advised by:

• Dr. Lorin Crawford (Microsoft Research and Brown University)
• Dr. Jeffrey Bailey (Center for Computational Biology, Brown University).
• Dr. Ashley Conard (Microsoft Research)
• Dr. Robert (Bob) Verity (Imperial College London)
• Dr. Oliver James (OJ) Watson (Imperial College London)
• Dr. Mary L. Gray (Microsoft Research)
• Dr. Franziska Michor (Department of Data Science, Dana-Farber Cancer Institute)
• Dr. Simona Cristea (Department of Data Science, Dana-Farber Cancer Institute)

Key research projects

Transformer-Based Generative Model for Population Genetic Inference

A forward-time generative simulator and transformer-based deep learning model for simulation-based inference of population genetic dynamics. Generates large-scale synthetic biological sequence trajectories under realistic evolutionary dynamics, enabling downstream ML inference without explicit likelihoods. A novel framework applicable to genomic and protein sequence modeling.

Multimodal AI for Personalized Blood Biomarker Prediction

A multimodal generative framework that integrates genomic, clinical, and environmental data to predict blood biomarkers, moving beyond genetic determinism to provide quantitative, modality-level interpretability. Enables actionable insights into which data types drive disease-linked predictions, directly supporting personalized medicine pipelines.

Spatiotemporal Bayesian Model for Drug Resistance Surveillance

A Bayesian spatiotemporal ML framework that maps the continuous prevalence of antimalarial drug resistance mutations across Africa from sparse, heterogeneous genomic surveillance data. Provides calibrated uncertainty quantification for early hotspot detection, before clinical efficacy is lost, enabling proactive drug strategy decisions at scale.

Preprint and Code.

Clinical Relevance of Toxicity Metrics in Drug Combination Models

Systematic evaluation of whether ML-driven synergy scores and toxicity metrics in cancer drug combination models reflect clinically observed adverse interactions. Integrated large-scale synergy datasets with curated clinical toxicity databases, revealing a critical gap: optimizing for synergy alone can inadvertently prioritize toxic combinations. Directly informs safer ML-guided drug discovery pipelines.

Paper in Bioinformatics and Code.

Bayesian Mixed-Effects Modeling for Drug-Resistance Forecasting

Bayesian mixed-effects model estimating the speed of antimalarial drug-resistance spread across Uganda and southeast Asia from sparse genomic surveillance data. Forecasts show combined resistance mutations could reach near fixation within a decade. These findings were cited in the WHO World Malaria Report 2025 and directly inform global drug policy. Demonstrates how calibrated, uncertainty-aware ML can turn incomplete real-world biological data into actionable public health decisions.

Paper in Lancet Microbe and Code.

Interpretable Evolutionary Modeling of Tumor Progression from scRNA-seq

An evolutionary ML framework integrating copy-number variation features from single-cell RNA-seq data to classify tumor evolutionary modes across cancer types. Random Forest ensemble learning captured non-linear interactions underlying clonal expansion, intratumoral heterogeneity, and treatment pressure — providing interpretable insights into cell-state dynamics for drug target identification.