Published Research

Six peer-reviewed papers on applying machine learning to cancer biology, immunotherapy prediction, and precision medicine. Published in Nature Communications, npj Precision Oncology, Genome Biology, and Cancer Research.

Pooling multimodal cancer data across unaligned embedding spaces maintains tumor of origin signal Latest

First Author

Bioinformatics Advances · July 2026

Developed machine learning algorithms that preserve the underlying biological signal when combining different types of cancer data, enabling more accurate analysis across heterogeneous datasets.

Read paper

Predicting anti-PD-1 immune checkpoint blockade response in melanoma patients with spatially aware machine learning models

First Author

npj Precision Oncology · January 2026

Used single-cell spatial proteomics alongside machine learning to predict how advanced melanoma patients respond to immunotherapy. Models correctly predicted treatment response in 11 of 12 patients by analyzing the tumor microenvironment.

Read paper

Imputing single-cell protein abundance in multiplex tissue imaging

First Author

Nature Communications · May 2025

Introduced a computational approach to estimate protein levels in tissue images, making complex biological analysis more accessible and interpretable for clinical research.

Read paper

SyntheVAEiser: Augmenting machine learning methods with VAE-based gene expression sample generation

Co-Author

Genome Biology · December 2024

Built a tool that generates synthetic cancer gene expression data to improve machine learning performance on rare cancer subtypes. Trained and validated on over 8,000 cancer samples.

Read paper

Tumor model to tumor treatment: Applying deep learning to map multimodal cancer data

Co-Author

Cancer Research · March 2024

Developed a deep learning strategy to correct for technical differences across cancer drug response datasets, improving the ability to predict how tumors respond to treatment across different experimental systems.

Read paper

Computational pipeline to identify gene signatures defining cancer subtypes

Senior Author

bioRxiv · November 2022

Created a pipeline to identify the genetic patterns that distinguish cancer subtypes, demonstrating how data-driven approaches can uncover clinically meaningful differences in tumor biology.

Read paper