Introduction

A cell is the basic unit of life, but how its components interact to ensure cell functioning remains incompletely understood. My motivation is to generate a mathematical model of the human cell, specifically a predictive understanding of how gene regulation at a genome-wide scale controls human cellular phenotypes. Throughout my scientific career, I have demonstrated repeatedly that my mathematical modeling and machine learning (AI/ML) approaches to answer questions in cell and RNA biology lead to novel discoveries. The key to this systematic success is that my modeling results in specific and often unintuitive mechanistic hypotheses, which, upon experimentally validation, turn out to have a high truth rate. My research program delivers theoretical innovations yet is complemented by experimental genomic technology development. Overall, I am uniquely positioned to expand this systems biology approach that paves the way for novel cell therapies, and ultimately improves human health.

Research highlights

Most recently, I have extended my graduate work to study the life cycle of all cellular RNAs simultaneously in human and mouse cells. Here, I quantified the kinetics of RNA flow across subcellular compartments through the development of novel integrated genomic and AI/ML technology, in an equal contribution collaboration in the Churchman lab. Preprint on bioRxiv.

During my postdoc, I have developed GeneWalk, an ML tool that identifies the most important genes and their relevant functions from any input list of experimental gene hits. GeneWalk is available as an open source software package for the biomedical research community, with a tutorial to get started.

In a collaboration between researchers from Novartis and academic institutions, we used ML to identify associations between drug target genes and adverse drug reactions. Our models can be used to predict at the preclinical stage whether therapeutic candidates are likely to cause adverse events. Detailed info in our EBioMedicine publication.