Michael Akpabey

I’m a Computational Biologist and trained Toxicologist at the Broad Institute of MIT and Harvard, working at the intersection of quantitative biology, machine learning, and drug discovery. I develop computational methods that turn complex biological data into actionable insights for functional genomics, target discovery, perturbation biology, and translational research.

At the Broad, I lead computational development for PROSPECT, a large-scale perturbation-screening platform supporting 100K+ compound screens from barcode-level QC and dose-response modeling through hit prioritization and mechanism-of-action inference. I also develop statistical approaches for gene-set analysis, benchmark ML virtual-screening models across billion-scale chemical libraries, and work closely with experimental scientists to translate computational results into assay decisions and mechanistic hypotheses.

My work spans functional genomics, CRISPRi, perturbation screening, multi-omics, transcriptomics, epigenomics, and microbial genomics, using approaches including statistical modeling, clustering, dimensionality reduction, enrichment analysis, concordance methods, and machine learning. Earlier research integrating RNA-seq and DNA methylation from human biospecimens, together with my training in toxicology and preclinical research, gives me additional perspective on biomarkers, dose response, pharmacology, and translational biology.

I also focus on turning analytical methods into reproducible scientific systems. I build scalable workflows in Python and R using Nextflow, Docker, HPC, and cloud infrastructure, and have developed provenance-aware biological knowledge systems and LLM-assisted workflows for integrating genomic, metabolic, pathway, and literature evidence.

I’m particularly interested in opportunities where computational biology, quantitative methods, and AI/ML directly inform biological and therapeutic decisions—especially in functional genomics, target discovery, perturbation biology, multi-omics, biomarker research, scientific AI, and scalable drug-discovery platforms.