Sequence-to-function deep learning decodes human cis-regulatory evolution
Predicting lineage-specific regulatory activity to uncover functional changes beyond elevated sequence divergence.
Research ScientistMassachusetts Institute
of Technology
Postdoctoral ScholarBroad Institute of MIT
and Harvard
Cambridge, Massachusetts
rimangan [at] mit [dot] eduEvolutionary & regulatory genomics
I seek to understand the gene regulatory changes that shaped human neurodevelopment and how their evolutionary history influences susceptibility to neurological and neuropsychiatric disease.
My research combines comparative genomics, functional genomics, and genomic deep learning to identify consequential regulatory variants, determine their molecular and cellular mechanisms, and uncover how regulatory innovations reshape functional constraint.
I work with Prof. Manolis Kellis in the MIT Computational Biology Group and at the Broad Institute. I received my Ph.D. in Molecular Genetics and Microbiology from Duke University, advised by Prof. Craig Lowe.
Identify evolutionary changes in regulatory function, from rapidly evolved DNA to consequential variants overlooked by sequence divergence alone.
Combine sequence-to-function deep learning with functional genomics, including in vivo scSTARR-seq, to reveal the mechanisms of regulatory change.
Investigate how regulatory innovations reshape functional constraint and influence susceptibility to neurological and neuropsychiatric disease.
* Co-first author · † Co-corresponding author
Predicting lineage-specific regulatory activity to uncover functional changes beyond elevated sequence divergence.
Using complete ape genomes and within-species variation to refine the search for human regulatory innovation.
Discovering Human Ancestor Quickly Evolved Regions (HAQERs) and testing their enhancer activity in the developing brain.
I am a lead developer of Gonomics, an open-source collection of high-performance genomics tools written in Go. The project brings together readable code and scalable methods for genomic data analysis.

I mentor undergraduate and graduate researchers, visiting scholars, and high-school students in computational genomics, evolutionary genomics, and machine learning. At MIT, I developed curriculum and lecture materials for Machine Learning for Computational Biology in 2024 and 2025 and mentored semester-long student projects. I also delivered an invited lecture on human evolutionary neuroscience at Davidson College.