Amarda Shehu
AI researcher, university AI leader, and computational biology expert
amarda-shehu
Fairfax, Virginia, USA
Joined May 2026
Summary
Institutional AI leader shaping university-wide AI strategy, governance, and education. gmu+1
Researcher bridging foundational AI methods and molecular/biological applications, with sustained work in computational biology, protein modeling, generative models, and AI for scientific discovery. gmu+2
Academic leader and mentor with a record of securing competitive research funding, directing centers and institutes, and serving in national research leadership roles (NSF program director). nd+2
Entrepreneur and executive advisor who translates technical depth into boardroom-ready AI strategy and risk counsel for industry, government, and investors. naril+1
Work
Education
Projects
Writing
Predicting epistasis across proteins by structural logic
January 1, 2026Study combining high-throughput mutational scanning and machine learning (protein language model embeddings) to predict intragenic complementation and epistatic interactions in proteins; demonstrates a structural basis for predictability and model transferability.
A population-based evolutionary search approach to the multiple minima problem in de novo protein structure prediction
Presents evolutionary search algorithms (memetic approaches) to sample diverse low-energy conformations for de novo protein structure prediction and addresses multiple-minima challenges in conformational sampling.
Amarda Shehu — Substack essays and stories
Long-form essays and reflections on interdisciplinary AI research, education, and institutional strategy posted on her Substack publication.