Sarah Gurev

Sarah Gurev

Postdoctoral Fellow

Harvard Medical School

I am an postdoctoral fellow in Debora Marks lab at Harvard. I am interested in deep generative models of proteins, especially modeling viral and antibody sequences and structures. My research develops models to learn viral antibody escape from fitness models and biophysical and structural information. These models can be used to identify high escape variants and to design and test future-proof vaccines. I recently completed my PhD in Electrical Engineering and Computer Science at MIT.

Interests
  • Machine learning for biology
  • Computational vaccine design
  • Viral evolution
  • Protein engineering
Education
  • PhD in Electrical Engineering and Computer Science, 2025

    Massachusetts Institute of Technology

  • MS in Electrical Engineering and Computer Science, 2023

    Massachusetts Institute of Technology

  • BS in Computer Science, 2020

    Stanford University

Publications

Quickly discover relevant content by filtering publications.
(2025). Computationally designed proteins mimic antibody immune evasion in viral evolution. Immunity.

PDF Cite Code

(2024). Alignment-based and protein foundation models for viral evolution, vaccines and vectors. Neurips Workshop on AI for New Drug Modalities.

PDF Cite

(2024). Future-proof vaccine design with a generative model of antibody cross-reactivity. ICLR Workshop on Generative and Experimental Perspectives for Biomolecular Design.

PDF Cite

(2023). Learning from prepandemic data to forecast viral escape. Nature.

PDF Cite Code Dataset

(2023). Removing bias in sequence models of protein fitness. BioRxiv.

PDF Cite

Contact