Pranav M. Murugan

I am a Senior Machine Learning Research Scientist at Genesis Therapeutics, working on novel generative and predictive machine learning methods for small-molecule drug discovery. I am currently leading a team to build foundation diffusion models to predict protein-ligand structures, iterating on architecture, inference protocol, and kernel optimizations. Previously, I spearheaded Genesis’s efforts in building rotationally-equivariant neural networks for state-of-the-art force fields for molecular simulation.

Prior to joining Genesis, I received my M.Eng. and S.B. from MIT with a double major in Physics and in Electrical Engineering and Computer Science. I primarily worked in the lab of Arup Chakraborty, where I studied computational models of the human immune system. I combined tools from statistical mechanics and information theory with massively parallel CUDA-accelerated stochastic simulation to design better vaccines and immunization protocols. Over the course of my undergraduate degree, I also worked as a machine learning research intern at Genesis Therapeutics, and in the labs of Mark Harnett (MIT) and Wesley Tansey (MSKCC). My work in these labs included deep learning classification of time-series neural signals and developing novel clustering-based methods for causal inference.

I was a member of the 2017 and 2018 US Physics Olympiad Team which was quite formative for my interest in physics; consider donating or getting involved if you’re interested!

In my free time, I dance and climb mountains with my friends.

Feel free to take a look around! I have a selected list of publications and projects in a wide range of topics highlighted in the respective sections. Currently, I’m best contacted at mpranav81 [at] gmail [dot] com.

selected publications

PhysRevE

Minimal framework for optimizing vaccination protocols targeting highly mutable pathogens

Saeed Mahdisoltani, Pranav Murugan, Arup K. Chakraborty, and 1 more author

Physical Review E Dec 2024
PNAS

Optimizing immunization protocols to elicit broadly neutralizing antibodies

Kayla G. Sprenger, Joy E. Louveau, Pranav M. Murugan, and 1 more author

Proceedings of the National Academy of Sciences Dec 2020

Abs HTML

Natural infections and vaccination with a pathogen typically stimulate the production of potent antibodies specific for the pathogen through a Darwinian evolutionary process known as affinity maturation. Such antibodies provide protection against reinfection by the same strain of a pathogen. A highly mutable virus, like HIV or influenza, evades recognition by these strain-specific antibodies via the emergence of new mutant strains. A vaccine that elicits antibodies that can bind to many diverse strains of the virus—known as broadly neutralizing antibodies (bnAbs)—could protect against highly mutable pathogens. Despite much work, the mechanisms by which bnAbs emerge remain uncertain. Using a computational model of affinity maturation, we studied a wide variety of vaccination strategies. Our results suggest that an effective strategy to maximize bnAb evolution is through a sequential immunization protocol, wherein each new immunization optimally increases the pressure on the immune system to target conserved antigenic sites, thus conferring breadth. We describe the mechanisms underlying why sequentially driving the immune system increasingly further from steady state, in an optimal fashion, is effective. The optimal protocol allows many evolving B cells to become bnAbs via diverse evolutionary paths.