Vignesh Ram Somnath

Bio

Hi! I am currently a 3rd year Ph.D. student at ETH Zurich, supervised by Andreas Krause and Kjell Jorner. The first 1.5 years of my PhD were spent at IBM Research Zurich, under Maria Rodriguez Martinez. Previously, I obtained my Masters degree in Computational Biology and Bioinformatics at ETH Zurich, and had the fortune of doing my Master Thesis with Regina Barzilay and Klavs Jensen at MIT.

My research aims to develop machine learning methods in scientific domains such as structural biology and chemistry. I enjoy learning the science behind the problem, and developing algorithms that strike a balance between being methodologically interesting yet practically applicable. I also am interested in finding connections between different fields that could provide a different perspective on the problem being investigated. My past experience revolves around the following domains:

Methods: Geometric Deep Learning, Generative Models, Riemannian Manifolds, (Stochastic) Optimal Control, Game Theory
Applications: Representation learning in proteins, Protein docking, Protein conformational changes and flexibility, Retrosynthesis

In addition, I also value a strong software engineering component to my research, trying to maintain readability and ease of extension in the code associated with my projects.

Publications

Most recent publications on Google Scholar.
^‡ indicates equal contribution.

Selected
All

DockGame: Cooperative Games for Multimeric Rigid Protein Docking

Vignesh Ram Somnath^‡, Pier Giuseppe Sessa^‡, Maria Rodriguez Martinez, Andreas Krause

arXiv

arXiv Code

Aligned Diffusion Schrödinger Bridges

Vignesh Ram Somnath^‡, Matteo Pariset^‡, Ya-Ping Hsieh, Maria Rodriguez Martinez, Andreas Krause, Charlotte Bunne

Conference on Uncertainty in Artificial Intelligence (UAI) 2023. Spotlight Presentation.

Proceedings arXiv Workshop Code

Isotropic Gaussian Processes on Spaces of Graphs

Viacheslav Borovitskiy^‡, Mohammad Reza Karimi^‡, Vignesh Ram Somnath^‡, Andreas Krause

Artificial Intelligence and Statistics (AISTATS) 2023

Proceedings arXiv Code

Multi-Scale Representation Learning on Proteins

Vignesh Ram Somnath^‡, Charlotte Bunne^‡, Andreas Krause

Neural Information Processing Systems (NeurIPS) 2021.

Proceedings arXiv Code

Learning Graph Models for Retrosynthesis Prediction

Vignesh Ram Somnath, Charlotte Bunne, Connor W. Coley, Andreas Krause, Regina Barzilay

Neural Information Processing Systems (NeurIPS) 2021.

Best Paper Award at ICML 2020 Graph Representation Learning & Beyond Workshop

Proceedings arXiv Code

Mixture-of-Experts Variational Autoencoder for Clustering and Generating from Similarity-Based Representations on Single Cell Data

Andreas Kopf, Vincent Fortuin, Vignesh Ram Somnath, Manfred Claassen

PLoS Computational Biology 2021

Proceedings arXiv Code

EquiReact: An equivariant neural network for chemical reactions

Puck van Gerwen, Ksenia R Briling, Charlotte Bunne, Vignesh Ram Somnath, Ruben Laplaza, Andreas Krause, Clemence Corminboeuf

arXiv

Paper

DockGame: Cooperative Games for Multimeric Rigid Protein Docking

Vignesh Ram Somnath^‡, Pier Giuseppe Sessa^‡, Maria Rodriguez Martinez, Andreas Krause

arXiv

Paper Code

Aligned Diffusion Schrödinger Bridges

Vignesh Ram Somnath^‡, Matteo Pariset^‡, Ya-Ping Hsieh, Maria Rodriguez Martinez, Andreas Krause, Charlotte Bunne

Conference on Uncertainty in Artificial Intelligence (UAI) 2023. Spotlight Presentation.

Paper Code

Isotropic Gaussian Processes on Spaces of Graphs

Viacheslav Borovitskiy^‡, Mohammad Reza Karimi^‡, Vignesh Ram Somnath^‡, Andreas Krause

Artificial Intelligence and Statistics (AISTATS) 2023

Paper Code

ChromFormer: A transformed-based model for 3D genome structure prediction

Henry Valeyre, Pushpak Pati, Federico Gossi, Vignesh Ram Somnath, Adriano Martinelli, Marianna Rapsomaniki

Learning Meaningful Representations of Life workshop, NeurIPS 2022

Paper

Multi-Scale Representation Learning on Proteins

Vignesh Ram Somnath^‡, Charlotte Bunne^‡, Andreas Krause

Neural Information Processing Systems (NeurIPS) 2021.

Paper Code

Learning Graph Models for Retrosynthesis Prediction

Vignesh Ram Somnath, Charlotte Bunne, Connor W. Coley, Andreas Krause, Regina Barzilay

Neural Information Processing Systems (NeurIPS) 2021.

Best Paper Award at ICML 2020 Graph Representation Learning & Beyond Workshop

Paper Code

Mixture-of-Experts Variational Autoencoder for Clustering and Generating from Similarity-Based Representations on Single Cell Data

Andreas Kopf, Vincent Fortuin, Vignesh Ram Somnath, Manfred Claassen

PLoS Computational Biology 2021

Paper Code

Vitæ

A PDF version of my CV can be found here.

Google DeepMind Aug 2023 - Nov 2023

Student Researcher
Computational Biology Team
ETH Zurich Sep 2021 - now

Ph.D. Student
Department of Computer Science (D-INFK)
ETH Zurich Jun 2020 - May 2021

Research Assistant
Department of Computer Science (D-INFK)
ETH Zurich Sep 2017 - May 2020

M.Sc. Student
Masters in Computational Biology and Bioinformatics
MIT Sep 2019 - Mar 2020

Visiting M.Sc. Student
CSAIL & ChemE
Google Summer of Code Summer 2019

Student Developer
DeepChem
IBM Research Zurich Spring 2019

Research Intern
Computational Systems Biology Group
Indian Institute of Technology Madras 2013 - 2017

B.Tech. Student
Chemical Engineering

Acknowledgements

This website uses the website design and template by Martin Saveski.