Mona Singh and Panel Discussion

Mona Singh and Panel Discussion with Princeton Postdocs and Schmidt Center Postdoctoral Fellows

Monday, November 4, 2024

Broad Institute - Yellowstone - 415 Main St.

Watch virtually via YouTube Livestream: broad.io/ewsc

‍

Panel Discussion: Foundation Models in Genomics

2:45-3:30 pm

In this panel discussion, Schmidt Center and Princeton postdocs will explore the significance of foundation models in genomics. Key topics include:

• Defining foundation models and their impact on genomics.
• Benchmarking insights and challenges from recent studies.
• Architectural innovations tailored for biological applications.

‍

Colloquium Talk: Mona Singh on "Protein Language Models: Their Power, Limitations, and Future Directions."

4:00-5:00 pm  (refreshments at 3:30 pm)

Abstract: Protein language models are one of the most exciting recent advances in modeling protein sequences.  Among other applications, they have proven to be powerful in predicting structures, interactions and functions, and in assessing the impact of amino acid mutations.  I will discuss how my group has is using protein language models in a variety of settings, including alignment and functional site predictions, and how we are “stress testing” their capabilities. I will conclude by discussing promising avenues for future work.

Mona Singh is a Professor of Computer Science and the Lewis Sigler Institute for Integrative Genomics. She works broadly in computational molecular biology, as well as its interface with machine learning and algorithms. Much of her work is on developing algorithms to decode genomes at the level of proteins and she is especially interested in developing data-driven methods for predicting and characterizing protein sequences, functions, interactions and networks, both in healthy and disease contexts.