Faculty Seminar Anne-Florence Bitbol: Modeling and harnessing biological evolution across scales
Abstract:
My group focuses on modeling evolution and extracting information from evolutionary data, from molecules to microbial populations.
Protein sequences are shaped by functional optimization and historical contingency. Functional constraints induce correlations in amino acid usage across sites in multiple sequence alignments, but correlations also arise from pure phylogeny. We showed that phylogenetic correlations are a double-edged sword: while they hinder the inference of structural contacts and functional amino-acid groups from sequences, they facilitate the inference of interacting partners. Advances in machine learning, particularly protein language models trained on multiple sequence alignments, offer exciting possibilities for coevolution-based inference. We exploited these models to predict interacting partners and generate new sequences from protein families.
Natural microbial populations often have complex spatial structures, which can affect the predictability of evolution. We developed a general model of populations on graphs, and determined the impact of such spatial structures on mutant fixation and on fitness landscape exploration. We investigated applications to the evolution of drug resistance and to the evolution of gut bacteria.
Bio:
Anne-Florence Bitbol studied physics at ENS Lyon and Université Paris-Diderot in France, and obtained her PhD in 2012, advised by Jean-Baptiste Fournier. As an HFSP postdoctoral fellow, she then joined the Biophysics Theory group at Princeton University (USA), led by William Bialek, Curtis Callan and Ned Wingreen. In 2016, she became an independent CNRS researcher at Sorbonne Université in Paris, before joining EPFL as a tenure-track assistant professor in early 2020. She is broadly interested in understanding biological phenomena through physical concepts and mathematical and computational tools, and she leads the Laboratory of Computational Biology and Theoretical Biophysics. She is investigating the impacts of optimization and historical contingency in biological systems, from the molecular to the population scale. She holds an ERC Starting Grant and has received the Early Career Scientist Prize in Biological Physics from the International Union of Pure and Applied Physics (IUPAP) and the EPFL Life Sciences Engineering Teaching Award.
This seminar is part of the evaluation of Prof Bitbol for the promotion to Associate Professor.
My group focuses on modeling evolution and extracting information from evolutionary data, from molecules to microbial populations.
Protein sequences are shaped by functional optimization and historical contingency. Functional constraints induce correlations in amino acid usage across sites in multiple sequence alignments, but correlations also arise from pure phylogeny. We showed that phylogenetic correlations are a double-edged sword: while they hinder the inference of structural contacts and functional amino-acid groups from sequences, they facilitate the inference of interacting partners. Advances in machine learning, particularly protein language models trained on multiple sequence alignments, offer exciting possibilities for coevolution-based inference. We exploited these models to predict interacting partners and generate new sequences from protein families.
Natural microbial populations often have complex spatial structures, which can affect the predictability of evolution. We developed a general model of populations on graphs, and determined the impact of such spatial structures on mutant fixation and on fitness landscape exploration. We investigated applications to the evolution of drug resistance and to the evolution of gut bacteria.
Bio:
Anne-Florence Bitbol studied physics at ENS Lyon and Université Paris-Diderot in France, and obtained her PhD in 2012, advised by Jean-Baptiste Fournier. As an HFSP postdoctoral fellow, she then joined the Biophysics Theory group at Princeton University (USA), led by William Bialek, Curtis Callan and Ned Wingreen. In 2016, she became an independent CNRS researcher at Sorbonne Université in Paris, before joining EPFL as a tenure-track assistant professor in early 2020. She is broadly interested in understanding biological phenomena through physical concepts and mathematical and computational tools, and she leads the Laboratory of Computational Biology and Theoretical Biophysics. She is investigating the impacts of optimization and historical contingency in biological systems, from the molecular to the population scale. She holds an ERC Starting Grant and has received the Early Career Scientist Prize in Biological Physics from the International Union of Pure and Applied Physics (IUPAP) and the EPFL Life Sciences Engineering Teaching Award.
This seminar is part of the evaluation of Prof Bitbol for the promotion to Associate Professor.
Practical information
- Informed public
- Free
Organizer
- Deanship SV
Contact
- Manuelle Mary