Statistical biology refers to statistical physics — the effort to explain the emergent properties of large composite systems (e.g. gas or liquids) from interactions between their constituents. To extend this framework to biological problems, we focus on the process that shapes living systems: Evolution.
Inference from evolutionarily related systems
How to describe biological systems with few relevant parameters ? We look for simplified representations by analyzing evolutionary conservation and coevolution in large datasets of related (homologous) biological sequences.
Protein sequencesApplied to proteins, this approach led us to propose that independent subsets of amino acids with no obvious structural signature can control different functional properties of a same protein domain (collaboration with the group of Rama Ranganathan).
- Evolution-based functional decomposition of proteins
- Elements of coevolution in biological sequences
- A model for the generation and transmission of variations in evolution
The value of information for populations in varying environments
Bacterial genomesApplied to genomes, the approach reveals a mode of gene regulation is bacteria that may be more fondamental than the regulation of operons by transcription factors (collaboration with the group of Ivan Junier).
Experimental molecular evolution
How to test theoretical predictions and go beyond natural evolution ? We analyze natural proteins and generate new ones by high-throughput screening and directed evolution.
Antibody evolution by phage displayWe use phage display to select and evolve antibodies with high affinity against a given molecular target.
Enzyme evolution by droplet microfluidicsWe use droplet microfluidics to screen and select enzymes.
Mathematical models of evolution
We are following inspiring examples from physics and develop abstract mathematical models to clarify fundamental concepts and draw connections between systems and problems.