Hugo Talibart’s defense on “Comparison of homologous protein sequences using direct coupling information by pairwise Potts model alignments” will take place on Wednesday 24th February 2021 at 14:30 (UTC+1).
The defense will be broadcasted live on youtube at: https://youtu.be/NtBNWGjifv8
Committe:
-
Sean EDDY, Professor at Harvard University, Cambridge, USA (Rapporteur)
-
Martin WEIGT, Professor at Sorbonne Université, Paris (Rapporteur)
-
Guillaume GRAVIER, Senior researcher CNRS, Rennes
-
Juliette MARTIN, Researcher CNRS, Lyon
-
Thomas SCHIEX, Senior researcher INRAE, Toulouse
-
Jacques NICOLAS, Senior researcher Inria, Rennes (thesis Director)
-
François COSTE, Researcher Inria, Rennes (thesis supervizor)
Abstract:
To assign structural and functional annotations to the ever increasing amount of sequenced proteins, the main approach relies on sequence-based homology search methods based on significant alignments of query sequences to annotated proteins or protein families. While powerful, existing approaches do not take coevolution between residues into account. Taking advantage of recent advances in the field of contact prediction, in this thesis we propose to represent proteins by Potts models, which model direct couplings between positions in addition to positional composition, and to compare proteins by aligning these models. This novel application of Potts models raised further requirements for their construction, and we identified several key points towards building more comparable Potts models, towards an ideal of canonicity. Due to non-local dependencies, the problem of aligning Potts models is NP-hard. Here, we introduced a method based on an Integer Linear Programming formulation of the problem which can be optimally solved in tractable time. Our first results suggest that taking pairwise couplings into account can improve the alignment of remote homologs and could thus improve remote homology detection.
