Description
Date depot: 4 avril 2019
Titre: New computational approaches to study the effect of genetic mutations on the topology of protein-protein interaction networks
Directrice de thèse:
Alessandra CARBONE (LCQB)
Directrice de thèse:
Elodie LAINE (LCQB)
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Non defini
Resumé:
In this thesis, we wish to develop the necessary computational framework to reconstruct the annotated PPI network (more than 10000 nodes and hundreds thousands of interactions) for the human reference genome and annotated PPI (sub)networks for individual genomes. This means to provide an ab initio method that reconstructs interaction networks and determine how mutations may affect the PPI topology. It will be based on the development of a deep learning approach discriminating protein docking conformations and protein partners’ conformations, and on a novel docking algorithm especially designed for the protein partners discrimination problem. These approaches will be built on fundamental results developed in the last years in the team. Namely, the dataset of protein-protein interactions generated from Complete Cross-Docking (CC-D) (Sacquin-Mora et al 2009, Lopes et al 2013, Laine & Carbone 2015), and the algorithmic tools LISA (Raucci et al Structure2018), CIPS (Nadalin & Carbone Bioinformatics 2018), JET2 (Laine & Carbone PLoS Computational Biology2016), which arethe state-of-the-art in the field today. The reduction of the computational cost of the docking strategy (obtained by integrating the tools cited above), the deep learning model especially designed to discriminate conformational docking poses and the improved computational facilities available will allow us to realize a breakthrough on CC-D experiments and reach the size of the human proteome (the order of 10000 protein structures).
Doctorant.e: Mohseni Behbahani Yasser