Mots-Clés

symbolic AI, computational biology, drug discovery, logic programming, Boolean networks

Description

Computational modeling of biased signaling in G protein-coupled receptors

The complexity of pharmacological efficacy at G protein-Coupled Receptors opens the possibility of selectively controlling their signaling pathways, i.e., the ability of a ligand to activate certain signal transduction pathways selectively compared to the native ligand acting at the same receptor. This potentially offers significant opportunities in various fields of biology, including mammalian reproduction. Kinetic experiments, measuring the activity of several downstream effectors of a receptor after ligand binding over time, are now widely available. The numerous cross-talks, redundancies, and regulatory mechanisms within signaling cascades hinder the use of a direct statistical approach to understand and infer ligand efficacy.

The objectives of this PhD thesis are to : 1) develop a tool using logic programming to infer models, representing ligand efficacy, 2) develop or extend an algorithm to sample networks, providing a comprehensive view of the solution space, 3) identify a way to characterize different dynamical behaviors among a set of networks, improving the understanding of pathway variability, and 4) develop an algorithm to select informative conditions, reducing the variability among inferred networks.

The PhD student will implement this methodology on several practical scenarios on interest in the BIOS team, in particular to gain knowledge on the gonadotropins signaling networks and to characterize the pharmacological efficacy of innovative ligands currently developed in the team.

Profil et compétences recherchées

* Background in Computer Science

* Python, C++, logic programming

* Machine learning

* Linux, Latex, GitHub

* Ability to work as independent as well as a part of a team

* Creative and good communication skills

* English proficiency