Intitulé du projet
GEVOLEX:
Learning and explaining the fine-scale evolutionary forces in
human genes
Learning and explaining the fine-scale evolutionary forces in
human genes
Nature du financement
Financement public - privé
État du projet
Refusé
Année de soumission
2026
Equipe(s) impliquée(s) dans le projet
StatInfOmics
Coordinateur·trice (nom et prénom)
Fanny Pouyet
Rôle de MaIAGE dans le projet
Responsable de Work Package
Nom(s) du(des) participant(s) - MaIAGE
G. KON KAM KING
Nom(s) du(des) partenaire(s) (nom, labo et localisation) - Hors MaIAGE
F. Pouyet - LISN - Université Paris-Saclay, L. Planche - AgroParisTech, A. F. Palazzo - University of Toronto Canada
Date de début du projet
Date de fin du projet
Résumé
Although broad genome-wide mutation patterns are known, fine-scale, position-dependent nucleotide composition along genes—especially GC content—remains poorly understood and inadequately modelled.
GC content reflects multiple interacting processes (mutation bias, gBGC, selection, biochemical constraints), but existing models and simulators either average at the gene level or lack mechanistic interpretability.
Initial models around transcription start sites reveal strong local GC variation and non-linear gene groups, yet fail to capture extreme GC values or intragenic heterogeneity.
GEVOLEX aims to build a position-aware, mechanistic framework that integrates genomic and epigenomic features to explain local base composition across full gene sequences.
Using non-homogeneous HMMs and evolutionary inference (ABC), GEVOLEX will quantify how mutation processes and epigenetic states jointly shape gene evolution over time.
GC content reflects multiple interacting processes (mutation bias, gBGC, selection, biochemical constraints), but existing models and simulators either average at the gene level or lack mechanistic interpretability.
Initial models around transcription start sites reveal strong local GC variation and non-linear gene groups, yet fail to capture extreme GC values or intragenic heterogeneity.
GEVOLEX aims to build a position-aware, mechanistic framework that integrates genomic and epigenomic features to explain local base composition across full gene sequences.
Using non-homogeneous HMMs and evolutionary inference (ABC), GEVOLEX will quantify how mutation processes and epigenetic states jointly shape gene evolution over time.
Champ thématique du contrat (MathNum)
Grand objectif concerné - principal - (MathNum)
Grand objectif concerné - secondaire - (MathNum)
Ce projet s'inscrit-il dans le périmètre scientifique du département MathNum ?