FrancaisEnglish (United Kingdom)
 

Responsable de publication :
Olivier Martin

Responsable éditoriale :
Rozenn Le Guyader

Administrateur de l'infoservice :
Thierry Balliau

SémIDEEV à l’UMR

Génétique Quantitative et Évolution – Le Moulon

Ferme du Moulon – Salle de conférences

Mardi 9 mai 2017

10h00

Mélina GALLOPIN
I2BC – Équipe Olivier Lespinet - Orsay

invitée par Élodie Marchadier
Cette adresse email est protégée contre les robots des spammeurs, vous devez activer Javascript pour la voir.

" Network inference from gene expression data "

Abstract :

Gaussian graphical models are widely utilized to infer networks from gene expression data. However, inferring the graph is difficult when the sample size is small compared to the number of genes. To reduce the number of parameters to estimate in the model, we propose a non-asymptotic model selection procedure supported by theoretical guarantees. The covariance matrix of the model is approximated by a block-diagonal matrix. The structure of this matrix is detected by thresholding the sample covariance matrix, where the threshold is selected using the slope heuristic. Based on the block-diagonal structure of the covariance matrix, the estimation problem is divided into several independent problems: subsequently, the network of dependencies between variables is inferred using the graphical lasso algorithm in each block. On a real RNA-seq gene expression dataset with a limited sample size, the dimension reduction allows attention to be objectively focused on interactions among smaller subsets of genes, leading to a more parsimonious and interpretable modular network.