Séminaire de Probabilités et Statistique :
Le 15 mars 2021 à 13:45 - UM - Bât 09 - Salle de conférence (1er étage)
Présentée par Varoquaux Nelle - TIMC - Université Grenoble Alpes
Functional clustering of time-course gene expression data
Understanding the dynamics of gene expression is a fundamental question of
biology. However, analysing time-course transcriptomic data raises unique
challenging statistical and computational questions, requiring the development
of novel methods. Here, we are interested the classification of genes into
different types of temporal changes. To do so, we propose a novel functional
clustering of time-course gene expression data that generalizes a
mixture-model clustering by Ma et al [2006]. Our clustering strategy estimates
a functional splines model for the mean of the cluster, in order to cluster
temporal patterns of features independent of scale, including
identifying genes with inverted patterns of expressions. We also present
results both on simulated data and real data from a time-course transcriptomic
experiment aiming to better understand the dynamical response of plants under
drought stress.
I will present results both on simulated data and real data from a time-course
transcriptomic experiment aiming to better understand the dynamical response
of plants under drought stress.
This work is done in collaboration with Stephanie DeGraaf and Elizabeth Purdom.
WEBINAIRE ouvert à toutes et tous : https://umontpellier-fr.zoom.us/j/85813807839