Title: Bayesian lasso time-course data clustering
Authors: Alejandro Murua - University of Montreal (Canada) [presenting]
Folly Adjogou - Universite de Montreal (Canada)
Wolfgang Raffelsberger - Institute de Genetique et de Biologie Moleculaire et Cellulaire (France)
Abstract: A flexible model is developed for the analysis and clustering of time-course or longitudinal data. The model combines functional analysis and model-based clustering. The functional framework is used to model time-course data. Principal functional components are described by score coefficients which embed the curves in a much lower-dimensional space. Model based clustering is performed on the score space, thus avoiding the curse of dimensionality in the curves' space. The model is embedded into a Bayesian framework. We first develop an approximation of the marginal log-likelihood MLL that allows us to perform a MLL based model selection. We then developed a Bayesian version of the lasso and elastic-net penalty in order to render the model selection step more efficient. The number of clusters as well as the dimension of the score space are determined via this Bayesian lasso penalty model. We show some applications to the analysis of gene-expression data.