PhD Student until October 2019.

UMR518 AgroParisTech/INRA 
Team Stat&Genome 
16 rue Claude Bernard 
75 231 Paris Cedex 05 
Tel : +33 (0)1 44 08 72 09 
marie.perrot-dockes@agroparistech.fr

Research

Thesis : Regularized methods to study multivariate data in high dimensional settings : theory and applications.

Advisors : Céline Lévy-Leduc, Jullien Chiquet, Larue Sansonnet

Keywords : Regularized methods, multivariate data, high dimension

Abstracct :  In this PhD thesis we study general linear model (multivariate linear model) in high dimensional settings. We propose a novel variable selection approach in the framework of multivariate linear models taking into account the dependence that may exist between the responses. It consists in estimating beforehand the co-variance matrix of the responses and to plug this estimator in a Lasso criterion, in order to obtain a sparse estimator of the coefficient matrix. The properties of our approach are investigated both from a theoretical and a numerical point of view. More precisely, we give  general conditions that the estimators of the covariance matrix and its inverse have to satisfy in order to recover the positions of the zero and non-zero entries of the coefficient matrix when the number of responses is not fixed and can tend to infinity. We also propose novel, efficient and fully data-driven approaches for estimating Toeplitz and large block structured sparse covariance matrices in the case where the number of variables is much larger than the number of samples without limiting ourselves to block diagonal matrices. These approaches are applied to different biological issues in metabolomics, in proteomics and in immunology.

Teaching

     - Tutorial of Inferential Statistic to first year student of AgroParisTech (2016- 2019)

     - Introduction to R langage to second year students of AgroParisTech  (2016- 2019)

     - Course of probability and statistic of master 2 student at Polytech

Publications

Article publiés

•  Journal of Multivariate Analysis M. Perrot-Dockès, C. Lévy-Leduc, L. Sansonnet, J. Chiquet, "Variable selection in multivariate linear models with high-dimensional covariance matrix estimation" , 166 :78 – 97, 2018.
• Statistical Applications in Genetics and Molecular Biology M. Perrot-Dockès, C. Lévy-Leduc, J. Chiquet, L. Sansonnet, M. Brégère, M.-P. Étienne, S. Robin, G. Genta-Jouve "A variable selection approach in the multivariate linear model : An application to LC-MS metabolomics data" 17(5), 2018.
• Cell M. Grandclaudon*, M. Perrot-Dockès*, C. Trichot,* O. Mostafa-Abouzid, W. Abou-Jaoudé, F. Berger, P. Hupé, D. Thieffry, L. Sansonnet, J. Chiquet, C. Lévy-Leduc, V. Soumelis A quantitative ultivariate model of human dendritic cell-Thelper cell communication. , 2019 * : ces auteurs ont contribué de manière égale à cette publication

Article soumis

• M. Perrot-Dockès, C. Lévy-Leduc "Estimation of large block structured covariance matrices : Application to “multi-omic” approaches to study seed quality"I

Chapitre de livre (à paraître prochainement)

• Willey M. Perrot-Dockès, C. Lévy-Leduc "Estimation of large block structured covariance matrices : Application to “multi-omic” approaches to study seed quality"I

Package R

• MultiVarSel disponible sur le CRAN.
• BlockCov disponible sur le CRAN.
• VariSel disponible sur github.