Modèles Structurés et Méthodes Algorithmiques pour l'analyse de Données Complexes et de Grande Dimension : Applications en Santé

Persyvact2 is a follow-up to the Persyvact exploratory project. Our collaborative research team aims at developing cutting edge data science methodologies to analyse large biomedical data. Persyvact2 consists of about 20 researchers from GIPSAlab, LJK and TIMC-IMAG. Coming from different fields related to data science (statistics, machine learning, image and signal processing), members of Persyvact2 will analyze biomedical data generated from neuroscience, genomics, and clinical trial research. The key structures of biomedical data that Persyvact 2 will exploit consist of graph structure, repeated experiments and their intrinsic lower dimensional representation.

The aim of Persyvact2 is to perform collaborative research and to bring together researchers of different scientific fields interested by data science. Persyvact2 intends to organize scientific events and an international workshop during its lifetime. Persyvact2 seeks to enhance the international visibility of data science in Grenoble.

Keywords: Computational and theoretical statistics; multimodal and heterogeneous data; image and signal processing; spatial analysis; graphical models; mixed models; learning; structure extraction; (non)super- vised classification; detection; point processes; large scale optimization; regularisation (convex or not, differentiable or not); Bayesian analysis; multiscale analysis; markov models; latent variables models; dimension reduction; sparseness; model selection; robustness; computational neuroscience; genetics; medical imaging. 

Seminars and meetings

• Cancer heterogeneity data challenge. December 10-14 2018

• Summer school signal processing on graph. September 10-14 2018

• Graduate training program Data Sciences for Geosciences. January 15-19 2018

• Epigenetic & High-Dimension Mediation Data Challenge. June 7-9 2017

• Journée "Modélisation en biologie et statistique de données biomédicales". November 2 2015

​Publications

• Choiruddin A, J.-F Coeurjolly, F. Letué (2018) Convex and non-convex regularization methods for spatial point processes intensity estimation. Electronic Journal of Statistics
• Prive F, H Aschard H, A Ziyatdinov, MGB Blum (2018) Efficient management and analysis of large-scale genome-wide data with two R packages: bigstatsr and bigsnpr. Bioinformatics 34:2781-2787
• Capblancq T, K Luu, MGB Blum, E Bazin (2018) How to make use of ordination methods to identify local adaptation: a comparison of genome scans based on PCA and RDA. Molecular Ecology Resources
• Dias-Alves T, J Mairal, MGB Blum (2018) Loter: A software package to infer local ancestry for a wide range of species. Molecular Biology and Evolution, msy 126
• Bacher R, Meillier C, Chatelain F, Michel O (2017) Robust control of varying weak hyperspectral target detection with sparse non-negative representation. IEEE Transactions on Signal Processing 65:3538-3550
• Cubry P, Vigouroux Y, François O (2017) The empirical distribution of singletons for geographic samples of DNA sequences. Frontiers in Genetic 8
• Gabriel Zebadua A, P-O Amblard, E Moisan, OJJ  Michel (2017) Compressed and Quantized Correlation Estimators. IEEE Transactions on Signal Processing 65:56-68
• Luu K, Bazin E, Blum MGB (2017) pcadapt: an R package to perform genome scans for selection based on principal component analysis. Molecular Ecology Resources 1:67-77
• Le Bihan N, F Chatelain, J Manton (2016) Isotropic Multiple Scattering Processes on Hyperspheres. IEEE Transactions on Information Theory 62:5740-5752
• Grenier E, Helbert C, Louvet V, Samson A, Vigneaux P (2016) Population parametrization of costly black box models using iterations between SAEM algorithm and kriging. Computational and Applied Mathematics.
• Delattre M, Genon-Catalot V, Samson A (2016) Mixtures of stochastic differential equations with  random effects: application to  data clustering. Journal of Statistical Planning and Inference.
• Harlé F, F Chatelain, C Gouy-Pailler, S Achard. Bayesian Model for Multiple Change-Points Detection in Multivariate Time Series (2016) Signal Processing, IEEE Transactions Signal Processing 
• Ollier E, Samson A, Delavenne X, Viallon V (2016) A SAEM Algorithm for Fused Lasso Penalized Non Linear Mixed Effect Models: Application to Group Comparison in Pharmacokinetic. Computational Statistics and Data Analysis. 96:207-221.
• Martins H, Caye K, Luu K, Blum MGB, François O (2016) Identifying outlier loci in admixed and in continuous populations using ancestral population differentiation statistics. Molecular Ecology 25:5029–5042
• Duforet-Frebourg N, Luu K, Bazin E, Blum MGB (2016). Detecting Genomic Signatures of Natural Selection with Principal Component Analysis: Application to the 1000 Genomes Data. Molecular Biology and Evolution. 33:1082-1093 
• Termenon M, C Delon-Martin, A Jaillard, Achard S (2016). Reliability of graph analysis of resting state fMRI using test-retest dataset from the human connectome project. Neuroimage 142: 172–187.
• Termenon M, S Achard, A Jaillard, Delon-Martin C (2016). The ”hub disruption index”, a reliable index sensitive to the brain networks reorganization. a study of the contra- lesional hemisphere in stroke. Frontiers in Computational Neuroscience 10.

Members

• Sophie Achard Gipsa-lab
• Pierre-Olivier Amblard Gipsa-lab
• Hacheme Ayasso Gipsa-lab
• Caroline Bazzoli LJK
• Michael Blum TIMC-IMAG
• Jean Marc Brossier Gipsa-lab
• Florent Chatelain Gipsa-lab
• Laurent Condat Gipsa-lab
• Michel Desvignes Gipsa-lab
• Rémy Drouilhet LJK
• Florence Forbes LJK
• Olivier François TIMC-IMAG
• Stéphane Girard LJK
• Sophie Lambert TIMC-IMAG
• Adeline Leclercq-Samson LJK
• Frédérique Letué LJK
• Julien Mairal LJK
• Marie-José Martinez LJK
• Olivier Michel Gipsa-lab
• Nicolas Thierry-Mieg TIMC-IMAG
• Laurent Zwald LJK

Coordinators

• Michael Blum TIMC-IMAG
• Adeline Leclercq-Samson LJK
• Pierre-Olivier Amblard Gipsa-lab