Ghislain Durif

Research Engineer

CNRS

IMAG

University of Montpellier

Montpellier Bio-Stats

About me

Here you can find information about my engineering and research work in applied mathematics, including statistics, machine learning, data science, bio-statistics, and more.

Starting from February 2019, I am a research engineer at CNRS, based at the Institut Montpelliérain Alexander Grothendieck (IMAG), the Mathematics lab from the University of Montpellier, located in the south of France.
Among other activities, I co-manage the local network Montpellier Bio-Stats.

Previously (2017-2019), I was a research engineer at Inria (Grenoble, France) in THOTH team, under the supervision of Julien Mairal.

I did my PhD (2013-2016) at the Biometry and Evolutionary lab (LBBE), at the Claude Bernard Lyon 1 University (Villeurbanne, France), under the supervision of Franck Picard and Sophie Lambert-Lacroix.

Interests

Applied mathematics
Statistics
Machine Learning
Data Science
Bio-statistics
Software Development
High Performance Computing

Education

PhD in Statistics, 2016

Claude Bernard Lyon 1 University
Master degree in Applied Mathematics, 2013

Claude Bernard Lyon 1 University

Experience

February 2019 – Present

Montpellier, France

Research Engineer

CNRS

Providing support for researchers in applied mathematics regarding:

mathematical engnineering
software development
high performance computing

February 2017 – January 2019

Grenoble, France

Research Engineer

Inria

Engineering and research work in machine learning

October 2013 – January 2017

Lyon, France

PhD

LBBE

PhD in computational statistics applied to genomic data analysis

News

Previous posts

Launch meeting Montpellier Bio-stats

Last updated on Nov 15, 2019 1 min read

Development

R

Python

C/C++

Software

More details are available here.

KeOps

Python package for haplotype phasing and local ancestry inference

Loter

Python package for haplotype phasing and local ancestry inference

pCMF

Probabilistic count matrix factorization for single cell transcriptomic data analyses (dimension reduction, visualization).

SPAMS

Optimization toolbox for sparse estimation, implementing algorithms that solve machine learning and signal processing problems involving sparse regularizations.

plsgenomics

Supervised methods for dimension reduction in classification and regression framework (in particular PLS-based routines for genomic data analyses).

ckntf

Re-implementation of Convolutional Kernel Network (CKN) from Mairal (2016) in Python based on the TensorFlow framework

Featured Publications

More can be found here.

Probabilistic Count Matrix Factorization for Single Cell Expression Data Analysis

Motivation: The development of high throughput single-cell sequencing technologies now allows the investigation of the population …

Ghislain Durif, Laurent Modolo, Jeff E Mold, Sophie Lambert-Lacroix, Franck Picard

Code DOI OUP arXiv BiorXiv HAL

High Dimensional Classification with Combined Adaptive Sparse PLS and Logistic Regression

MotivationThe high dimensionality of genomic data calls for the development of specific classification methodologies, especially to …

Ghislain Durif, Laurent Modolo, Jakob Michaelsson, Jeff E. Mold, Sophie Lambert-Lacroix, Franck Picard

Code DOI OUP arXiv HAL

Featured Talks

Recent/upcoming and all past talks

Seamless Kernel Operations on GPU, with auto-differentiation and without memory overflows

Jan 7, 2020 14:00 — 15:00

Scidolyse Meeting, Lyon University

Slides

Multivariate analysis of single cell genomic expression data

Oct 4, 2018 13:00

Paris Single Cell Day (PASIC) 2018, Muséum National d’Histoire Naturelle de Paris

PASIC

Probabilistic Count Matrix Factorization for Single Cell Expression Data Analysis

Apr 23, 2018 15:20

Research in Computational Molecular Biology, 22nd Annual International Conference RECOMB 2018, Pierre et Marie Curie Campus, Sorbonne University

RECOMB

Count-based Probabilistic PCA for single-cell data analysis

Apr 3, 2017 11:25

ISCB NGS 2017 Structural Variation and Population Genomics, Barcelona Biomedical Research Park

NGS17

Count Matrix Factorization and Single Cell Data Analysis

Jun 30, 2016 15:00

Journées Ouvertes de Biologie Informatique et Mathématiques (JOBIM) 2016, ENS Lyon

JOBIM

Adaptive Sparse PLS for Logistic Regression

Jun 2, 2015 17:15

47èmes Journées de Statistique de la SFdS, Lille University

JDS

Ghislain Durif

Research Engineer

CNRS

IMAG

University of Montpellier

Montpellier Bio-Stats

About me

Interests

Education

Experience

Research Engineer

CNRS

Research Engineer

Inria

PhD

LBBE

News

Launch meeting Montpellier Bio-stats

Development

R

Python

C/C++

Software

Featured Publications

Probabilistic Count Matrix Factorization for Single Cell Expression Data Analysis

High Dimensional Classification with Combined Adaptive Sparse PLS and Logistic Regression

Featured Talks

Seamless Kernel Operations on GPU, with auto-differentiation and without memory overflows

Multivariate analysis of single cell genomic expression data

Probabilistic Count Matrix Factorization for Single Cell Expression Data Analysis

Count-based Probabilistic PCA for single-cell data analysis

Count Matrix Factorization and Single Cell Data Analysis

Adaptive Sparse PLS for Logistic Regression

Tags

Contact