Software

I am the developer, maintainer or contributor of 16 R packages. Several have been published in Journal of Statistical Software or The R Journal, and are widely used in epidemiology, public health and bioinformatics.

Available on CRAN

• combss — Continuous Optimisation Towards Best Subset Selection. Maintainer, with Anant Mathur and Sarat Moka.
• sgPLS — sparse group partial least squares regression models. Maintainer, with Pierre Lafaye de Micheaux.
• MBSGS — multivariate Bayesian sparse group selection with spike and slab. Maintainer, with Matthew Sutton.
• GCPBayes — cross-phenotype genetic association studies with group-level Bayesian meta-analysis. With Taban Baghfalaki, Pierre-Emmanuel Sugier, Yazdan Asgari and Thérèse Truong.
• CEoptim — the cross-entropy method for optimisation problems. Maintainer, with Tim Benham, Qibin Duan and Dirk P. Kroese.
• lcmm — estimation of latent class mixed models and joint latent class mixed models. Developer, with Cécile Proust-Lima.
• ClustOfVar — cluster analysis of a set of variables, quantitative, qualitative or mixed. Developer, with Marie Chavent, Vanessa Kuentz and Jérôme Saracco.
• PCAmixdata — multivariate analysis of mixed data. Contributor, with Marie Chavent, Vanessa Kuentz, Amaury Labenne and Jérôme Saracco.
• ConvergenceConcepts — check and visualise convergence of a sequence of random variables in law, in probability, in r-th mean and almost surely. Developer, with Pierre Lafaye de Micheaux.
• CPMCGLM — p-value correction after multiple coding in generalized linear models. Contributor, with Jérémie Riou.
• rPowerSampleSize — sample size determination with multiple correlated endpoints. Contributor, with Jérémie Riou and Pierre Lafaye de Micheaux.

Available elsewhere

• mixOmics — statistical integrative techniques for highly dimensional data sets. Past Contributor from 2011 to 2012, developed multilevel approaches . Now distributed through Bioconductor.
• COMBSS — best subset selection via continuous optimisation. Maintainer of the code, on GitHub.
• bigsgPLS and sgsPLS — big-data and subgroup extensions of sparse group PLS, on GitHub.
• SGMT — Sparse Group Multi-Task methods for pleiotropy, on GitHub.
• CPMTL — Fortran program to correct the p-value after multiple coding of an explanatory variable in logistic regression. Developer.

Archived on CRAN

Still installable from the CRAN archive, and the methods remain described in the associated publications.

• R2GUESS — GPU-based Bayesian variable selection regression of multiple correlated phenotypes. Maintainer, with Marc Chadeau-Hyam, Gianluca Campanella and Sylvia Richardson.
• PIGE — pathway interaction gene environment, an extension of the ARTP method. Maintainer, with Thérèse Truong.
• edrGraphicalTools — tools for dimension reduction methods. Maintainer, with Jérôme Saracco and Raphaël Coudret.

Programming

R, Python, C, Fortran 77/90 and Delphi. Advanced use of R, S-Plus, SPSS, SAS and Minitab, and of modern machine learning and deep learning frameworks in Python.

All repositories on GitHub