I spread most of my code as `R`

packages, with underlying routines in
`C/C/++`

relying on the great
Rcpp package
and the outstanding armadillo library.

@soft{PLNModels, author = {Chiquet, J. and Mariadassou, M. and Robin, S.}, title = {PLNmodels: Poisson lognormal models}, howpublished = {\url{https://github.com/jchiquet/PLNmodels}}, download = {https://github.com/jchiquet/PLNmodels}, year = {2017} }

- Chiquet J, Dervieux V, Rigaill G:
**aricode:a package for efficient computations of standard clustering comparison measures**, 2017

download@soft{aricode, author = {Chiquet, J. and Dervieux, V. and Rigaill, G.}, title = {aricode:a package for efficient computations of standard clustering comparison measures}, howpublished = {\url{https://github.com/jchiquet/aricode}}, download = {https://github.com/jchiquet/aricode}, year = {2017} }

- Brault V, Chiquet J:
**blockseg: two Dimensional Change-Points Detection**, 2016

Segments a matrix in blocks with constant values. The underlying algorithm is a Lars-type algorithm where all the matrix operation can be computed explicitly.

download@soft{blockseg, author = {Brault, V. and Chiquet, J.}, title = {blockseg: two Dimensional Change-Points Detection}, howpublished = {\url{https://CRAN.R-project.org/package=blockseg}}, download = {https://CRAN.R-project.org/package=blockseg}, year = {2016}, note = {Segments a matrix in blocks with constant values. The underlying algorithm is a Lars-type algorithm where all the matrix operation can be computed explicitly.} }

- Bouveyron C, Chiquet J, Latouche P, Mattei P-A:
**spinyReg: Sparse Generative Model and Its EM Algorithm**, 2015

Implements a generative model that uses a spike-and-slab like prior distribution obtained by multiplying a deterministic binary vector. Such a model allows an EM algorithm, optimizing a type-II log-likelihood.

download@soft{spinyreg, author = {Bouveyron, C. and Chiquet, J. and Latouche, P. and Mattei, P.-A.}, title = {spinyReg: Sparse Generative Model and Its EM Algorithm}, howpublished = {\url{https://cran.r-project.org/web/packages/spinyReg/}}, download = {https://cran.r-project.org/web/packages/spinyReg/}, year = {2015}, note = {Implements a generative model that uses a spike-and-slab like prior distribution obtained by multiplying a deterministic binary vector. Such a model allows an EM algorithm, optimizing a type-II log-likelihood.} }

- Chiquet J:
**SPRING: Structured selection of Primordial Relationships IN the General linear model**, 2014

This package fits multivariate regression models using sparse conditional Gaussian graphical modeling with Laplacian regularization.

download@soft{spring, author = {Chiquet, J.}, title = {SPRING: Structured selection of Primordial Relationships IN the General linear model}, howpublished = {\url{https://r-forge.r-project.org/projects/spring-pkg/}}, download = {https://r-forge.r-project.org/projects/spring-pkg/}, year = {2014}, note = {This package fits multivariate regression models using sparse conditional Gaussian graphical modeling with Laplacian regularization.} }

- Gutierrez P, Rigaill G, Chiquet J:
**Fused-Anova**, 2013

This package adjusts a penalized ANOVA model with Fusion penalities, i.e. a sum of weighted l1-norm on the difference of each coefficient. The fitting procedure is accompanied by a highly efficient cross-validation method.

download@soft{fusedanova, author = {Gutierrez, P. and Rigaill, G. and Chiquet, J.}, title = {Fused-Anova}, howpublished = {\url{https://r-forge.r-project.org/projects/fusedanova/}}, download = {https://r-forge.r-project.org/projects/fusedanova/}, year = {2013}, note = {This package adjusts a penalized ANOVA model with Fusion penalities, i.e. a sum of weighted l1-norm on the difference of each coefficient. The fitting procedure is accompanied by a highly efficient cross-validation method.} }

- Chiquet J:
**Quadrupen: Sparsity by Worst-Case Quadratic Penalties**, 2012

This package fits classical sparse regression models with efficient active set algorithms by solving quadratic problems. It also provides a few methods for model selection purposes (cross-validation, stability selection).

download@soft{quadrupen, author = {Chiquet, J.}, title = {Quadrupen: Sparsity by Worst-Case Quadratic Penalties}, howpublished = {\url{http://cran.r-project.org/web/packages/quadrupen/}}, download = {http://cran.r-project.org/web/packages/quadrupen/}, year = {2012}, note = {This package fits classical sparse regression models with efficient active set algorithms by solving quadratic problems. It also provides a few methods for model selection purposes (cross-validation, stability selection).} }

- Chiquet J:
**Scoop: Sparse Cooperative Regression**, 2011

This R package fits coop-Lasso, group-Lasso and tree-group Lasso variants for linear regression and logistic regression. The cooperative-Lasso (in short, coop-Lasso) may be viewed as a modification of the group-Lasso penalty that promotes sign coherence and that allows zeros within groups.

download web page@soft{scoop, author = {Chiquet, J.}, title = {Scoop: Sparse Cooperative Regression}, howpublished = {\url{http://julien.cremeriefamily.info/scoop}}, download = {http://julien.cremeriefamily.info/scoop}, web = {http://julien.cremeriefamily.info/scoop}, year = {2011}, note = {This R package fits coop-Lasso, group-Lasso and tree-group Lasso variants for linear regression and logistic regression. The cooperative-Lasso (in short, coop-Lasso) may be viewed as a modification of the group-Lasso penalty that promotes sign coherence and that allows zeros within groups.} }

- Chiquet J, Grasseau G, Ambroise C, Charbonnier C:
**SIMoNe: Statistical Inference for MOdular NEtworks**, 2010

SIMoNe (Statistical Inference for MOdular NEtworks) is an R package which implements the inference of co-regulated networks based on partial correlation coefficients from either steady-state or time-course transcriptomic data. This package can deal with samples collected in different experimental conditions. In this particular case, multiple related graphs are inferred simultaneously. The underlying statistical tools enter the framework of Gaussian graphical models (GGM). Basically, the algorithm searches for a latent clustering of the network to drive the selection of edges through an adaptive l1-penalization of the model likelihood.

download web page@soft{simone, author = {Chiquet, J. and Grasseau, G. and Ambroise, C. and Charbonnier, C.}, title = {SIMoNe: Statistical Inference for MOdular NEtworks}, howpublished = {\url{http://julien.cremeriefamily.info/simone}}, download = {http://cran.r-project.org/web/packages/simone/}, web = {http://julien.cremeriefamily.info/simone}, year = {2010}, note = {SIMoNe (Statistical Inference for MOdular NEtworks) is an R package which implements the inference of co-regulated networks based on partial correlation coefficients from either steady-state or time-course transcriptomic data. This package can deal with samples collected in different experimental conditions. In this particular case, multiple related graphs are inferred simultaneously. The underlying statistical tools enter the framework of Gaussian graphical models (GGM). Basically, the algorithm searches for a latent clustering of the network to drive the selection of edges through an adaptive l1-penalization of the model likelihood.} }