From b90dea14dc7755c1572eb4ebcca600510f08693e Mon Sep 17 00:00:00 2001 From: Julien Chiquet Date: Fri, 19 Jun 2026 09:59:41 +0200 Subject: [PATCH 1/4] correct link in README.qmd --- README.md | 5 +++-- README.qmd | 2 +- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index 9d71f590..dfc6fd8b 100644 --- a/README.md +++ b/README.md @@ -18,8 +18,9 @@ stable](https://img.shields.io/badge/lifecycle-stable-blue.svg)](https://lifecyc > of multivariate problems when count data are at play. This package > implements efficient variational algorithms to fit such models, > accompanied with a set of functions for visualization and diagnostic. -> See [this deck of slides](https://pln-team.github.io/slideshow/slides) -> for a comprehensive introduction. +> See [all the dedicated +> vignettes](https://pln-team.github.io/PLNmodels/articles/) for a +> comprehensive introduction. **PLNmodels** covers the following models, all built around the multivariate Poisson-lognormal distribution and sharing a common diff --git a/README.qmd b/README.qmd index cf68c0a5..17a2d9c3 100644 --- a/README.qmd +++ b/README.qmd @@ -23,7 +23,7 @@ knitr::opts_chunk$set( ## Description -> The Poisson lognormal model and variants[^1] can be used for a variety of multivariate problems when count data are at play. This package implements efficient variational algorithms to fit such models, accompanied with a set of functions for visualization and diagnostic. See [this deck of slides](https://pln-team.github.io/slideshow/slides) for a comprehensive introduction. +> The Poisson lognormal model and variants[^1] can be used for a variety of multivariate problems when count data are at play. This package implements efficient variational algorithms to fit such models, accompanied with a set of functions for visualization and diagnostic. See [all the dedicated vignettes](https://pln-team.github.io/PLNmodels/articles/) for a comprehensive introduction. **PLNmodels** covers the following models, all built around the multivariate Poisson-lognormal distribution and sharing a common formula-based interface (covariates, offsets, weights) and a choice of optimization backends (a fast built-in Newton solver, NLOPT, and an experimental torch backend): From b9189c87120fa0d3a2589dfa8eda46d2b7d32d6c Mon Sep 17 00:00:00 2001 From: Julien Chiquet Date: Fri, 19 Jun 2026 13:18:57 +0200 Subject: [PATCH 2/4] Cite the ZIPLN and ZIPLNnetwork methodology papers - Add bib entries for Batardiere, Chiquet, Gindraud & Mariadassou (2025, Statistics and Computing) and Tous, Chiquet, Deacon, Fontrodona-Eslava, Fraser & Magurran (2025, bioRxiv). - ZIPLN vignette: cite the dedicated ZIPLN paper where the model is introduced, and the ZIPLNnetwork bioRxiv preprint alongside the existing PLNnetwork citation. - README.qmd: add a footnote for ZIPLNnetwork citing the same bioRxiv preprint, renumbering the trailing barents footnote accordingly. --- README.md | 12 +++++++++--- README.qmd | 11 ++++++----- vignettes/ZIPLN.Rmd | 4 ++-- vignettes/article/PLNreferences.bib | 17 +++++++++++++++++ 4 files changed, 34 insertions(+), 10 deletions(-) diff --git a/README.md b/README.md index dfc6fd8b..5a8640c4 100644 --- a/README.md +++ b/README.md @@ -41,7 +41,7 @@ experimental torch backend): of PLN models. - **ZIPLN**[^8]: a zero-inflated extension of PLN for data with excess zeros, with the same family of covariance structures and an optional - sparse (`ZIPLNnetwork`) variant. + sparse (`ZIPLNnetwork`[^9]) variant. ## Installation @@ -57,7 +57,7 @@ remotes::install_github("pln-team/PLNmodels@tag_number") # a specific tagged re ## Illustration -We illustrate the main models on the `barents` data set[^9]: the +We illustrate the main models on the `barents` data set[^10]: the abundance of 30 fish species observed in 89 sites in the Barents sea, along with depth, temperature and geographic coordinates for each site. @@ -273,6 +273,12 @@ table(cluster = myMixture$memberships, zone = barents$zone) Statistics and Computing, 35, 2025. [doi:10.1007/s11222-025-10729-0](https://doi.org/10.1007/s11222-025-10729-0) -[^9]: Fossheim, M., Nilssen, E. M. and Aschan, M. Fish assemblages in +[^9]: Tous, J., Chiquet, J., Deacon, A. E., Fontrodona-Eslava, A., + Fraser, D. F. and Magurran, A. E. A JSDM with zero-inflation to + improve inference of association networks from count community data + with structural zeros. bioRxiv preprint, 2025. + [doi:10.1101/2025.07.24.666553](https://doi.org/10.1101/2025.07.24.666553) + +[^10]: Fossheim, M., Nilssen, E. M. and Aschan, M. Fish assemblages in the Barents Sea. Marine Biology Research, 2(4), 2006. [doi:10.1080/17451000600815698](https://doi.org/10.1080/17451000600815698) diff --git a/README.qmd b/README.qmd index 17a2d9c3..8fffee0f 100644 --- a/README.qmd +++ b/README.qmd @@ -27,12 +27,12 @@ knitr::opts_chunk$set( **PLNmodels** covers the following models, all built around the multivariate Poisson-lognormal distribution and sharing a common formula-based interface (covariates, offsets, weights) and a choice of optimization backends (a fast built-in Newton solver, NLOPT, and an experimental torch backend): -- **PLN**[^2]: unpenalized multivariate Poisson regression, with several covariance structures (full, diagonal, spherical, fixed, or a genetic/heritability structure). +- **PLN**[^1][^2]: unpenalized multivariate Poisson regression, with several covariance structures (full, diagonal, spherical, fixed, or a genetic/heritability structure). - **PLNPCA**[^3]: probabilistic Poisson PCA — a rank-constrained covariance for dimension reduction and visualization. -- **PLNLDA**: Poisson lognormal discriminant analysis[^4] for the supervised classification of count data. +- **PLNLDA**[^1]: Poisson lognormal discriminant analysis[^4] for the supervised classification of count data. - **PLNnetwork**[^5]: sparse inverse-covariance (network) inference via a graphical-lasso-like penalty[^6]. - **PLNmixture**: model-based clustering[^7] of count data via a mixture of PLN models. -- **ZIPLN**[^8]: a zero-inflated extension of PLN for data with excess zeros, with the same family of covariance structures and an optional sparse (`ZIPLNnetwork`) variant. +- **ZIPLN**[^8]: a zero-inflated extension of PLN for data with excess zeros, with the same family of covariance structures and an optional sparse (`ZIPLNnetwork`[^9]) variant. [^1]: J. Chiquet, M. Mariadassou and S. Robin: The Poisson-lognormal model as a versatile framework for the joint analysis of species abundances, Frontiers in Ecology and Evolution, 2021. [doi:10.3389/fevo.2021.588292](https://www.frontiersin.org/articles/10.3389/fevo.2021.588292/full) [^2]: Aitchison, J. and Ho, C. H. The multivariate Poisson-log normal distribution. Biometrika, 76(4), 1989, 643–653. @@ -42,6 +42,7 @@ knitr::opts_chunk$set( [^6]: Friedman, J., Hastie, T. and Tibshirani, R. Sparse inverse covariance estimation with the graphical lasso. Biostatistics, 9(3), 2008. [^7]: Fraley, C. and Raftery, A. E. MCLUST: Software for model-based cluster analysis. Journal of Classification, 16(2), 1999. [^8]: Batardière, B., Chiquet, J., Gindraud, F. and Mariadassou, M. Zero-inflation in the multivariate Poisson lognormal family. Statistics and Computing, 35, 2025. [doi:10.1007/s11222-025-10729-0](https://doi.org/10.1007/s11222-025-10729-0) +[^9]: Tous, J., Chiquet, J., Deacon, A. E., Fontrodona-Eslava, A., Fraser, D. F. and Magurran, A. E. A JSDM with zero-inflation to improve inference of association networks from count community data with structural zeros. bioRxiv preprint, 2025. [doi:10.1101/2025.07.24.666553](https://doi.org/10.1101/2025.07.24.666553) ## Installation @@ -55,9 +56,9 @@ remotes::install_github("pln-team/PLNmodels@tag_number") # a specific tagged re ## Illustration -We illustrate the main models on the `barents` data set[^9]: the abundance of 30 fish species observed in 89 sites in the Barents sea, along with depth, temperature and geographic coordinates for each site. +We illustrate the main models on the `barents` data set[^10]: the abundance of 30 fish species observed in 89 sites in the Barents sea, along with depth, temperature and geographic coordinates for each site. -[^9]: Fossheim, M., Nilssen, E. M. and Aschan, M. Fish assemblages in the Barents Sea. Marine Biology Research, 2(4), 2006. [doi:10.1080/17451000600815698](https://doi.org/10.1080/17451000600815698) +[^10]: Fossheim, M., Nilssen, E. M. and Aschan, M. Fish assemblages in the Barents Sea. Marine Biology Research, 2(4), 2006. [doi:10.1080/17451000600815698](https://doi.org/10.1080/17451000600815698) ```{r load} library(PLNmodels) diff --git a/vignettes/ZIPLN.Rmd b/vignettes/ZIPLN.Rmd index c8322382..2c6aeddf 100644 --- a/vignettes/ZIPLN.Rmd +++ b/vignettes/ZIPLN.Rmd @@ -50,7 +50,7 @@ mean(microcosm$Abundance == 0) ### Mathematical background -The zero-inflated PLN model (ZIPLN) combines the Poisson lognormal model [@AiH89] -- see [the PLN vignette](PLN.html) -- with a zero-inflation mechanism: each count $Y_{ij}$ is either a structural zero (with probability $\pi_{ij}$) or drawn from the usual PLN generative process: +The zero-inflated PLN model (ZIPLN) [@ZIPLN] combines the Poisson lognormal model [@AiH89] -- see [the PLN vignette](PLN.html) -- with a zero-inflation mechanism: each count $Y_{ij}$ is either a structural zero (with probability $\pi_{ij}$) or drawn from the usual PLN generative process: \begin{equation} \begin{array}{rcl} \text{latent space } & \mathbf{Z}_i \sim \mathcal{N}\left({\boldsymbol\mu},\boldsymbol\Sigma\right) & \\ @@ -66,7 +66,7 @@ Just like PLN, ${\boldsymbol\mu}$ generalizes to $\mathbf{o}_i + \mathbf{x}_i^\t - `"col"`: one $\pi_j$ per species. - covariates: $\text{logit}(\pi_{ij}) = \mathbf{x}_{0,i}^\top\mathbf{B}_{0,j}$, specified with the formula syntax `Y ~ PLN effect | ZI effect` (see below). -`ZIPLNnetwork` further adds a sparsity penalty on $\boldsymbol\Omega = \boldsymbol\Sigma^{-1}$, exactly as `PLNnetwork` does for PLN (see [the PLNnetwork vignette](PLNnetwork.html) and @PLNnetwork), so that both the excess of zeros and the residual dependency structure between taxa are accounted for. +`ZIPLNnetwork` further adds a sparsity penalty on $\boldsymbol\Omega = \boldsymbol\Sigma^{-1}$, exactly as `PLNnetwork` does for PLN (see [the PLNnetwork vignette](PLNnetwork.html) and @PLNnetwork), so that both the excess of zeros and the residual dependency structure between taxa are accounted for. See @ZIPLNnetwork for an application to species association networks from count data with structural zeros. ## Analysis of microcosm with ZIPLN diff --git a/vignettes/article/PLNreferences.bib b/vignettes/article/PLNreferences.bib index a4bd9069..c5c7cc2c 100644 --- a/vignettes/article/PLNreferences.bib +++ b/vignettes/article/PLNreferences.bib @@ -44,6 +44,23 @@ @InProceedings{PLNnetwork url = {http://proceedings.mlr.press/v97/chiquet19a.html}, } +@Article{ZIPLN, + author = {Batardière, Bastien and Chiquet, Julien and Gindraud, François and Mariadassou, Mahendra}, + title = {Zero-inflation in the multivariate Poisson lognormal family}, + journal = {Statistics and Computing}, + year = {2025}, + volume = {35}, + doi = {10.1007/s11222-025-10729-0}, +} + +@Unpublished{ZIPLNnetwork, + author = {Tous, Jeanne and Chiquet, Julien and Deacon, Amy E. and Fontrodona-Eslava, Ada and Fraser, Douglas F. and Magurran, Anne E.}, + title = {A JSDM with zero-inflation to improve inference of association networks from count community data with structural zeros}, + year = {2025}, + note = {bioRxiv preprint}, + doi = {10.1101/2025.07.24.666553}, +} + @inproceedings{trichoptera, title={Influence des facteurs météorologiques sur les résultats de piégeage lumineux}, author={Usseglio-Polatera, P. and Auda, Y.}, From d160c28d24c38f16b6e11a37dc0b6393c87ab13c Mon Sep 17 00:00:00 2001 From: Julien Chiquet Date: Fri, 19 Jun 2026 13:47:56 +0200 Subject: [PATCH 3/4] Fix broken CRAN badge in README --- README.md | 50 ++++++++++++++++++++++++++++++-------------------- README.qmd | 6 ++++-- 2 files changed, 34 insertions(+), 22 deletions(-) diff --git a/README.md b/README.md index 5a8640c4..26998046 100644 --- a/README.md +++ b/README.md @@ -6,7 +6,7 @@ [![R-CMD-check](https://github.com/PLN-team/PLNmodels/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/PLN-team/PLNmodels/actions/workflows/R-CMD-check.yaml) [![Coverage status](https://codecov.io/gh/pln-team/PLNmodels/branch/master/graph/badge.svg)](https://codecov.io/github/pln-team/PLNmodels?branch=master) -[![CRAN_Status_Badge](http://www.r-pkg.org/badges/version/PLNmodels.png)](https://cran.r-project.org/package=PLNmodels) +[![CRAN_Status_Badge](https://www.r-pkg.org/badges/version/PLNmodels)](https://cran.r-project.org/package=PLNmodels) [![Lifecycle: stable](https://img.shields.io/badge/lifecycle-stable-blue.svg)](https://lifecycle.r-lib.org/articles/stages.html) [![](https://img.shields.io/github/last-commit/pln-team/PLNmodels.svg)](https://github.com/pln-team/PLNmodels/commits/master) @@ -28,20 +28,20 @@ formula-based interface (covariates, offsets, weights) and a choice of optimization backends (a fast built-in Newton solver, NLOPT, and an experimental torch backend): -- **PLN**[^2]: unpenalized multivariate Poisson regression, with several - covariance structures (full, diagonal, spherical, fixed, or a +- **PLN**[^2][^3]: unpenalized multivariate Poisson regression, with + several covariance structures (full, diagonal, spherical, fixed, or a genetic/heritability structure). -- **PLNPCA**[^3]: probabilistic Poisson PCA — a rank-constrained +- **PLNPCA**[^4]: probabilistic Poisson PCA — a rank-constrained covariance for dimension reduction and visualization. -- **PLNLDA**: Poisson lognormal discriminant analysis[^4] for the +- **PLNLDA**[^5]: Poisson lognormal discriminant analysis[^6] for the supervised classification of count data. -- **PLNnetwork**[^5]: sparse inverse-covariance (network) inference via - a graphical-lasso-like penalty[^6]. -- **PLNmixture**: model-based clustering[^7] of count data via a mixture +- **PLNnetwork**[^7]: sparse inverse-covariance (network) inference via + a graphical-lasso-like penalty[^8]. +- **PLNmixture**: model-based clustering[^9] of count data via a mixture of PLN models. -- **ZIPLN**[^8]: a zero-inflated extension of PLN for data with excess +- **ZIPLN**[^10]: a zero-inflated extension of PLN for data with excess zeros, with the same family of covariance structures and an optional - sparse (`ZIPLNnetwork`[^9]) variant. + sparse (`ZIPLNnetwork`[^11]) variant. ## Installation @@ -57,7 +57,7 @@ remotes::install_github("pln-team/PLNmodels@tag_number") # a specific tagged re ## Illustration -We illustrate the main models on the `barents` data set[^10]: the +We illustrate the main models on the `barents` data set[^12]: the abundance of 30 fish species observed in 89 sites in the Barents sea, along with depth, temperature and geographic coordinates for each site. @@ -243,42 +243,52 @@ table(cluster = myMixture$memberships, zone = barents$zone) abundances, Frontiers in Ecology and Evolution, 2021. [doi:10.3389/fevo.2021.588292](https://www.frontiersin.org/articles/10.3389/fevo.2021.588292/full) -[^2]: Aitchison, J. and Ho, C. H. The multivariate Poisson-log normal +[^2]: J. Chiquet, M. Mariadassou and S. Robin: The Poisson-lognormal + model as a versatile framework for the joint analysis of species + abundances, Frontiers in Ecology and Evolution, 2021. + [doi:10.3389/fevo.2021.588292](https://www.frontiersin.org/articles/10.3389/fevo.2021.588292/full) + +[^3]: Aitchison, J. and Ho, C. H. The multivariate Poisson-log normal distribution. Biometrika, 76(4), 1989, 643–653. -[^3]: J. Chiquet, M. Mariadassou and S. Robin: Variational inference for +[^4]: J. Chiquet, M. Mariadassou and S. Robin: Variational inference for probabilistic Poisson PCA, the Annals of Applied Statistics, 12: 2674–2698, 2018. [doi:10.1214/18-AOAS1177](http://dx.doi.org/10.1214/18%2DAOAS1177) -[^4]: Fisher, R. A. The use of multiple measurements in taxonomic +[^5]: J. Chiquet, M. Mariadassou and S. Robin: The Poisson-lognormal + model as a versatile framework for the joint analysis of species + abundances, Frontiers in Ecology and Evolution, 2021. + [doi:10.3389/fevo.2021.588292](https://www.frontiersin.org/articles/10.3389/fevo.2021.588292/full) + +[^6]: Fisher, R. A. The use of multiple measurements in taxonomic problems. Annals of Eugenics, 7(2), 1936; Rao, C. R. The utilization of multiple measurements in problems of biological classification. JRSS B, 10(2), 1948. -[^5]: J. Chiquet, M. Mariadassou and S. Robin: Variational inference for +[^7]: J. Chiquet, M. Mariadassou and S. Robin: Variational inference for sparse network reconstruction from count data, Proceedings of the 36th International Conference on Machine Learning (ICML), 2019. [link](http://proceedings.mlr.press/v97/chiquet19a.html) -[^6]: Friedman, J., Hastie, T. and Tibshirani, R. Sparse inverse +[^8]: Friedman, J., Hastie, T. and Tibshirani, R. Sparse inverse covariance estimation with the graphical lasso. Biostatistics, 9(3), 2008. -[^7]: Fraley, C. and Raftery, A. E. MCLUST: Software for model-based +[^9]: Fraley, C. and Raftery, A. E. MCLUST: Software for model-based cluster analysis. Journal of Classification, 16(2), 1999. -[^8]: Batardière, B., Chiquet, J., Gindraud, F. and Mariadassou, M. +[^10]: Batardière, B., Chiquet, J., Gindraud, F. and Mariadassou, M. Zero-inflation in the multivariate Poisson lognormal family. Statistics and Computing, 35, 2025. [doi:10.1007/s11222-025-10729-0](https://doi.org/10.1007/s11222-025-10729-0) -[^9]: Tous, J., Chiquet, J., Deacon, A. E., Fontrodona-Eslava, A., +[^11]: Tous, J., Chiquet, J., Deacon, A. E., Fontrodona-Eslava, A., Fraser, D. F. and Magurran, A. E. A JSDM with zero-inflation to improve inference of association networks from count community data with structural zeros. bioRxiv preprint, 2025. [doi:10.1101/2025.07.24.666553](https://doi.org/10.1101/2025.07.24.666553) -[^10]: Fossheim, M., Nilssen, E. M. and Aschan, M. Fish assemblages in +[^12]: Fossheim, M., Nilssen, E. M. and Aschan, M. Fish assemblages in the Barents Sea. Marine Biology Research, 2(4), 2006. [doi:10.1080/17451000600815698](https://doi.org/10.1080/17451000600815698) diff --git a/README.qmd b/README.qmd index 8fffee0f..4cd6335f 100644 --- a/README.qmd +++ b/README.qmd @@ -1,12 +1,14 @@ --- title: "PLNmodels: Poisson lognormal models for multivariate count data" -format: gfm +format: + gfm: + default-image-extension: "" --- [![R-CMD-check](https://github.com/PLN-team/PLNmodels/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/PLN-team/PLNmodels/actions/workflows/R-CMD-check.yaml) [![Coverage status](https://codecov.io/gh/pln-team/PLNmodels/branch/master/graph/badge.svg)](https://codecov.io/github/pln-team/PLNmodels?branch=master) -[![CRAN_Status_Badge](http://www.r-pkg.org/badges/version/PLNmodels)](https://cran.r-project.org/package=PLNmodels) +[![CRAN_Status_Badge](https://www.r-pkg.org/badges/version/PLNmodels)](https://cran.r-project.org/package=PLNmodels) [![Lifecycle: stable](https://img.shields.io/badge/lifecycle-stable-blue.svg)](https://lifecycle.r-lib.org/articles/stages.html) [![](https://img.shields.io/github/last-commit/pln-team/PLNmodels.svg)](https://github.com/pln-team/PLNmodels/commits/master) From 32c77447c2becf8ebe0917607aa78be0bca3cb54 Mon Sep 17 00:00:00 2001 From: Julien Chiquet Date: Fri, 19 Jun 2026 13:53:24 +0200 Subject: [PATCH 4/4] Fix duplicated Frontiers 2021 reference in README footnotes --- README.md | 48 +++++++++++++++++++----------------------------- README.qmd | 4 ++-- 2 files changed, 21 insertions(+), 31 deletions(-) diff --git a/README.md b/README.md index 26998046..bc348b6b 100644 --- a/README.md +++ b/README.md @@ -28,20 +28,20 @@ formula-based interface (covariates, offsets, weights) and a choice of optimization backends (a fast built-in Newton solver, NLOPT, and an experimental torch backend): -- **PLN**[^2][^3]: unpenalized multivariate Poisson regression, with - several covariance structures (full, diagonal, spherical, fixed, or a +- **PLN**[^2]: unpenalized multivariate Poisson regression, with several + covariance structures (full, diagonal, spherical, fixed, or a genetic/heritability structure). -- **PLNPCA**[^4]: probabilistic Poisson PCA — a rank-constrained +- **PLNPCA**[^3]: probabilistic Poisson PCA — a rank-constrained covariance for dimension reduction and visualization. -- **PLNLDA**[^5]: Poisson lognormal discriminant analysis[^6] for the +- **PLNLDA**: Poisson lognormal discriminant analysis[^4] for the supervised classification of count data. -- **PLNnetwork**[^7]: sparse inverse-covariance (network) inference via - a graphical-lasso-like penalty[^8]. -- **PLNmixture**: model-based clustering[^9] of count data via a mixture +- **PLNnetwork**[^5]: sparse inverse-covariance (network) inference via + a graphical-lasso-like penalty[^6]. +- **PLNmixture**: model-based clustering[^7] of count data via a mixture of PLN models. -- **ZIPLN**[^10]: a zero-inflated extension of PLN for data with excess +- **ZIPLN**[^8]: a zero-inflated extension of PLN for data with excess zeros, with the same family of covariance structures and an optional - sparse (`ZIPLNnetwork`[^11]) variant. + sparse (`ZIPLNnetwork`[^9]) variant. ## Installation @@ -57,7 +57,7 @@ remotes::install_github("pln-team/PLNmodels@tag_number") # a specific tagged re ## Illustration -We illustrate the main models on the `barents` data set[^12]: the +We illustrate the main models on the `barents` data set[^10]: the abundance of 30 fish species observed in 89 sites in the Barents sea, along with depth, temperature and geographic coordinates for each site. @@ -243,52 +243,42 @@ table(cluster = myMixture$memberships, zone = barents$zone) abundances, Frontiers in Ecology and Evolution, 2021. [doi:10.3389/fevo.2021.588292](https://www.frontiersin.org/articles/10.3389/fevo.2021.588292/full) -[^2]: J. Chiquet, M. Mariadassou and S. Robin: The Poisson-lognormal - model as a versatile framework for the joint analysis of species - abundances, Frontiers in Ecology and Evolution, 2021. - [doi:10.3389/fevo.2021.588292](https://www.frontiersin.org/articles/10.3389/fevo.2021.588292/full) - -[^3]: Aitchison, J. and Ho, C. H. The multivariate Poisson-log normal +[^2]: Aitchison, J. and Ho, C. H. The multivariate Poisson-log normal distribution. Biometrika, 76(4), 1989, 643–653. -[^4]: J. Chiquet, M. Mariadassou and S. Robin: Variational inference for +[^3]: J. Chiquet, M. Mariadassou and S. Robin: Variational inference for probabilistic Poisson PCA, the Annals of Applied Statistics, 12: 2674–2698, 2018. [doi:10.1214/18-AOAS1177](http://dx.doi.org/10.1214/18%2DAOAS1177) -[^5]: J. Chiquet, M. Mariadassou and S. Robin: The Poisson-lognormal - model as a versatile framework for the joint analysis of species - abundances, Frontiers in Ecology and Evolution, 2021. - [doi:10.3389/fevo.2021.588292](https://www.frontiersin.org/articles/10.3389/fevo.2021.588292/full) - -[^6]: Fisher, R. A. The use of multiple measurements in taxonomic +[^4]: Fisher, R. A. The use of multiple measurements in taxonomic problems. Annals of Eugenics, 7(2), 1936; Rao, C. R. The utilization of multiple measurements in problems of biological classification. JRSS B, 10(2), 1948. -[^7]: J. Chiquet, M. Mariadassou and S. Robin: Variational inference for +[^5]: J. Chiquet, M. Mariadassou and S. Robin: Variational inference for sparse network reconstruction from count data, Proceedings of the 36th International Conference on Machine Learning (ICML), 2019. [link](http://proceedings.mlr.press/v97/chiquet19a.html) -[^8]: Friedman, J., Hastie, T. and Tibshirani, R. Sparse inverse +[^6]: Friedman, J., Hastie, T. and Tibshirani, R. Sparse inverse covariance estimation with the graphical lasso. Biostatistics, 9(3), 2008. -[^9]: Fraley, C. and Raftery, A. E. MCLUST: Software for model-based +[^7]: Fraley, C. and Raftery, A. E. MCLUST: Software for model-based cluster analysis. Journal of Classification, 16(2), 1999. -[^10]: Batardière, B., Chiquet, J., Gindraud, F. and Mariadassou, M. +[^8]: Batardière, B., Chiquet, J., Gindraud, F. and Mariadassou, M. Zero-inflation in the multivariate Poisson lognormal family. Statistics and Computing, 35, 2025. [doi:10.1007/s11222-025-10729-0](https://doi.org/10.1007/s11222-025-10729-0) -[^11]: Tous, J., Chiquet, J., Deacon, A. E., Fontrodona-Eslava, A., +[^9]: Tous, J., Chiquet, J., Deacon, A. E., Fontrodona-Eslava, A., Fraser, D. F. and Magurran, A. E. A JSDM with zero-inflation to improve inference of association networks from count community data with structural zeros. bioRxiv preprint, 2025. [doi:10.1101/2025.07.24.666553](https://doi.org/10.1101/2025.07.24.666553) -[^12]: Fossheim, M., Nilssen, E. M. and Aschan, M. Fish assemblages in +[^10]: Fossheim, M., Nilssen, E. M. and Aschan, M. Fish assemblages in the Barents Sea. Marine Biology Research, 2(4), 2006. [doi:10.1080/17451000600815698](https://doi.org/10.1080/17451000600815698) diff --git a/README.qmd b/README.qmd index 4cd6335f..812156f4 100644 --- a/README.qmd +++ b/README.qmd @@ -29,9 +29,9 @@ knitr::opts_chunk$set( **PLNmodels** covers the following models, all built around the multivariate Poisson-lognormal distribution and sharing a common formula-based interface (covariates, offsets, weights) and a choice of optimization backends (a fast built-in Newton solver, NLOPT, and an experimental torch backend): -- **PLN**[^1][^2]: unpenalized multivariate Poisson regression, with several covariance structures (full, diagonal, spherical, fixed, or a genetic/heritability structure). +- **PLN**[^2]: unpenalized multivariate Poisson regression, with several covariance structures (full, diagonal, spherical, fixed, or a genetic/heritability structure). - **PLNPCA**[^3]: probabilistic Poisson PCA — a rank-constrained covariance for dimension reduction and visualization. -- **PLNLDA**[^1]: Poisson lognormal discriminant analysis[^4] for the supervised classification of count data. +- **PLNLDA**: Poisson lognormal discriminant analysis[^4] for the supervised classification of count data. - **PLNnetwork**[^5]: sparse inverse-covariance (network) inference via a graphical-lasso-like penalty[^6]. - **PLNmixture**: model-based clustering[^7] of count data via a mixture of PLN models. - **ZIPLN**[^8]: a zero-inflated extension of PLN for data with excess zeros, with the same family of covariance structures and an optional sparse (`ZIPLNnetwork`[^9]) variant.