Name Mode Size
R 040000
data 040000
docs 040000
inst 040000
man 040000
manuscript 040000
tests 040000
vignettes 040000
.Rbuildignore 100644 0 kb
.gitignore 100644 0 kb
.travis.yml 100644 1 kb 100644 4 kb
DESCRIPTION 100644 2 kb
LICENSE 100644 0 kb 100644 1 kb
Makefile 100644 1 kb
NAMESPACE 100644 1 kb 100644 4 kb
README.Rmd 100644 6 kb 100644 8 kb
_pkgdown.yml 100644 0 kb
appveyor.yml 100644 2 kb
codecov.yml 100644 0 kb
<!-- is generated from README.Rmd. Please edit that file --> # R/`scPCA` [![Travis CI Build Status](]( [![AppVeyor Build Status](]( [![Codecov test coverage](]( [![Project Status: Active – The project has reached a stable, usable state and is being actively developed.](]( [![BioC status](]( [![Bioc Time](]( [![status](]( [![MIT license](]( > Sparse Contrastive Principal Component Analysis for Computational > Biology **Authors:** [Philippe Boileau](, [Nima Hejazi](, [Sandrine Dudoit]( ------------------------------------------------------------------------ ## What’s `scPCA`? The exploration and analysis of modern high-dimensional biological data regularly involves the use of dimension reduction techniques in order to tease out meaningful and interpretable information from complex experimental data, often subject to batch effects and other noise. In tandem with the development of sequencing technology (e.g., RNA-seq, scRNA-seq), many variants of PCA have been developed in attempts to remedy deficiencies in interpretability and stability that plague vanilla PCA. Such developments have included both various forms of sparse PCA (SPCA) (Zou, Hastie, and Tibshirani 2006; Erichson et al. 2018), which increase the stability and interpretability of principal component loadings in high dimensions, and, more recently, contrastive PCA (cPCA) (Abid et al. 2018), which captures relevant information in the target (experimental) data set by eliminating technical noise through comparison to a so-called background data set. While SPCA and cPCA have both individually proven useful in resolving distinct shortcomings of PCA, neither is capable of simultaneously tackling the issues of interpretability, stability and relevance simultaneously. The `scPCA` package implements *sparse contrastive PCA* (Boileau, Hejazi, and Dudoit 2020) to accomplish these tasks in the context of high-dimensional biological data. In addition to implementing this newly developed technique, the `scPCA` package implements cPCA and generalizations thereof. ------------------------------------------------------------------------ ## Installation For standard use, install from [Bioconductor]( using [`BiocManager`]( ``` r if (!requireNamespace("BiocManager", quietly=TRUE)) { install.packages("BiocManager") } BiocManager::install("scPCA") ``` To contribute, install the bleeding-edge *development version* from GitHub via [`remotes`]( ``` r remotes::install_github("PhilBoileau/scPCA") ``` Current and prior [Bioconductor]( releases are available under branches with numbers prefixed by “RELEASE\_”. For example, to install the version of this package available via Bioconductor 3.10, use ``` r remotes::install_github("PhilBoileau/scPCA@RELEASE_3_10") ``` ------------------------------------------------------------------------ ## Example For details on how to best use the `scPCA` R package, please consult the most recent [package vignette]( available through the [Bioconductor project]( ------------------------------------------------------------------------ ## Issues If you encounter any bugs or have any specific feature requests, please [file an issue]( ------------------------------------------------------------------------ ## Contributions Contributions are welcome. Interested contributors should consult our [contribution guidelines]( prior to submitting a pull request. ------------------------------------------------------------------------ ## Citation Please cite the first paper below after using the `scPCA` R software package. Please also make sure to cite the article describing the statistical methodology when using scPCA or cross-validated cPCA as part of an analysis. @article{boileau2020scPCAjoss, doi = {10.21105/joss.02079}, url = {}, year = {2020}, publisher = {The Open Journal}, volume = {5}, number = {46}, pages = {2079}, author = {Philippe Boileau and Nima Hejazi and Sandrine Dudoit}, title = {scPCA: A toolbox for sparse contrastive principal component analysis in R}, journal = {Journal of Open Source Software} } @article{boileau2020scPCA, author = {Boileau, Philippe and Hejazi, Nima S and Dudoit, Sandrine}, title = "{Exploring High-Dimensional Biological Data with Sparse Contrastive Principal Component Analysis}", journal = {Bioinformatics}, year = {2020}, month = {03}, issn = {1367-4803}, doi = {10.1093/bioinformatics/btaa176}, url = {}, note = {btaa176}, eprint = {}, } ------------------------------------------------------------------------ ## License © 2019-2023 [Philippe Boileau]( The contents of this repository are distributed under the MIT license. See file `LICENSE` for details. ------------------------------------------------------------------------ ## References <div id="refs" class="references csl-bib-body hanging-indent"> <div id="ref-abid2018exploring" class="csl-entry"> Abid, Abubakar, Martin J Zhang, Vivek K Bagaria, and James Zou. 2018. “Exploring Patterns Enriched in a Dataset with Contrastive Principal Component Analysis.” *Nature Communications* 9 (1): 2134. </div> <div id="ref-boileau2020" class="csl-entry"> Boileau, Philippe, Nima S Hejazi, and Sandrine Dudoit. 2020. “<span class="nocase">Exploring High-Dimensional Biological Data with Sparse Contrastive Principal Component Analysis</span>.” *Bioinformatics*, March. <>. </div> <div id="ref-erichson2018sparse" class="csl-entry"> Erichson, N. Benjamin, Peng Zeng, Krithika Manohar, Steven L. Brunton, J. Nathan Kutz, and Aleksandr Y. Aravkin. 2018. “Sparse Principal Component Analysis via Variable Projection.” *ArXiv* abs/1804.00341. </div> <div id="ref-zou2006sparse" class="csl-entry"> Zou, Hui, Trevor Hastie, and Robert Tibshirani. 2006. “Sparse Principal Component Analysis.” *Journal of Computational and Graphical Statistics* 15 (2): 265–86. </div> </div>