AC-PCA simultaneously performs dimension reduction and adjustment for confounding variation.
If you use AC-PCA in published research, please cite: Z. Lin, C. Yang, Y. Zhu, J. C. Duchi, Y. Fu, Y. Wang, B. Jiang, M. Zamanighomi, X. Xu, M. Li, N. Sestan, H. Zhao, W. H. Wong: AC-PCA: simultaneous dimension reduction and adjustment for confounding variation bioRxiv, http://dx.doi.org/10.1101/040485
R package: make sure you have the latest R version
A) To install the R package:
-
In R/Rstudio, type install.packages("RSpectra") to install the package "RSpectra";
-
Download 'acPCA_version_number.tar.gz' in "linzx06/AC-PCA/R_package/";
-
In R/Rstudio, type install.packages("acPCA_version_number.tar.gz", repos = NULL, type="source")
B) User's guide for the R package, "linzx06/AC-PCA/R_user_guide.pdf"
C) The data examples ('.rda') are wrapped in "acPCA_version_number.tar.gz". In case you just want to download the data, it's in "linzx06/AC-PCA/R_package/acPCA/data/"
Matlab source code
A) Functions ('.m') are in "linzx06/AC-PCA/Matlab_code/"
B) The data examples ('.mat') are also in "linzx06/AC-PCA/Matlab_code/". These are the same examples as in the R package.
C) 'data_description.m' in "linzx06/AC-PCA/Matlab_code/" provides decription for the data examples.
D) 'examples_implementation.m' in "linzx06/AC-PCA/Matlab_code/" provides examples for running the functions.
F) The matlab functions match one to one with the R functions. "linzx06/AC-PCA/R_user_guide.pdf" is also instructive.