sva

Surrogate Variable Analysis

Bioconductor version: Release (2.12)

The sva package contains functions for removing batch effects and other unwanted variation in high-throughput experiment. Specifically, the sva package contains functions for the identifying and building surrogate variables for high-dimensional data sets. Surrogate variables are covariates constructed directly from high-dimensional data (like gene expression/RNA sequencing/methylation/brain imaging data) that can be used in subsequent analyses to adjust for unknown, unmodeled, or latent sources of noise. The sva package can be used to remove artifacts in two ways: (1) identifying and estimating surrogate variables for unknown sources of variation in high-throughput experiments (Leek and Storey 2007 PLoS Genetics,2008 PNAS) and (2) directly removing known batch effects using ComBat (Johnson et al. 2007 Biostatistics). Removing batch effects and using surrogate variables in differential expression analysis have been shown to reduce dependence, stabilize error rate estimates, and improve reproducibility, see (Leek and Storey 2007 PLoS Genetics, 2008 PNAS or Leek et al. 2011 Nat. Reviews Genetics). Surrogate variable analysis and ComBat were developed in the context of microarray experiments, but may be used as a general tool for high throughput data sets where dependence may be involved.

Author: Jeffrey T. Leek <jleek at jhsph.edu>, W. Evan Johnson <wej at bu.edu>, Hilary S. Parker <hiparker at jhsph.edu>, Andrew E. Jaffe <ajaffe at jhsph.edu>, John D. Storey <jstorey at princeton.edu>,

Maintainer: Jeffrey T. Leek <jleek at jhsph.edu>

To install this package, start R and enter:

    source("http://bioconductor.org/biocLite.R")
    biocLite("sva")

To cite this package in a publication, start R and enter:

    citation("sva")

Documentation

Details

Package Downloads