README.md

In kokrah/cbcbSEQ: RNAseq analysis pipeline for UMD CBCB collaborators with batch effect assessment and correction

cbcbSEQ: RNAseq analysis for UMD CBCB collaborators

The purpose of this pipeline is to streamline the process for analyzing RNA-seq data with potential batch effects.

The pipeline includes:

1. quantile normalization
2. log-transformation of counts
3. ComBat (location) batch correction
4. voom calculation of weights.

The functions in this package can be grouped into two main categories:

1. The functions used for assessing batch effects. a. makeSVD b. pcRes c. plotPC
2. The functions for removing batch effect and computing weights for limma. a. qNorm b. log2CPM c. voomMod d. combatMod e. batchSEQ

batchSEQ is the pipeline function. It combines qNorm, log2CPM, voomMod, and combatMod into one step.

Installation

To begin, install Bioconductor along with a few dependencies that cbcbSEQ uses:

source("http://bioconductor.org/biocLite.R")
biocLite(c('limma', 'preprocessCore', 'sva', 'pasilla'))

Next, use devtools to install the latest version of cbcbSEQ from Github:

require(devtools)
install_github("cbcbSEQ", user="kokrah")

If all went well you should now be able to load cbcbSEQ:

require(cbcbSEQ)
vignette('cbcbSEQIntro', package='cbcbSEQ')

kokrah/cbcbSEQ documentation built on May 20, 2019, 12:54 p.m.