README.md

In rnaseqWrapper: Wrapper for several R packages and scripts to automate RNA-seq analysis

To install from CRAN

This should download the rnaseqWrapper package and all of its smaller dependencies:

install.packages("rnaseqWrapper")

However, the three larger dependencies, including two from Bioconductor, are not automatically installed, though you will be prompted to install them if you call a function that needs them. To install them on your own use:

install.packages("seqinr")

source("http://bioconductor.org/biocLite.R")
biocLite(c("DESeq","topGO"))

Note that these are large packages, and may take a while to install.

To install from this development version

• Download the tar.gz file from the "downloads" tab of the Bitbucket repo
• Open R, and setwd() to the location of the tar file
• Run the following:

Code:

### Install dependencies ###
install.packages(c("ecodist","gplots","seqinr","gdata"))

source("http://bioconductor.org/biocLite.R")
biocLite(c("DESeq","topGO"))

## note that DESeq, topGO and seqinr are only "suggested" and only needed for certain functions

### Install this package ###
install.packages("rnaseqWrapper_1.0.tar.gz",repos=NULL,type="source")

Version notes

• Development Version

  • No current development differences from the official CRAN release.

• Version 1.0-1

  • Minor update to the help pages to avoid running when a required package is not available.

• Version 1.0

  • First official release version.

To start using

Load the package

library(rnaseqWrapper)

List functions available in rnaseqWrapper

lsf.str("package:rnaseqWrapper")

Available functions

I will write some sort of vignette in the near(ish) future (I hope). For now, here is a brief description of the functions, broken down by area.

Merge count data

• mergeCountFiles - Reads all the count files from a directory and merges them into a single file.

Run DESeq

• DESeqWrapper - runs DESeq with sensible defaults and outputs the results and some plots

Make basic calculations on data

These represent the most commonly done/used calculations on basic read/FPKM data, and (hopefully) make it easier to calculate them

• calcLogVal - Calculates the log value (your choice of base) for multiple similarly named columns at once
• calcCombVals - Calculates summary stats for basic columns (e.g. reads) based on matches, thus allowing separate calculations by treatment.
• calcBasicDE - Calculates the difference between treatments, (generally by the mean of log values, calculated with the above)

A distance function

• calcHornMatrix - Calculates a distance measure between samples, which is useful for plotting. (See the help for a good example.)

A slight tweak to a plot

• heatmap.mark - This marginally tweaks the heatmap.2 function, but allows a bit more control useful for DESeq.

Reading in variant files

These are just two common variant formats, but most of the other possible formats won't need to be modified for downstream use in the package.

• readVariantFiles - Reads in individual variant files, and can merge them. This is especially useful for CLC genomics, and other formats run separately on different samples.
• parseVarScan - Separates the columns of varScan output format, which improves some usability. This step is not needed for downstream use, but may still prove helpful for other applications.

Variant Analysis

These functions should work on nearly any input format of variant information, and (I hope) represent the most commonly used basic analyses. All but calculateThirdPosBias require a reference file, and even calculateThirdPosBias is only sort of worthwhile without one

• kaksFromVariants - Calculate Ka/Ks ratios
• nSynNonSites - Calculate the number of synonymous and non-synonymous sites in genes from a reference. Useful to complement the Ka/Ks output for genes with no identified variants.
• determineSynonymous - Determines whether each variant is synonymous or non-synonymous compared to the reference position. Also calculates dN/dS (wihout respect to sites).
• calculateThirdPosBias - Calculate the portion of variants (in each gene) at each codon position. A nominal proxy for Ka/Ks and dN/dS, if needed. When no reference is used, it assumes the most common variant position is the third position.

Miscellaneous info

For more usage information, or to report bugs and enhancement requests, please contact the authors. This package was very recently converted from in-house scripts, so some idiosyncrasies may persist.

rnaseqWrapper documentation built on May 2, 2019, 5:58 a.m.