pathVar (version 1.2.0)

pathVarTwoSamplesDisc: Compares the number of genes in clusters in each pathway for two groups of samples that you define.

Description

Compares the distribution of genes in each pathway for two groups of samples that you define.

Usage

pathVarTwoSamplesDisc(dat.mat,pways,groups,perc=c(1/3,2/3), test=c("chisq", "exact"),varStat=c("sd", "mean", "mad","cv"))

Arguments

dat.mat
matrix with the genes on the rows and the samples on the columns.
pways
list which contains a vector of pathway IDs, a vector of pathway names, and a list of genes in each pathway.
groups
vector indicating the amount of samples and replicates of each sample.
perc
numeric vector of probabiliities with values between 0 and 1. Used to put genes into clusters
test
a string, either "exact" or "chisq" which are tests to see if clusters in the 2 samples are sig. different from each other
varStat
a string specifying the type of variability summary statistic to perform. The options are "sd", "mean", "mad", or "cv".

Value

A geneDistributionSet3 object is returned.

Details

This function splits the samples into two groups that you define. It computes the variability (sd, mad, cv, or mean) for each gene in each group. Then it classifies the genes with respect to the variability in at most 4 clusters. For each pathway, we extract the gene in our dataset and in which cluster they belong. Then for each pathway we look at the gene counts in each category and compare the 2 samples to each other with all the genes from the data set with the Chi-Squared or exact test.

Examples

Run this code
# we run the 2 samples analysis on the first 10 pathways from kegg
pways.kegg.10pways <- lapply(pways.kegg, function(x) x[1:10])
results_2samples=pathVarTwoSamplesDisc(bock,pways.kegg.10pways,groups=as.factor(c(rep(1,10),rep(2,10))),perc=c(1/3,2/3),test="exact",varStat="sd")

Run the code above in your browser using DataCamp Workspace