clusterDEgenes: Cluster DE genes by fold change from multiple files

In vivekbhr/vivlib: An R package for useful bioinformatics wrappers

Description

Cluster DE genes by fold change from multiple files

Usage

clusterDEgenes(DEoutList, sampleNames, FDRcutoff = 0.05,
  method = "correlation", cut_cluster = NA, row_annotation = NULL,
  keepNAs = TRUE, outFile_prefix = NULL)

Arguments

DEoutList	A vector, with names of DESeq2 output files to use.
sampleNames	Name of samples corresponding to the DESeq output file list.
FDRcutoff	FDR cutoff to select DE genes from the list
method	Which clustering method to use. "correlation" or "biclustering", will allow output of gene names per cluster. Other methods are also supported (all methods of hclust + kmeans), but won't output genes per cluster
cut_cluster	A number to which the cluster dendrogram will be cut into (NA means do not cut clusters)
row_annotation	A data frame for the annotation of genes, with rownames corresponding to the Row.names column of the DESeq2 output. The columns can have annotations like chromosome, gene type etc..
keepNAs	Many genes will have fold change = NA in some samples after merging, since they are undetected in some samples. Select this if you want to still keep those genes (NAs will be converted to zeros for clustering).
outFile_prefix	A prefix for output files.

Value

A pdf with clustered heatmap, an .Rdata file with the hclust objects and the genes sorted by clustered output, and text file with genes divided by clusters (if cut_clusters is selected).

Examples

## Not run: 
clusterDEgenes(DEoutList, sampleNames, FDRcutoff = 0.05, method = "correlation")

## End(Not run)

vivekbhr/vivlib documentation built on May 3, 2019, 6:13 p.m.