getPCA {qsea} | R Documentation |
The getPCA() function performs a Principle Component Analysis (PCA) of the coverage profiles from a qsea object for exploratory data analysis.
getPCA(qs, chr= getChrNames(qs),ROIs, minRowSum=20, keep , norm_method=normMethod(logRPM = c("log", "library_size", "cnv", "preference", "psC10")), topVar=1000, samples=getSampleNames(qs))
qs |
DIPSset (mandatory) |
chr |
chromosomes to consider |
ROIs |
If specified, only windows overlapping ROIs are considered. |
minRowSum |
minimal number of total read counts per window over all samples |
keep |
windows to consider |
norm_method |
name of predefined normalization (e.g. "beta"), or user defined normalization by calling normMethod() function |
topVar |
only the top variable windows are considered |
samples |
names of samples to be considered |
The principle component analysis is calculated using the singular value decomposition (svd).
getPCA() returns a list object, containing the svd and information on the selected windows.
Mathias Lienhard
plotPCA
qs=getExampleQseaSet( repl=5) pca=getPCA(qs, norm_method="beta") colors=c(rep("red", 5), rep("green", 5)) plotPCA(pca, bgColor=colors) #plotPCAfactors is more interesting, if ROIs have been specified in getPCA plotPCAfactors(pca)