discriminCNV: Discrimination Analysis for CNV data matrices

Description Usage Arguments Value References Examples

Description

An ade4 wrapping function to performs a Discriminat Analysis based on multivariate dimensionality reduction such as multiple correspondenc analysis, correspondence analysis or principal component analysis. Computes the correlation of each variable to the axes defined by each population in the principal component subspace.

Usage

1
2
3
4
5
6
discriminCNV(mat.f, pop.cla, select = NULL, type = "acm", verbose = TRUE, ...)

## S3 method for class 'cnv.discrimin'
print(x, ...)
## S3 method for class 'cnv.discrimin'
plot(x, all=FALSE, xax = 1, yax = 2, ...)

Arguments

mat.f

data.frame typically obtained from the matrix reduction performed by reduceMatrix or getReducedData

pop.cla

factors vector with population labeling of each subject. The subject entries must match the entries for the rows of mat.f

select

list of CNVs to be considered in the discrimination

type

for dimensionality reduction "acm" for multiple correspondence analysis "coa" for correspondence analysis "pca" for principal component analysis

verbose

whether to print detailed progress on R prompt.

x

an object of class 'cnv.discrim'

all

similar to 'plot.discrim' of ade4 package. The default is FALSE

xax

number of component plotted in the x-axis. Default is 1st

yax

number of component plotted in the y-axis. Default is 2nd

...

additional parameters for ade4 fucntions

Value

Return a list of class cnv.discrimin with the result of the discrimination and the correlation between the CNVs and the directions of all centroid axes.

proj

correlation of each CNV onto the centroid directions

cen

centroid directions

mat.f

matrix input

pop

population labeling -input

References

Caceres A, Basagana X, Gonzalez JR: Multiple Correspondence Discriminant Analysis: An Application to Detect Stratication in Copy Number Variation 2009. [ISCB Prague, Abstract: S05.03, submitted to the special issue of Statistics in Medicine].

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
## Not run: 

# 'HapMap270reducedData.RData' file can be downloaded from ...

load("HapMap270reducedData.RData")
cp<-discrimin.cnv(mat.f,pop.cla)
plot(cp)


## End(Not run)

gada documentation built on May 2, 2019, 6:10 p.m.