Repitools (version 1.18.0)

abcdDNA: A wrapper for fitting the offset-adjusted ABCD-DNA GLM

Description

This function performs differential analyses, given a QdnaData object with the sample-specific offsets already calculated (i.e. call getSampleOffsets before calling abcdDNA), a coefficient (or set of coefficients) to test and dispersion(s). In essence, the function is a wrapper for constructing the offset matrix, fitting the generalized linear model and performing a likelihood ratio test.

Usage

abcdDNA(obj, coef = ncol(obj$design), dispersion = NULL)

Arguments

obj

a QdnaData object

coef

coefficient (or coefficients) of the design matrix to test

dispersion

estimate(s) of dispersion to use for negative binomial testing

Value

a DGEGLM (see the edgeR package) containing the results of the differential comparison

Details

This function is simply a wrapper for taking the details in an QdnaData object and perform the differential analyses, adjusting for copy number if specified.

References

http://imlspenticton.uzh.ch/robinson_lab/ABCD-DNA/ABCD-DNA.html

See Also

QdnaData,

Examples

# library(Repitools)
# qd <- QdnaData(counts=counts, regions=gb, design=design, 
#                cnv.offsets=cn, neutral=(regs=="L=4 P=2"))
# qd <- getSampleOffsets(qd,ref=1)
# plotQdnaByCN(qd,cnv.group=regs,idx.ref=3,idx.sam=2)
# f <- abcdDNA(qd, dispersion=.05, coef=2)
# topTags(f)