deDGE {edgeR} | R Documentation |
Runs weighted likelihood calculation for moderated estimates of dispersion, and tests for differences in 'tag' abundance between groups
deDGE(object,alpha=500,doPoisson=FALSE,verbose=TRUE)
object |
DGEList containing elements data (matrix: rows-tags, columns-libraries), lib.size group indicating class |
alpha |
weight to put on the individual tag's likelihood |
doPoisson |
logical, whether to fit Poisson model instead of Negative Binomial, default FALSE |
verbose |
logical, whether to write comments, default TRUE |
deDGEList
with elements lr
(likelihood ratio test), r
(estimates of 1/overdispersion), ps
(list containing proportion estimates)
Mark Robinson
Robinson MD, Smyth GK. 'Small-sample estimation of negative binomial dispersion, with applications to SAGE data.' Biostatistics. 2008 Apr;9(2):321-32.
Robinson MD, Smyth GK. 'Moderated statistical tests for assessing differences in tag abundance.' Bioinformatics. 2007 Nov 1;23(21):2881-7.
# generate raw data from NB, create list object y<-matrix(rnbinom(20,size=1,mu=10),nrow=5) d<-DGEList(data=y,group=rep(1:2,each=2),lib.size=rep(c(1000:1001),2)) # find alpha and call main procedure to find differences alpha<-alpha.approxeb(d) ms<-deDGE(d,alpha=alpha$alpha)