Learn R Programming

EBSeqHMM (version 1.6.0)

GetConfidentCalls: Obtain confident gene calls for classifying genes into expression paths

Description

Obtain confident gene calls for classifying genes into expression paths

Usage

GetConfidentCalls(EBSeqHMMOut, FDR=.05, cutoff=0.5, OnlyDynamic=TRUE,Paths=NULL)

Arguments

EBSeqHMMOut
output from EBSeqHMMTest function
FDR
Target FDR, default is 0.05.
cutoff
cutoff to use for defining a confident call. Genes with PP_path greater or equal to cutoff will be called as a confident call. Default is 0.5.
OnlyDynamic
if specifies as T, only dynamic paths will be shown
Paths
paths that are of interest. Default is NULL. If it is not specified, all possible paths will be considered.

Value

Overall: a list of genes/isoforms that are identified as DE under the target FDR, shown are their names and PPs; EachPath: a list object, each sublist contains confident calls (genes/isoforms) that have PP(path)>=cutoff for a particular expression path, shown are their names and PPs; NumEach: length of each sublist in EachPath. EachPathName: gene/isoform names in each of the sublists in EachPath

Details

Function GetConfidentCalls() can be used to obtain a list of DE genes/isoforms with user specific cutoffs. To obtain a list of DE genes/isoforms with a target FDR alpha, the user may specify FDR=alpha. To further choose genes/isoforms with high posterior probability of being its most likely path, the user may specify the option cutoff (default is 0.5). Then genes or isoforms with PP(most likely path ) > = 0.5 will be selected

Examples

Run this code
data(GeneExampleData)
CondVector <- rep(paste("t",1:5,sep=""),each=3)
Conditions <- factor(CondVector, levels=c("t1","t2","t3","t4","t5"))
Sizes <- MedianNorm(GeneExampleData)
EBSeqHMMGeneOut <- EBSeqHMMTest(Data=GeneExampleData, sizeFactors=Sizes, Conditions=Conditions,
          UpdateRd=2)
GeneDECalls <- GetDECalls(EBSeqHMMGeneOut, FDR=.05)
GeneConfCalls <- GetConfidentCalls(EBSeqHMMGeneOut, FDR=.05,cutoff=.5, OnlyDynamic=TRUE)

Run the code above in your browser using DataLab