EDDA (version 1.10.0)

testDATs: Run differential abundance testings

Description

Perform differential abundance testing on simulated count data.

Usage

testDATs(data, numCores=10, minCountsThreshold=0,
	DE.methods=c("Cuffdiff","DESeq","baySeq","edgeR","MetaStats","NOISeq"), 
	nor.methods=c("default","Mode","UQN","NDE"),method.list=NULL)

Arguments

data

Data object from generateData() function or predifined data object similar to the output of generateData().

numCores

Number of cores for parallelization. Default is 10.

minCountsThreshold

Minimum counts threshold for filtering. Default is 0.

DE.methods

Method list for differential expression tests. Methods currently available include "Cuffdiff","DESeq","baySeq","edgeR","MetaStats","NOISeq".

nor.methods

Normalization method list. Methods currently available include "default" (default normalization for each DE method),"Mode"(Mode normalization), "UQN"(Upper quartile normalization), "NDE"(non-differential expression normalization).

method.list

The method list for the combination of DE.methods and nor.methods. Default is NULL.

Value

data

Data object from generateData() function.

filterCounts

filtered count data.

Cuffdiff

Result form Cuffdiff with default normalization.

Cuffdiff_uqn

Result form Cuffdiff with Upper quartile normalization normalization.

Cuffdiff_Mode

Result form Cuffdiff with Mode normalization.

Cuffdiff_nde

Result form Cuffdiff with non-differential expression normalization.

DESeq

Result form DESeq with default normalization.

DESeq_uqn

Result form DESeq with Upper quartile normalization normalization.

DESeq_Mode

Result form DESeq with Mode normalization.

DESeq_nde

Result form DESeq with non-differential expression normalization.

baySeq

Result form baySeq with default normalization.

baySeq_uqn

Result form baySeq with Upper quartile normalization normalization.

baySeq_Mode

Result form baySeq with Mode normalization.

baySeq_nde

Result form baySeq with non-differential expression normalization.

edgeR

Result form edgeR with default normalization.

edgeR_uqn

Result form edgeR with Upper quartile normalization normalization.

edgeR_Mode

Result form edgeR with Mode normalization.

edgeR_nde

Result form edgeR with non-differential expression normalization.

MetaStats

Result form MetaStats with default normalization.

MetaStats_uqn

Result form MetaStats with Upper quartile normalization normalization.

MetaStats_Mode

Result form MetaStats with Mode normalization.

MetaStats_nde

Result form MetaStats with non-differential expression normalization.

NOISeq

Result form NOISeq with default normalization.

NOISeq_uqn

Result form NOISeq with Upper quartile normalization normalization.

NOISeq_Mode

Result form NOISeq with Mode normalization.

NOISeq_nde

Result form NOISeq with non-differential expression normalization.

References

Luo Huaien, Li Juntao,Chia Kuan Hui Burton, Shyam Prabhakar, Paul Robson, Niranjan Nagarajan, The importance of study design for detecting differentially abundant features in high-throughput experiments, under review.

Examples

data <- generateData(EntityCount=100)
test.obj <- testDATs(data,nor.methods="default")
test.obj <- testDATs(data,DE.methods="DESeq")


# test data with input count.
x <- matrix(rnbinom(1000*15,size=1,mu=10), nrow=1000, ncol=15);
x[1:50,11:15] <- x[1:50,11:15]*10
x.name=paste("g",1:1000,sep="");
write.table(cbind(x.name,x),"count.txt",row.names =FALSE, sep ='\t')

x <- read.table("count.txt",head=TRUE,sep='\t')
x.count <- x[,2:16]
x.lable=c(rep(0,10),rep(1,5))
row.names(x.count) <- x[,1]
data <- list(count=x.count,dataLabel=x.lable)
test.obj <- testDATs(data,DE.methods=c("DESeq","edgeR"),nor.methods="default")