ABSSeq (version 1.2.0)

GPTest: Testing the differential expression

Description

Fitting generalized Poisson model and calculating p-value for each gene

Usage

GPTest(object, pPrior=0.3, dcut=0)

Arguments

object

an ABSDataSet object.

pPrior

the parameter for estimating rates of generalized Possion model for each gene, default is 0.30. The predefined parameter for estimating GP rate according to expression level of each gene. It is estimated by reducing the type I error rate in a acceptable level, i.e. 10% for perPrior=0.30 or 5% for perPrior=0.40. For details, please see the references,

dcut

the parameter for fitting GP model, default is 0.

Value

ABSDataSet object with additional elements: pvalue and adjusted p-value, denoted by pvalue and adj-pvalue, respectively. Use the results to get access it.

Details

This function firstly fits the generalized Possion model used absolute differences between two groups, then calculates the pvalue for each gene and finally adjusts the pvalues via BH method.

Examples

obj <- ABSDataSet(counts=simuN5$counts, groups=factor(simuN5$groups))
obj <- normalFactors(obj)
obj <- calPara(obj)
obj <- GPTest(obj)
head(results(obj))