fitComputeVPC.lmer: Fit linear mixed models (LMM) and compute the VPC values for...

In KechrisLab/HeritSeq: Heritability of Gene Expression for Next-Generation Sequencing

Description

Fit the Gaussian-like data to LMM and compute the VPC values for one or more features.

Usage

fitComputeVPC.lmer(CountMatrix, Strains, PriorWeights = NULL, test = FALSE,
  VPCname = "LMM")

Arguments

CountMatrix	Sequencing count matrix for one or more features. Each row is for one feature, and the columns are for samples.
Strains	Strain labels for the samples.
PriorWeights	Weights used in the lmer function in the package lme4. It is an optional vector used in the fitting process.
test	TRUE or FALSE (default). Test the presence of heritability through examining the random effect variance .
VPCname	Name of the VPC result, default = "LMM".

Value

A list with two objects. The first object is a 1 x G vector indicating the variance partition coefficients (VPC). If the argument test is set to be true, the second object of the list consists of p-values for testing the hypothesis that random effects sigma_a2 = 0; otherwise, the second object is NULL.

Examples

## Compute VPC for the first two features under linear mixed models for Gaussian-like datasets. 

## Provide normalized data and include hypothesis testing on presence of
## heritability:
result.vst <- fitComputeVPC.lmer(simData_vst[1:2,], strains, test = TRUE)
## Extract parameters
vpc.vst <- result.vst[[1]]
## Extract p-values
pval.vst <- result.vst[[2]]

## Visulize the distribution of p-values.
hist(pval.vst, breaks = 30, col = "cyan")

KechrisLab/HeritSeq documentation built on May 8, 2019, 4:48 p.m.