glmmHeritability {SAIGEgds}R Documentation

Heritability estimation

Description

Get the heritability estimate from the SAIGE model.

Usage

glmmHeritability(modobj, adjust=TRUE)

Arguments

modobj

an R object for SAIGE model parameters

adjust

if TRUE and binary outcomes, uses adjusted tau estimate for the heritability estimation

Details

In SAIGE, penalized quasi-likelihood (PQL) is used to estimate the variance component parameter tau. It is known to produce biased estimate of the variance component tau using PQL. If adjust=TRUE for binary outcomes, tau is adjusted based prevalence and observed tau using the data in Supplementary Table 7 (Zhou et al. 2018) to reduce the bias of PQL estimate of variance component.

Value

Return a liability scale heritability.

Author(s)

Xiuwen Zheng

See Also

seqFitNullGLMM_SPA

Examples

# open a GDS file
fn <- system.file("extdata", "grm1k_10k_snp.gds", package="SAIGEgds")
gdsfile <- seqOpen(fn)

# load phenotype
phenofn <- system.file("extdata", "pheno.txt.gz", package="SAIGEgds")
pheno <- read.table(phenofn, header=TRUE, as.is=TRUE)
head(pheno)

# fit the null model
glmm <- seqFitNullGLMM_SPA(y ~ x1 + x2, pheno, gdsfile, trait.type="binary")

glmmHeritability(glmm)

seqClose(gdsfile)

[Package SAIGEgds version 1.8.1 Index]