R2GCTA: The GCTA approach with bootstrap variance estimate

In hychen-uic/TEV: Total Explained Variation

The GCTA approach with bootstrap variance estimate

Description

This function implements the genetic complex trait analysis of Yang et al (2010) approach assuming independent covariates and normally distributed errors, with bootstrap variance estimate of Schweiger et al (2016).

Usage

R2GCTA(y, x, alpha = c(0.05), niter = 10, bt = NULL, nbt = 1000)

Arguments

y	outcome: a vector of length n
x	covariates: a matrix of nxp dimension
alpha	a vector of type I error for create the confidence intervals.
niter	number of iterations
bt	variable specifying whether to compute bootstrap variance. Default is FALSE
nbt	bootstrap sample size

Details

The function uses the singular value decomposition for estimation and bootstrap sampling approach for variance estimation under normal random errors.

Value

Estimate of proportion of the explained variation, and bootstrap estimate and variance and confidence intervals if bt=T

References

Schweiger, R., Kaufman, S., Laaksonen, R., Kleber, M. E., Marz, W., Eskin, E., Rosset, S., Halperin, E. (2016). Fats and ac-curate construction of confidence intervals for heritability. The American Journal of Human Genetics, 98, 1181-1192.

Yang, J., Lee, S. H., Wray, N. R., Goddard, M. E., Visscher, P. (2016). GCTA-GREML accounts for linkage disequilibrium when estimating genetic variance from genome-wide SNPs. Proceedings of the National Academy of Sciences, 113, E4579-E4580.

Examples

## Not run: R2GCTA(y, x)

hychen-uic/TEV documentation built on Jan. 24, 2025, 9:14 p.m.