lme.int.batch: function to test gene-environment or gene-gene interactions...

In GWAF: Genome-Wide Association/Interaction Analysis and Rare Variant Analysis with Family Data

Description

Fit linear mixed effects model (LME) to test gene-environment or gene-gene interactions for a continuous phenotype and all SNPs in a genotype file in family data under additive genetic model. The interaction term is the product of SNP genotype and a covariate for interaction (cov.int). The covariate for interaction (cov.int) can be SNP genotype (gene-gene interaction) or an environmental factor (gene-environment interaction). Only one interaction term is allowed. When cov.int is dichotomous, stratified analyses can be requested by specifying sub="Y". The covariance between the main effect (SNP) and the interaction effect is provided in the output when stratified analysis is not requested. The SNP genotype and the interaction are treated as fixed effects, and a random effect correlated according to degree of relatedness within a family is also fitted. In each test for interaction, the lmekin function from package coxme is used.

Usage

lmepack.int.batch(phenfile,genfile,pedfile,phen,kinmat,covars,cov.int,sub="N",
outfile,col.names=T,sep.ped=",",sep.phe=",",sep.gen=",")

Arguments

genfile	a character string naming the genotype file for reading (see format requirement in details)
phenfile	a character string naming the phenotype file for reading (see format requirement in details)
pedfile	a character string naming the pedigree file for reading (see format requirement in details)
outfile	a character string naming the result file for writing
phen	a character string for a phenotype name in phenfile
covars	a character vector for covariates in phenfile
cov.int	a character string naming the covariate for interaction, the covariate has to be included in covars
sub	"N" (default) for no stratified analysis, and "Y" for requesting stratified analyses (only when cov.int is dichotomous)
kinmat	a character string naming the file where kinship coefficient matrix is kept
col.names	a logical value indicating whether the output file should contain column names
sep.ped	the field separator character for pedigree file
sep.phe	the field separator character for phenotype file
sep.gen	the field separator character for genotype file

Details

The lmepack.int.batch function first reads in and merges phenotype-covariates, genotype and pedigree files, then tests gene-environment or gene-gene interaction for phen against all SNPs in genfile. Only one interaction term is allowed, so is the covariate for interaction (cov.int). When cov.int is dichotomous, stratified analyses can be requested by specifying sub="Y". The covariance between the main effect (SNP) and the interaction effect is provided in the output when stratified analysis is not requested. genfile contains unique individual id and genotype data, with the column names being "id" and SNP names. For each SNP, the genotype data should be coded as 0, 1, 2 indicating the numbers of the coded alleles. The SNP name in genotype file should not have any dash, '-' and other special characters(dots and underscores are OK). phenfile contains unique individual id, phenotype and covariate data, with the column names being "id" and phenotype and covaraite names. pedfile contains pedigree informaion, with the column names being "famid","id","fa","mo","sex". In all files, missing value should be an empty space, except missing parental id in pedfile. SNPs with low genotype counts (especially minor allele homozygote) may be omitted. The lmepack.int.batch function fits LME model using lmekin function from coxme package.

Value

No value is returned. Instead, results are written to outfile. If stratified analyses are requested, the result file will include the following columns. Otherwise, cov_beta_snp_beta_int will be included instead of the results from stratified analyses, that is, (beta_snp_cov0, se_snp_cov0, pval_snp_cov0, beta_snp_cov1, se_snp_cov1, and pval_snp_cov1).

phen	phenotype name
snp	SNP name
covar_int	the covariate for interaction
n	sample size used in analysis
AF	allele frequency of the coded allele
model	genetic model used in analysis, additive model only
beta_snp	regression coefficient of SNP covariate
se_snp	standard error of beta_snp
pval_snp	p-value of testing beta_snp not equal to zero
beta_snp_cov0	regression coefficient of SNP covariate in stratified analysis using the subset where cov.int level is 0
se_snp_cov0	standard error of beta_snp_cov0
pval_snp_cov0	p-value of testing beta_snp_cov0 not equal to zero
beta_snp_cov1	regression coefficient of SNP covariate in stratified analysis using the subset where cov.int level is 1
se_snp_cov1	standard error of beta_snp_cov1
pval_snp_cov1	p-value of testing beta_snp_cov1 not equal to zero
beta_int	regression coefficient of the interaction term
se_int	standard error of beta_int
pval_int	p-value of testing beta_int not equal to zero

Author(s)

Qiong Yang <[e-mail]> and Ming-Huei Chen <[e-mail]>

References

coxme package: mixed-effects Cox models, sparse matrices, and modeling data from large pedigrees. Beth Atkinson ([e-mail]) for pedigree functions.Terry Therneau ([e-mail]) for all other functions. 2007. Ref Type: Computer Program http://cran.r-project.org/.

Abecasis, G. R., Cardon, L. R., Cookson, W. O., Sham, P. C., & Cherny, S. S. Association analysis in a variance components framework. Genet Epidemiol, 21 Suppl 1, S341-S346 (2001).

Examples

## Not run: 
lmepack.int.batch(phenfile="simphen.csv",genfile="simgen.csv",pedfile="simped.csv",
phen="SIMQT",kinmat="simkmat.Rdata",outfile="simout.csv",covars=c("age","sex"),
cov.int="sex",sub="Y",sep.ped=",",sep.phe=",",sep.gen=",")

## End(Not run)

GWAF documentation built on May 2, 2019, 2:47 p.m.