GSAgm (version 1.0)

PCgamma: PC gamma

Description

For GSA of SNP data, the following two-step procedure is implemented (see Biernacka et al[1] for more details on the method). Step 1: Principal components analysis for SNPs within a gene is completed with the components needed to explain 80 percent of the variation retained. Using these components, a gene-level association test is completed to determine the association of the gene with the phenotype. Step 2: The gene-level p values for genes within a given gene set are combined using the Gamma Method, a variation of Fisher's Method, to determine the association of the gene set with the phenotype. The GSA function for SNP data allow quantitative, binary and time-to-event phenotypes (i.e., linear models, logistic models, Cox proportional hazard models).

Usage

PCgamma(formula,data,snpprefix="snp",gene,PCpctVar = 80,
gammaShape = 1, STT=NULL, pheno.type = c("case.control", "quantitative", "survival"),
perm = T, n.perm = 1000, seed = 12212012)

Arguments

formula

formula for model, include phenotype and covars. SNPs will be added by function

data

All data including matrix of genetic markers, each marker represented by the dosage of some allele, could also be CNV, treated as continuous and covariates

snpprefix

prefix for SNP variable, defaults to "snp"

gene

vector disignating the gene each marker belongs to, must be in same order as SNPs

PCpctVar

numeric indicating the percent of variation (in percent) in the genetic markers that is to be explained by PCs

gammaShape

numeric indicating the gamma shape parameter to be used for p-value summarization

STT

numeric indicating soft truncation threshold to be used, will calculate gamma parameter (must be

pheno.type

type of phenotype, case-control results in logistic regression, quantitative results in OLS, and survival results in cox model

perm

boolean indicating whether permutation p-value are to be used for the gamma summary method

n.perm

numeric indicating number of permutations to be used

seed

numeric to set RNG for reproducability

Value

This functions returns a list.

gamma.pvalue

Gamma P value

perm.pvalue

Gamma permutation p value, if specified. Else NA

gene.info

Info for each gene

Examples

###Case Control (logistic) example
data(testdata)
data(gene_example)
PCgamma(pheno~strata(study)+age,
       data=testdata,gene=gene_example,pheno.type="case.control",
       STT = 0.2, gammaShape = NULL,
       perm=FALSE, n.perm = 10, seed = 12212012)

##Here is a survival example
set.seed(1234)
time_example <- rnorm(150, m=50, sd=10)  
event_example <- rbinom(150, 1, 0.3)
testdata <- cbind(testdata,time_example,event_example)

PCgamma(Surv(time_example,event_example)~strata(study)+age,
       data=testdata,gene=gene_example,pheno.type="survival",
       STT = 0.2, gammaShape = NULL,
       perm=FALSE, n.perm = 10, seed = 12212012)