GenoScan.SingleWindow: Test the association between an quantitative/dichotomous...

In GenoScan: A Genome-Wide Scan Statistic Framework for Whole-Genome Sequence Data Analysis

Description

Once the preliminary work is done by "GenoScan.prelim()", this function tests a single window. This is often used to double-check significant windows identified by GenoScan.Region or GenoScan.VCF.chr, with an increased number of resampling replicates in GenoScan.prelim.

Usage

GenoScan.SingleWindow(result.prelim,G,Gsub.id=NULL,Z=NULL,MAF.weights='beta',
test='combined',MAF.threshold=1,impute.method='fixed')

Arguments

result.prelim	The output of function "GenoScan.prelim()"
G	Genetic variants in the target region, an n*p matrix where n is the subject ID and p is the total number of genetic variants.
Gsub.id	The subject id corresponding to the genotype matrix, an n dimensional vector. Each ID corresponds to a row in the genotype matrix. This is used to match phenotype with genotype. The default is NULL, where the matched phenotype and genotype matrices are assumed.
Z	Weight matrix for functional annotations, an p*q matrix where p is the total number of genetic variables and q is the number of weights. This is used to incorperate functional annotations. The default is NULL, where minor allele frequency weighted (see MAF.weights) dispersion and/or burden tests are applied.
MAF.weights	Minor allele frequency based weight. Can be 'beta' to up-weight rare variants or 'equal' for a flat weight. The default is 'beta'.
test	Can be 'dispersion', 'burden' or 'combined'. The test is 'combined', both dispersion and burden tests are applied. The default is 'combined'.
MAF.threshold	Threshold for minor allele frequency. Variants above MAF.threshold are ignored. The default is 1.
impute.method	Choose the imputation method when there is missing genotype. Can be "random", "fixed" or "bestguess". Given the estimated allele frequency, "random" simulates the genotype from binomial distribution; "fixed" uses the genotype expectation; "bestguess" uses the genotype with highest probability.

Value

n.marker	Number of tested variants in the window (heterozygous variants below MAF threshold).
p.value	P-value(s) of the window (dispersion p-value(s), then burden p-values(s))

Examples

## GenoScan.prelim does the preliminary data management.
# Input: Y, X (covariates)
## GenoScan.Region scans a region.
# Input: G (genetic variants), pos (position) Z (weights) and result of GenoScan.prelim

library(GenoScan)

# Load data example
# Y: outcomes, n by 1 matrix where n is the total number of observations
# X: covariates, n by d matrix
# G: genotype matrix, n by p matrix where n is the total number of subjects
# pos: positions of genetic variants, p dimention vector
# Z: functional annotation matrix, p by q matrix

data(GenoScan.example)
Y<-GenoScan.example$Y;X<-GenoScan.example$X
G<-GenoScan.example$G;pos<-GenoScan.example$pos
Z<-GenoScan.example$Z

# Preliminary data management
result.prelim<-GenoScan.prelim(Y,X=X,out_type="C",B=5000)

# Scan the region with functional annotations defined in Z
result<-GenoScan.SingleWindow(result.prelim,G,Z=Z)

GenoScan documentation built on May 2, 2019, 12:45 a.m.