Computes (average) Idenity-by-State for a set of people and markers

ibs.old {GenABEL}R Documentation

Computes (average) Idenity-by-State for a set of people and markers

Description

Given a set of SNPs, computes a matrix of average IBS for a group of people

Usage

ibs.old(data, snpsubset, idsubset, weight="no")

Arguments

data object of snp.data-class
snpsubset Index, character or logical vector with subset of SNPs to run analysis on. If missing, all SNPs from data are used for analysis.
idsubset Index, character or logical vector with subset of IDs to run analysis on. If missing, all people from data are used for analysis.
weight "no" for direct IBS computations, "freq" to weight by allelic frequency

Details

This function facilitates quality control of genomic data. E.g. people with exteremly high (close to 1) IBS may indicate duplicated samples (or twins), simply high values of IBS may indicate relatives.

When weight "freq" is used, IBS for a pair of people i and j is computed as

f_{i,j} = Σ_k \frac{(x_{i,k} - p_k) * (x_{j,k} - p_k)}{(p_k * (1 - p_k))}

where k changes from 1 to N = number of SNPs GW, x_{i,k} is a genotype of ith person at the kth SNP, coded as 0, 1/2, 1 and p_k is the frequency of the "+" allele. This apparently provides an unbiased estimate of the kinship coefficient.

Only with "freq" option monomorphic SNPs are regarded as non-informative.

ibs() operation may be very lengthy for a large number of people.

Value

A (Npeople X Npeople) matrix giving average IBS (kinship) values between a pair below the diagonal and number of SNP genotype measured for both members of the pair above the diagonal.

On the diagonal, homozygosity (0.5+inbreeding) is provided.

Author(s)

Yurii Aulchenko

See Also

check.marker, summary.snp.data, snp.data-class

Examples

data(ge03d2c)
a <- ibs(data=ge03d2c,ids=c(1:10),snps=c(1:1000))
a
# compute IBS based on a random sample of 1000 autosomal marker
a <- ibs(ge03d2c,snps=sample(ge03d2c@gtdata@snpnames[ge03d2c@gtdata@chromosome!="X"],1000,replace=FALSE),weight="freq")
mds <- cmdscale(as.dist(1-a))
plot(mds)
# identify smaller cluster of outliers
km <- kmeans(mds,centers=2,nstart=1000)
cl1 <- names(which(km$cluster==1))
cl2 <- names(which(km$cluster==2))
if (length(cl1) > length(cl2)) cl1 <- cl2;
cl1
# PAINT THE OUTLIERS IN RED
points(mds[cl1,],pch=19,col="red")

[Package GenABEL version 1.6-7 Index]