snpgdsLDMat: Linkage Disequilibrium (LD) analysis

In SNPRelate: Parallel Computing Toolset for Genome-Wide Association Studies (GWAS)

Description

Return a LD matrix for SNP pairs.

Usage

snpgdsLDMat(gdsobj, sample.id=NULL, snp.id=NULL, slide=250,
	method=c("composite", "r", "dprime", "corr"), num.thread=1, verbose=TRUE)

Arguments

gdsobj	a GDS file object (gds.class)
sample.id	a vector of sample id specifying selected samples; if NULL, all samples are used
snp.id	a vector of snp id specifying selected SNPs; if NULL, all SNPs are used
slide	# of SNPs, the size of sliding window, see details
method	"composite", "r", "dprime", "corr", see details
num.thread	the number of CPU cores used
verbose	if TRUE, show information

Details

Four methods can be used to calculate linkage disequilibrium values: "composite" for LD composite measure, "r" for R coefficient (by EM algorithm assuming HWE, it could be negative), "dprime" for D', and "corr" for correlation coefficient. The method "corr" is equivalent to "composite", when SNP genotypes are coded as: 0 – BB, 1 – AB, 2 – AA.

if slide <= 0, output a n-by-n LD matrix; otherwise, output a m-by-n LD matrix, where n is the total number of SNPs, m is the size of sliding window. For n-by-n matrix, the value of i row and j column is LD of i and j SNPs. For m-by-n matrix, the value of i row and j column is LD of j and j+i SNPs.

Value

Return a list:

sample.id	the sample ids used in the analysis
snp.id	the SNP ids used in the analysis
LD	a matrix of LD values
slide	the size of sliding window

Author(s)

Xiuwen Zheng

References

Weir B: Inferences about linkage disequilibrium. Biometrics 1979; 35: 235-254.

Weir B: Genetic Data Analysis II. Sunderland, MA: Sinauer Associates, 1996.

Weir BS, Cockerham CC: Complete characterization of disequilibrium at two loci; in Feldman MW (ed): Mathematical Evolutionary Theory. Princeton, NJ: Princeton University Press, 1989.

See Also

snpgdsLDpair, snpgdsLDpruning

Examples

library(lattice)

# open an example dataset (HapMap)
genofile <- openfn.gds(snpgdsExampleFileName())

# chromosome 15
snpset <- read.gdsn(index.gdsn(genofile, "snp.id"))[
	read.gdsn(index.gdsn(genofile, "snp.chromosome")) == 15]
length(snpset)

# LD matrix
ld.slide <- snpgdsLDMat(genofile, snp.id=snpset, method="composite")

# plot
levelplot(t(ld.slide$LD^2), col.regions = terrain.colors)


# LD matrix
ld.noslide <- snpgdsLDMat(genofile, snp.id=snpset, slide=-1, method="composite")

# plot
levelplot(t(ld.noslide$LD^2), col.regions = terrain.colors)


# close the genotype file
closefn.gds(genofile)

Example output

Loading required package: gdsfmt
SNPRelate -- supported by Streaming SIMD Extensions 2 (SSE2)
[1] 262
Hint: it is suggested to call `snpgdsOpen' to open a SNP GDS file instead of `openfn.gds'.
Linkage Disequilibrium (LD) estimation on genotypes:
Working space: 279 samples, 262 SNPs
    using 1 (CPU) core.
    sliding window size: 250 
    method: composite
LD matrix:    the sum of all selected genotypes (0,1,2) = 73989
Hint: it is suggested to call `snpgdsOpen' to open a SNP GDS file instead of `openfn.gds'.
Linkage Disequilibrium (LD) estimation on genotypes:
Working space: 279 samples, 262 SNPs
    using 1 (CPU) core.
    method: composite
LD matrix:    the sum of all selected genotypes (0,1,2) = 73989

SNPRelate documentation built on May 2, 2019, 4:56 p.m.