SHvr: A robust and efficient LD score regression (LDSC) for...

Description Usage Arguments Value References

Description

We follow the Bulik-Sullivan et al. (2015) approach but use a robust LS estimation without filtering large summary statistics. The input r2 is the averaged reference LD scores (scaled by the total number of SNPs used to compute the LD scores). The weight W is typically based on the LD scores computed using HapMap3 common SNPs. The purpose is to correct over-counting. LDSC is a variance regression (VR) approach: regressing chi-square statistics on LD scores.

Usage

1
SHvr(Z, r2, N, W = NULL)

Arguments

Z

summary Z-statistics for M variants

r2

average reference LD scores for M variants

N

GWAS sample size for each variant (could be different across variants)

W

variant weight

Value

h2

SNP heritability

v0

intercept in the LDSC, related to the genomic control (GC) parameter

References

Bulik-Sullivan,B.K. et al. (2015) LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nature Genetics, 47, 291-295.

Guo,B. and Wu,B. (2018) Principal component based adaptive association test of multiple traits using GWAS summary statistics. bioRxiv 269597; doi: 10.1101/269597


baolinwu/MTAR documentation built on May 14, 2019, 6:03 a.m.