z0_pp: Find PPs for SNPs and null model from Z-scores and MAFs

In corrcoverage: Correcting the Coverage of Credible Sets from Bayesian Genetic Fine Mapping

Description

Posterior probabilities of causality from marginal Z-scores with prior SD as a parameter

Usage

z0_pp(z, f, type, N, s, W = 0.2, p1 = 1e-04)

Arguments

z	Marginal Z-scores of SNPs
f	Minor allele frequencies
type	Type of experiment ('quant' or 'cc')
N	Total sample size
s	Proportion of cases (N1/N0+N1), ignored if type=='quant'
W	Prior for the standard deviation of the effect size parameter, beta (default 0.2)
p1	Prior probability a SNP is associated with the trait (default 1e-4)

Details

Converts Z-scores to posterior probabilities of causality, including the null model of no genetic effects

Value

Posterior probabilities of null model (no genetic effect) and causality for each SNP

Author(s)

Anna Hutchinson

Examples

set.seed(1)
nsnps = 100
N0 = 5000
N1 = 5000
z_scores <- rnorm(nsnps, 0, 3)

## generate example LD matrix and MAFs
library(mvtnorm)
nsamples = 1000

simx <- function(nsnps, nsamples, S, maf=0.1) {
    mu <- rep(0,nsnps)
    rawvars <- rmvnorm(n=nsamples, mean=mu, sigma=S)
    pvars <- pnorm(rawvars)
    x <- qbinom(1-pvars, 1, maf)
}

S <- (1 - (abs(outer(1:nsnps,1:nsnps,`-`))/nsnps))^4
X <- simx(nsnps,nsamples,S)
maf <- colMeans(X)

res <- z0_pp(z = z_scores, f = maf, type = "cc", N = N0+N1, s = N1/(N0+N1))
sum(res)
res

corrcoverage documentation built on Dec. 7, 2019, 1:07 a.m.