Estimate percent familial risk explained by a list of loci from genome-wide association studies. Input file is a file with columns for annotation, risk allele frequency, and odds ratio for risk allele.

familial_risk_snps.R

# Import command line arguments
args <- commandArgs(trailingOnly=T)
snp_filename <- args[1]
fd_fam_risk <- as.numeric(args[2])

# Import SNP data
snps <- read.table(snp_filename, as.is=T, header=F)
names(snps) <- c("id", "p", "r")
attach(snps)

# Observed familial risk to first degree relatives
print("Familial Risk Estimate (OR):")
fd_fam_risk

# Calculate familial risk due to each locus
snps$lambda <- (p*r**2+(1-p))/((p*r+(1-p))**2)
snps

# Contribution of known SNPs to familial risk
print("Familial Risk Explained (%):")
sum(log(snps$lambda))/log(fd_fam_risk)*100