The cross-trait LDSC approach (Bulik-Sullivan et al., 2015a, Bulik-Sullivan et al., 2015b) requires only GWAS summary statistics in samples of unrelated individuals and is not biased by sample overlap (Bulik-Sullivan et al., 2015b). The approach replaces the χ2 by the z scores from both studies (i.e., from sample summary statistics) and the genetic covariance is then estimated using the slope from the regression of both z scores on LD scores. Normalizing genetic covariance by SNP heritability yields genetic correlation. Analyses were conducted to estimate the genetic correlation between PTSD and AD using the open-source LDSC pipeline, version 1.0.0 (github.com/bulik/ldsc). Analyses were performed using HapMap3 variants and EA linkage disequilibrium (LD) scores from the 1000 Genomes project (Abecasis et al., 2010). The PGC summary statistic files have previously been processed through a comprehensive quality control pipeline (Sullivan, 2010) including filtering to remove SNPs with imputation information value < 0.90 and minor allele frequency < 0.01. The LDSC pipeline filtering process further removes SNPs based on a minimum N and variants that are either not SNPs or are strand-ambiguous.
