The following scripts are used to run the PGC PTSD EWAS Working Group's cross-sectional analysis based on the methods outlined in Ratanatharathorn et al. (2017). There are two steps to the analysis. First, the PGC_PTSD_EWAS_limma.R script is run by each contributing cohort. The results files from this script is sent to the PGC PTSD EWAS Working Group for meta-analysis, where a a series of meta-analysis scripts are performed. Those scripts are:

1. 01_PGC_PTSD_EWAS_loadData.R - this script loads data from each contributing cohort and saves a combined Rdata file with all the data as well as outputting a table with a summary of each cohort's top results. The example script here includes results for the 10 studies contributing to the main (i.e., smoking not included as a covariate) analysis.

2. 02_PGC_PTSD_EWAS_meta.R - this script loads the data created by the first script and runs the inverse-normal meta-analysis for all CpG sites.

3. 03_PGC_PTSD_EWAS_summary.R - this script takes the meta-analysis results and raw data and creates a QQ plot, Manhattan plot, forest plots of signfiicant sites, and a table of the results. In addition, the script looks at the top CpG sites associated with smoking from the CHARGE consortium.