Fourth, ENIGMA did not use the “mega-analysis” model adopted by the Psychiatric Genomics Consortium, where all phenotypic and genotype data are sent to a centralized site for analysis. Instead, ENIGMA uses a “meta-analysis” concept: GWAS was run locally using pre-agreed covariates—often with and without adjustments that may affect interpretation (e.g., adjustments for overall head size). Subsequently, and after quality control at each site, the p-values and regression coefficients were combined and meta-analyzed in a way that weights the results based on the sample sizes of each contributing cohort.
