GAWMerge is a protocol that we developed to integrate array and WGS genotyping technologies that minimizes false positives while discovering true association signals. Details of the protocol development process are provided in the Methods section. The final protocol consists of eight major steps (Fig. 1): (1) select control dataset(s) with WGS genotype data; (2) extract the SNPs from the WGS data of the control samples that match those for the array-genotyped case samples; (3) independently subject the case and control samples to the same quality control (QC) procedure (further details in the Methods); (4) phase the case and control samples with the same software (further details in the Methods); (5) merge the phased case and control data and impute to the desired reference genome (e.g., 1000Genome, TOPMed reference panel); (6) filter out genotyped SNPs with low quality (empirical ER2 < 0.9)19 and re-impute; (7) test SNP associations with phenotype of interest in case and control samples combined; and (8) filter association results for minor allele frequency (MAF), imputation quality (R2), and difference in imputation quality.
