For the cases, genomic DNA was extracted from bloodspots78 (link) and genotyped using the HumanOmni2.5-4 v1 BeadChip (Illumina, San Diego, CA), as previously described56 (link). For the controls, saliva was collected and genotyped by the NIH Center for Inherited Disease Research using the same Human Omni2.5-4 v1 BeadChip methodology, as previously described.
We merged raw intensities data (*.idat) files of cases and controls and carried out joint calling, to reduce batch effects. Clustering and genotyping were performed using GenomeStudio software 2011.1 (Illumina, 2011). Quality control procedures were performed, including filtering by call rate, mismatch between observed and reported gender and possible gender abnormalities as described in Wang et al.56 (link), and tri-allelic and ambiguous (AT/CG) SNPs were removed. Overall, a total of 2,015,750 SNPs were used in our downstream analyses.
We merged raw intensities data (*.idat) files of cases and controls and carried out joint calling, to reduce batch effects. Clustering and genotyping were performed using GenomeStudio software 2011.1 (Illumina, 2011). Quality control procedures were performed, including filtering by call rate, mismatch between observed and reported gender and possible gender abnormalities as described in Wang et al.56 (link), and tri-allelic and ambiguous (AT/CG) SNPs were removed. Overall, a total of 2,015,750 SNPs were used in our downstream analyses.
Free full text: Click here
