Humanomniexpress beadchip kit

5262 DACHS samples were genotyped using the whole-genome Illumina CytoSNP assay (Illumina, San Diego, CA, USA) for patients recruited 2003–2007, the Illumina HumanOmniExpress BeadChip Kit for patients recruited 2008–2014, and the Illumina Infinium OncoArray-500K BeadChip for those recruited 2015 or the Infinium Global Screening Array for patients recruited 2016–2017 as described previously [38 (link)]. Missing genotypes were imputed using the Haplotype Reference Consortium as reference panel l (HRC r1.1 2016).

Whole-genome genotyping was performed for all of the subjects with available DNA. The HumanOmniexpress BeadChip Kit (Illumina^®, San Diego, CA, USA) was used to type ~720,000 markers, and the samples were imaged with the Illumina^® iScan following the manufacturer protocol. Quality control was performed as described elsewhere [23 (link)]. A pruning procedure was performed, removing markers with Mendelian errors, minor allele frequency (MAF) < 0.4, and evidence of Hardy–Weinberg disequilibrium (p < 0.00001) and markers with missing genotypes (>5%). A thinning procedure was then performed, to reduce the marker density to five SNPs per cM (centimorgan), obtaining a final set of ~30,000 markers. Given the absence of an a priori disease model, to compute the LOD score, namely the logarithm (base 10) of odds, we used the multipoint non-parametric exponential linkage approach (NPL). Linkage disequilibrium was modeled by clustering SNPs, and haplotypes were generated from di-allelic markers using the Merlin r² option with a threshold of 0.15. We computed the theoretical sample power and the relative maximum expected information content, and applied both the linear and exponential model according to Kong and Cox [24 (link)], with all statistics giving more weight to markers identical by descent (IBD) shared among more than two affected relatives.