Snpchip

Sequenom MassARRAY technology (http://www.sequenom.com/iplex) was employed to regenotype the selected subset of SNPs in the majority of samples previously included in the GWA analysis (n = 96). The regenotyped SNPs were subsequently phased and imputed in the few missing samples (n_cases = 8, n_controls = 3) using Beagle v3.0 [78 (link), 79 (link)], as well as employing a reference dataset comprising of the Illumina SNPChip (see Methods, section “Genotyping and quality control”) and Sequenom MassARRAY regenotyped SNPs, in which variants were pruned based on MAF < 0.001. We subsequently performed a two-step filtering removing SNPs with imputation likelihood lower than empirically defined thresholds. Firstly, SNPs with an allelic squared correlation (R²) value lower than 0.75 were discarded. Secondly, genotypes with imputation probability values lower than 0.8 were labeled as missing. The filtered imputed data were merged with the Illumina SNPChip and regenotyping data produced in previous steps, in order to obtain a comprehensive dataset including all the study samples and all genotyped SNPs. We then performed fine-mapping of the determined genome-wide associated region using GenABEL [65 (link)] with the quality control steps, statistical model, conditional analysis and LD estimation procedures described above.

This replication cohort consisted of 255 healthy volunteers from the publicly available Philadelphia Neurodevelopment Cohort (PNC) [51 (link)] obtained from dbgap (accession number phs000607; mean age 16.0 ± 3.2 [std. dev.]; 125 males, 130 females). Participants were included in this analysis if they had the following: (i) a high-quality structural scan without evidence of significant artifacts based on visual inspection; (ii) high-quality genetic data from an Illumina SNP chip that clustered with the CEU and TSI HapMap3 populations, based on a principal components analysis of all genetic samples; and (iii) no significant past medical or neurological history. MRIs were collected on a 3-Tesla Siemens scanner as described elsewhere [51 (link)]. Genotyping, image processing and analysis were as described above for the NIMH GP cohort structural images. Polygenic-based scores for predicted LIMK1 expression were also computed as above for the NIMH GP cohort.