paperKB
coga / coga-kb
Processing
Help
Sign in

Chunk #7 — Results — 54% of cortex cis-eQTLs have independent associations

Source
Brain expression quantitative trait locus and network analyses reveal downstream effects and putative drivers for brain-related diseases.
Embedded
yes

Text

Within each discovery dataset, we performed a sample size-weighted cis-eQTL meta-analysis on common variants (minor allele frequency (MAF) > 1%), within 1 megabase (Mb) of the transcription start site (TSS) of a protein-coding gene. We identified 1,880 Basal ganglia-EUR, 10,577 Cerebellum-EUR, 4,797 Cortex-AFR, 16,169 Cortex-EUR, 1,265 Hippocampus-EUR and 998 Spinal cord-EUR cis-eQTL genes (q-value < 0.05; Fig. 3a and Supplementary Table 2). The observed eQTLs were consistent between datasets (Supplementary Fig. 8) but showed some sensitivity to RNA-seq alignment strategies (Supplementary Note and Supplementary Figs. 9,10). We next performed conditional analysis to identify independent associations in each cis-eQTL locus. In Cortex-EUR, 8,815 genes had a significant secondary cis-eQTL (54% of cis-eQTL genes identified in this dataset), 4,489 genes had tertiary and 2,065 had quaternary cis-eQTLs. We also identified secondary associations for the other discovery datasets, albeit to a lesser extent (Fig. 3a and Supplementary Tables 2,3). The properties of the Cortex-EUR cis-eQTLs conform to studies performed in blood14 and brain15: primary lead cis-eQTL single nucleotide polymorphisms (SNPs) were generally located close to the TSS (median distance, 33.6 kilobases (kb)) and