LAUSR

Background Intelligence is a fundamental human characteristic, central to social interaction and frequently disrupted in mental illness. A recent GWAS meta-analysis identified 18 loci associated with IQ at genome-wide levels of significance. Adding data from a GWAS of extremely high IQ individuals to this meta-analysis, we integrated localised gene expression data to determine tissue-specific enrichment of genomic signatures in IQ. Methods GWAS data from a population study of 78,308 individuals (Sniekers et al., 2017) was meta-analysed with an extreme-case sampling study of 1,247 individuals with IQ > 145 and 8,185 population controls (total Neff = 82,637). The enrichment of genetic associations to genomic annotations (using partitioned LDScore) and tissue-specific gene-expression levels (using MAGMA and LDScore) was assessed. Tissue-specific annotations were assessed body-wide and between brain tissues, using reference data from GTEx. Predicted effects of genetic variation at IQ-associated loci on gene expression in GTEx brain tissues was assessed using MetaXcan. Increased granularity of enrichment was obtained using brain cell-specific analyses from a large single-cell RNA sequencing reference dataset. Results Genomic loci significant in the IQ GWAS were enriched in regions conserved across mammalian species. Tissue-specific analyses demonstrated significant enrichment of brain-expressed genes in GWAS IQ loci using multiple methods. Between brain tissues, significant enrichment was identified in MAGMA analyses for the frontal cortex. Within the frontal cortex, RNF123 and RBM6 were significantly down-regulated and up-regulated respectively. Cell-specific analyses identified enrichment in pyramidal neurons of the somatosensory cortex and CA1 region of the hippocampus, as well as embryonic GABAergic neurons. Discussion Integrating results from GWAS of IQ with tissue- and cell-specific measures of gene expression implicates pyramidal neurons, and regions of the prefrontal cortex and hippocampus in IQ. One locus identified by GWAS may function to alter the expression of multiple genes, including RNF123, RBM6 and AMT. Depending on the spatial and temporal effect of this locus, this could have downstream consequences for several processes, including glycine metabolism, dendrite development and the immune response. Such functional examination of GWAS results can generate hypotheses for further neurobiological exploration.