Regulome-wide association study identifies enhancer properties associated with risk for schizophrenia

Abstract

Genetic risk for complex traits is strongly enriched in non-coding genomic regions involved in gene regulation, especially enhancers. However, we lack adequate tools to connect the characteristics of these disruptions to genetic risk. Here, we propose RWAS (Regulome Wide Association Study), a new framework to identify the characteristics of enhancers that contribute to genetic risk for disease. Applying our technique to interrogate genetic risk for schizophrenia, we found that risk-associated enhancers in this disease are predominantly active in the brain, evolutionarily conserved, and AT-rich. The association between AT percentage and risk corresponds to an overrepresentation in risk-associated enhancers for the binding sites of transcription factors that recognize AT-rich cis-regulatory motifs. Several of the TFs identified in our model as being overrepresented in risk-associated enhancers, including MEF2C, are master regulators of neuronal development. The genes that encode several of these TFs are themselves located at genetic risk loci for schizophrenia. This list also includes brain-expressed TFs that have not previously been linked to schizophrenia. In summary, we developed a generalizable approach that integrates GWAS summary statistics with enhancer characteristics to identify risk factors in tissue-specific regulatory regions.

Genome-wide significant risk loci for mood disorders in the Old Order Amish founder population

“…MAGMA gene set enrichment analyses were performed with default parameters. We studied 21 gene sets with prior evidence for association with neuropsychiatry, as described previously (56,57). Briefly, these gene sets were derived from the following sources: genes identified through GWAS of MDD(7), BD(8), SCZ(58), and neuroticism(59); genes identified through genetic association studies of rare variants, including exome and genome sequencing studies of SCZ(45), autism spectrum disorders (ASD) (46), or other developmental disorders(60), as well as genes intolerant to loss-of-function mutations(61); genes that are differentially expressed in the prefrontal cortex of individuals with BD, SCZ, or ASD (44); genes that have been identified as targets of the RNA binding proteins FMRP, RBFOX2, RBFOX1/3, and CELF4, of the chromatin remodeling genes CHD8, and of the microRNA miR-137 (57,62); genes localized to synapses from SynaptomeDB (63).…”

“…Next, we asked whether these OOA risk genes overlap genes and gene networks previously implicated in neuropsychiatric disorders, using 21 gene sets derived from psychiatric GWAS, exome sequencing, post-mortem prefrontal cortex gene expression, and analyses of disease-associated gene networks (56,57). We tested both for direct overlap of OOA risk genes with these gene sets, as well as network overlap, in which OOA risk genes interact with established neuropsychiatry genes via protein-protein interactions (Methods).…”