The Genotype-Tissue Expression (GTEx) project established a data resource and tissue bank to study the relationship between genetic variation and gene expression in multiple human tissues, collecting whole-genome sequence and RNA-sequence from ~850 deceased adult donors. The Enhancing GTEx (eGTEx) project extends this resource by performing multiple molecular assays on tissue from GTEx donors [1].
As part of eGTEx, we have performed whole-genome bisulfite sequencing (WGBS) on 224 GTEx samples to measure DNA methylation and analyse its relationship with gene expression, genetic variation, and the other molecular marks assayed by eGTEx. Three-quarters of these samples come from 8 brain regions, reflecting the ambition of the eGTEx project to help contextualize results from GWAS of neurological and neuropsychiatric traits, and our group’s interest in neuroepigenetics. Indeed, the eGTEx study design builds upon our recent work studying DNA methylation using WGBS from 72 human samples, covering 4 brain regions and 3 cell populations: nuclei positive for the neuronal marker NeuN (RBFOX3), negative for NeuN, and bulk tissue [2].
Here, we present results from [2] – including analyses of WGBS, ATAC-seq and RNA-seq – alongside the initial results for the 224 WGBS eGTEx samples. We identified tens-to-hundreds of megabases of differentially methylated regions (DMRs) and found that brain region-specific DMRs are almost only found in NeuN+ nuclei. These DMRs have characteristics of enhancers and are highly enriched for heritability of neuropsychiatric disease, such as schizophrenia, as estimated using genome-wide association study data.
Finally, we combined data from [2] with the eGTEx dataset to design capture probes (~46 Mb) for use with a bisulfite-sequencing assay targeting brain-specific DMRs, variably methylated regions (VMRs), and other brain-specific methylation loci. We are using this capture design to assay ~800 GTEx samples to explore how genetic variation influences DNA methylation in the adult human brain.
[1] https://doi.org/10.1038/ng.3969
[2] https://www.biorxiv.org/content/early/2017/03/24/120386