Most brain disease-associated and eQTL haplotypes are not located within transcription factor DNase-seq footprints in brain

Abstract

Dense genotyping approaches have revealed much about the genetic architecture both of gene expression and disease susceptibility. However, assigning causality to genetic variants associated with a transcriptomic or phenotypic trait presents a far greater challenge. The development of epigenomic resources by ENCODE, the Epigenomic Roadmap and others has led to strategies that seek to infer the likely functional variants underlying these genome-wide association signals. It is known, for example, that such variants tend to be located within areas of open chromatin, as detected by techniques such as DNase-seq and FAIRE-seq. We aimed to assess what proportion of variants associated with phenotypic or transcriptomic traits in the human brain are located within transcription factor binding sites. The bioinformatic tools, Wellington and HINT, were used to infer transcription factor footprints from existing DNase-seq data derived from central nervous system tissues with high spatial resolution. This dataset was then employed to assess the likely contribution of altered transcription factor binding to both expression quantitative trait loci (eQTL) and genome-wide association study (GWAS) signals. Surprisingly, we show that most haplotypes associated with GWAS or eQTL phenotypes are located outside of DNase-seq footprints. This could imply that DNase-seq footprinting is too insensitive an approach to identify a large proportion of true transcription factor binding sites. Importantly, this suggests that prioritising variants for genome engineering studies to establish causality will continue to be frustrated by an inability of footprinting to identify the causative variant within a haplotype.

Atlas of Transcription Factor Binding Sites from ENCODE DNase Hypersensitivity Data across 27 Tissue Types

“…The aggregate models obtain good performance over a relatively wide range of thresholds compared with the models using individual methods. most haplotypes with cis-acting effects on gene expression (eQTLs) contain SNPs that are located within DNase I hypersensitive regions (Handel et al, 2017). However, DHS regions span a large fraction of the genome, and many SNPs within DHS regions have no evidence for influencing gene expression.…”

“…However, DHS regions span a large fraction of the genome, and many SNPs within DHS regions have no evidence for influencing gene expression. It remains controversial whether footprints more precisely capture the causal variants on eQTL haplotypes: some recent studies found that only a small fraction of eQTL haplotypes overlap footprints (Handel et al, 2017;Moyerbrailean et al, 2016), whereas others have suggested stronger enrichment (Degner et al, 2012;Schwessinger et al, 2017). To address this question, we examined overlap between footprints in our database with eQTLs from the Genotype-Tissue Expression (GTEx) consortium.…”

Open Chromatin Profiling in hiPSC-Derived Neurons Prioritizes Functional Noncoding Psychiatric Risk Variants and Highlights Neurodevelopmental Loci

“…However, postmortem brain data are well known to be confounded by cell heterogeneity and environmental factors (Lipska et al, 2006) and do not capture changes at early neurodevelopmental stages (Brennand et al, 2015). Enrichment analyses of GWAS risk variants for SZ and other brain disorders using postmortem brain open chromatin data (DNase I hypersensitive sites [DHSs]) yield inconsistent findings (Schizophrenia Working Group of the Psychiatric Genomics Consortium, 2014; Handel et al, 2017). The lack of open chromatin profiles in appropriate cellular models has thus been a roadblock in interpreting functional noncoding sequences and GWAS risk variants for neuropsychiatric disorders.…”

Annotation and assessment of functional variants in regulatory regions using epigenomic data in farm animals

“…Although eQTLs are reportedly associated with gene expression ( 35, 36 ), some TF ChIP-seq and DNase-seq studies in humans suggest that more than 50% of eQTLs are not associated with TF binding sites ( 19, 37 ), implying that genomic regulatory features could be used to assess the potential regulatory function of eQTLs. To test this possible use of our Functional Confidence scoring system, we obtained cis-eQTL data from adipose, liver, spleen, hypothalamus, kidney, lung, muscle, and rumen of cattle from the farmGTEx database (https://www.farmgtex.org/).…”