Most regulatory interactions are not in linkage disequilibrium

Abstract

Linkage disequilibrium (LD) and genomic proximity are commonly used to map non-coding variants to genes, despite increasing examples of causal variants outside the LD block of the gene they regulate. We compared chromatin contacts in 22 cell types to LD across billions of pairs of loci in the human genome and found no concordance, even at genomic distances below 25 kilobases where both tend to be high. Gene expression and ontology data suggest that chromatin contacts identify regulatory variants more reliably than do LD and genomic proximity. We conclude that the genomic architectures of genetic and physical interactions are independent, with important implications for gene regulatory evolution and precision medicine. 2. CC-BY-NC-ND 4.0 International license It is made available under a (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint . http://dx.doi.org/10.1101/272245 doi: bioRxiv preprint first posted online Feb. 26, 2018; Genetic variants ranging from large scale chromosomal rearrangements to single nucleotide polymorphisms (SNPs) can impact gene function by altering exonic sequence or by changing gene regulation. Recent studies estimate that 93% of disease-associated variants are in non-coding DNA [1] and 60% of causal variants map to regulatory elements [2], accounting for 79% of phenotypic variance [3]. Additionally, diseaseassociated variants are enriched in regulatory regions [4], especially those from tissues relevant to the phenotype [5]. Functionally annotating non-coding variants and correctly mapping causal variants to the genes and pathways they affect is critical for understanding disease mechanisms and using genetics in precision medicine [6][7][8][9].Common practice associates non-coding variants with the closest gene promoter or promoters within the same LD block. However, regulatory variants can affect phenotypes by changing the expression of target genes up to several megabases (mb) away [10][11][12][13], well beyond their LD block (median length ≈ 1-2kb, Supplementary Table 1b). This prompted Corradin and colleagues to conclude that a gene's regulatory program is not related to local haplotype structure [14]. Even when a GWAS SNP is in LD with a gene that has a strong biological link to the phenotype, the causal variant may be in a nearby non-coding region regulating a different gene [15,16]. Highlighting the long range of regulatory interactions, recent work in T cells found that only 14% of 684 autoimmune variants targeted their closest gene; 86% skipped one or more intervening genes to reach their target, and 64% of variants interacted with multiple genes [17]. Thus, many non-coding variants are far away and in low LD with the promoters they regulate.Distal non-coding variants can cause changes in gene regulation and phenotypes via three-dimensional (3D) chromatin interactions. For example, the obesity-associated FTO variant (rs1421085) was found to disrupt an ARID5...

Using three-dimensional regulatory chromatin interactions from adult and fetal cortex to interpret genetic results for psychiatric disorders and cognitive traits

“…TADs are defined by 3D chromatin interactions in cell nuclei whereas LD reflects historical events for a given human sample. We confirmed that TADs and LD blocks identify different genomic regions 17,46 . Fifth, we conducted evolutionary analyses for brain FIREs and found that these regions have stronger evidence for ancient and recent positive selection, less population differentiation, and fewer singleton/doubleton single nucleotide variants (Table S5).…”

Defining the Genetic, Genomic, Cellular, and Diagnostic Architectures of Psychiatric Disorders

“…These distinct data types-often derived in different laboratories in different samples-show significant overlap in regulatory predictions (Gusev et al, 2018). This is in contrast to comparisons relying on LD blocks or the assignment by the closest gene, where the overlaps with methods that directly assess chromatin are less substantial (Short et al, 2018;Whalen and Pollard, 2018).…”

“…A complicating factor is that assigning even well-annotated genomic regions to specific genes is not as simple as choosing the closest gene or genes containing variation that is highly correlated with the associated SNPs, which is usually the default approach (Whalen et al, 2016;Won et al, 2016). Rather, as suggested by studies of brain eQTL (Battle et al, 2017;Hauberg et al, 2017) and chromatin structure (de la Torre-Ubieta et al, Won et al, 2016), nearly half of the target genes of human regulatory variation are not in genomic loci defined by linkage disequilibrium (LD) (Whalen and Pollard, 2018) (Table 3). Thus, ''4D mapping'' of chromatin interactions (i.e., brain regions across developmental time) is critical for understanding the functional relationships of regulatory regions to genes (Dekker et al, 2017).…”

Gene-centric functional dissection of human genetic variation uncovers regulators of hematopoiesis

“…Finally, evidence has recently been published that the targets of identified non-coding variants are occasionally not within linkage disequilibrium blocks in the genome (Whalen and Pollard, 2018). This does not necessarily conflict with our results, since we identify hits at only 38 of 75 examined loci and provides an intriguing direction for further work that may elucidate how Our data shows that gene-centric screens are valuable for GWAS follow-up.…”