Index and biological spectrum of accessible DNA elements in the human genome

Abstract

DNase I hypersensitive sites (DHSs) are generic markers of regulatory DNA and harbor disease-and phenotypic trait-associated genetic variation. We established high-precision maps of DNase I hypersensitive sites from 733 human biosamples encompassing 439 cell and tissue types and states, and integrated these to precisely delineate and numerically index ~3.6 million DHSs encoded within the human genome, providing a common coordinate system for regulatory DNA. Here we show that the expansive scale of cell and tissue states sampled exposes an unprecedented degree of stereotyped actuation of large sets of elements, signaling the operation of distinct genome-scale regulatory programs. We show further that the complex actuation patterns of individual elements can be captured comprehensively by a simple regulatory vocabulary reflecting their dominant cellular manifestation. This vocabulary, in turn, enables comprehensive and quantitative regulatory annotation of both protein-coding genes and the vast array of well-defined but poorly-characterized non-coding RNA genes. Finally, we show that the combination of high-precision DHSs and regulatory vocabularies markedly concentrate disease-and trait-associated non-coding genetic signals both along the genome and across cellular compartments. Taken together, our results provide a common and extensible coordinate system and vocabulary for human regulatory DNA, and a new global perspective on the architecture of human gene regulation.

Human and rat skeletal muscle single-nuclei multi-omic integrative analyses nominate causal cell types, regulatory elements, and SNPs for complex traits

“…A tightly linked SNP, rs227731 (R 2 = 0.99), had a probability of 0.221; no other SNPs had probability greater than 0.01. SNP rs227731 is not in an ATAC-seq peak in any of the muscle cell types we identified nor is it in any of ENCODE's 1.3 million candidate cisregulatory elements (55,56) or any of the approximately 3.6 million DNaseI hypersensitive sites (DHS) annotated in (57), suggesting that the index SNP rs227727 is indeed the causal SNP. A previous study found that the A allele of rs227727 was associated with higher activity in an allelic luciferase assay in both human fetal oral epithelial cells (GMSM-K) and murine osteoblastic cells (MC3T3) (58).…”

Atlas and developmental dynamics of mouse DNase I hypersensitive sites

“…Among primary samples, we observed the highest number of DHSs in samples of early fetal brain tissue (321,338), and the lowest (34,693) in CD4+ T cells, a highly specialized cell type. From a cumulative total of 24,524,056 DHSs in individual samples, we merged all tissues and developmental stage/time samples to create a comprehensive atlas comprising 1,802,603 distinct DHSs, each of which was identified independently in one or more samples using the Index pipeline (Methods, Supplementary File 1) 17 . Within the atlas, over 467,900 elements (26%) were unique for developing or neonatal samples and were not found in DHSs from 28 adult tissues 1,16 , revealing a large untapped regulatory compartment within developmental stages.…”

“…To address this, we first identified all human variants for each phenotype within the NHGRI GWAS catalog 30 and intersected these with human neural component DHSs from the ENCODE project 17 . As anticipated and previously described 7,27,28,29 , significantly enriched phenotypes included neurological traits and neuropsychiatric disorders comprising, among others, schizophrenia, neuroticism, and attention deficit hyperactivity disorder (Figure 4C).…”

“…[B] Flowchart of the analysis procedure to obtain GWAS-associated TF families. Starting with all unique SNP-trait associations present in the NHGRI-EBI GWAS catalog, we filtered for overlap with neural component DHSs (obtained as described in Meuleman et al, 2019) and TF motifs. This analysis highlighted several neural traits and TF motif families.…”

Global reference mapping and dynamics of human transcription factor footprints

“…DNase I footprints define sites of direct regulatory factor occupancy on DNA and can be 10 used to discriminate sites of direct vs. indirect occupancy within orthogonal data from chromatin 11 immunoprecipitation and sequencing (ChIP-seq) experiments. Cognate transcription factors 12 (TFs) can be reliably assigned to DNase I footprints based on matching to consensus 13 sequences, enabling TF-focused analysis of gene regulation and regulatory networks 9 , and the 14 evolution of regulatory factor binding patterns within regulatory DNA 10 . DNase I is a small 15 enzyme, roughly the size of a typical transcription factor that recognizes the minor groove of 16 DNA and hydrolyzes single-stranded cleavages that, in principle, reflect both the topology and 17 the kinetics of DNA-protein interaction.…”