ATAC-seq has been widely adopted to identify accessible chromatin regions across the genome. However, current data analysis still utilizes approaches initially designed for ChIP-seq or DNase-seq, without considering the transposase digested DNA fragments that contain additional nucleosome positioning information. We present the first dedicated ATAC-seq analysis tool, a semi-supervised machine learning approach named HMMRATAC. HMMRATAC splits a single ATAC-seq dataset into nucleosome-free and nucleosome-enriched signals, learns the unique chromatin structure around accessible regions, and then predicts accessible regions across the entire genome. We show that HMMRATAC outperforms the popular peak-calling algorithms on published human ATAC-seq datasets. We find that single-end sequenced or size-selected ATAC-seq datasets result in a loss of sensitivity compared to paired-end datasets without size-selection.

9 1 0ATAC-seq has been widely adopted to identify accessible chromatin regions across the genome.1 1 However, current data analysis still utilizes approaches initially designed for ChIP-seq or DNase-1 2 seq, without taking into account the transposase digested DNA fragments that contain additional 1 3 nucleosome positioning information. We present the first dedicated ATAC-seq analysis tool, a 1 4 semi-supervised machine learning approach named HMMRATAC. HMMRATAC splits a single 1 5 ATAC-seq dataset into nucleosome-free and nucleosome-enriched signals, learns the unique 1 6 chromatin structure around accessible regions, and then predicts accessible regions across the 1 7 entire genome. We show that HMMRATAC outperforms the popular peak-calling algorithms on 1 8 published human and mouse ATAC-seq datasets.

Background Genetic variants in the human leukocyte antigen (HLA) locus contribute to the risk for developing scleroderma/systemic sclerosis (SSc). However, there are other replicated loci that also contribute to genetic risk for SSc, and it is unknown whether genetic risk in these non-HLA loci acts primarily on the vasculature, immune system, fibroblasts, or other relevant cell types. We used the Cistrome database to investigate the epigenetic landscapes surrounding 11 replicated SSc associated loci to determine whether SNPs in these loci may affect regulatory elements and whether they are likely to impact a specific cell type. Methods We mapped 11 replicated SNPs to haplotypes and sought to determine whether there was significant enrichment for H3K27ac and H3K4me1 marks, epigenetic signatures of enhancer function, on these haplotypes. We queried pathologically relevant cell types: B cells, endothelial cells, fibroblasts, monocytes, and T cells. We then identified the topologically associated domains (TADs) that encompass the SSc risk haplotypes in primary T cells to identify the full range of genes that may be influenced by SSc causal SNPs. We used gene ontology analyses of the genes within the TADs to gain insight into immunologic functions that might be affected by SSc causal SNPs. Results The SSc-associated haplotypes were enriched (p value < 0.01) for H3K4me1/H3K27ac marks in monocytes. Enrichment of one of the two histone marks was found in B cells, fibroblasts, and T cells. No enrichment was identified in endothelial cells. Ontological analyses of genes within the TADs encompassing the risk haplotypes showed enrichment for regulation of transcription, protein binding, activation of T lymphocytes, and proliferation of immune cells. Conclusions The 11 non-HLA SSc risk haplotypes queried are highly enriched for H3K4me1/H3K27ac-marked regulatory elements in a broad range of immune cells and fibroblasts. Furthermore, in immune cells, the risk haplotypes belong to larger chromatin structures encompassing genes that regulate a wide array of immune processes associated with SSc pathogenesis. Though importance of the vasculature in the pathobiology of SSc is widely accepted, we were unable to find evidence for genetic influences on endothelial cell function in these regions.

Juvenile idiopathic arthritis (JIA) is one of the most common chronic diseases in children. While clinical outcomes for patients with juvenile JIA have improved, the underlying biology of the disease and mechanisms underlying therapeutic response/non-response are poorly understood. We have shown that active JIA is associated with distinct transcriptional abnormalities, and that the attainment of remission is associated with reorganization of transcriptional networks. In this study, we used a multi-omics approach to identify mechanisms driving the transcriptional abnormalities in peripheral blood CD4+ T cells of children with active JIA. We demonstrate that active JIA is associated with alterations in CD4+ T cell chromatin, as assessed by ATACseq studies. However, 3D chromatin architecture, assessed by HiChIP and simultaneous mapping of CTCF anchors of chromatin loops, reveals that normal 3D chromatin architecture is largely preserved. Overlapping CTCF binding, ATACseq, and RNAseq data with known JIA genetic risk loci demonstrated the presence of genetic influences on the observed transcriptional abnormalities and identified candidate target genes. These studies demonstrate the utility of multi-omics approaches for unraveling important questions regarding the pathobiology of autoimmune diseases.