This composite shows TAD domains ("contact domains") called by the ENCODE uniform Hi-C pipeline across 117 human biosamples on hg38. Contact domains are regions that preferentially self-interact, called by the Arrowhead algorithm (Juicer). Each subtrack is one biosample, browsable with a faceted selector (filter by organ, biosample type, assay, life stage, and call type).
ENCODE provides contact domains for many human biosamples across hundreds of files. This track shows one representative experiment per biosample; by default a set of 14 reference biosamples is displayed (GM12878, K562, HepG2, HCT116, IMR-90, A549, HL-60/S4, heart left/right ventricle, dorsolateral prefrontal cortex, ovary, pancreas, transverse colon, and motor neuron), and the remaining biosamples are turned off and reached through the faceted selector. The complete set is available at the ENCODE portal.
Each domain is drawn as a box, and subtracks are colored by organ. Mousing over a domain shows the biosample and the Arrowhead corner score (the likelihood that the spot is a contact-domain corner); the details page also reports Arrowhead's upper/lower variance and sign scores. These calls use a different algorithm and a finer resolution (5 kb, calling sub-TAD contact domains) than the other TAD tracks, and are not directly comparable to them.
Use the faceted selector on the track configuration page to choose biosamples by Organ, Biosample type (cell line, tissue, primary cell, in vitro differentiated), Assay (intact or in situ Hi-C), Life stage, and Calls. Where a biosample had both an untreated baseline and a perturbed (treated or genetically modified) experiment, the baseline was chosen; a handful of biosamples that exist only in a stimulated state (e.g. activated immune cells) are named accordingly. The Calls facet distinguishes the 112 biosamples with native hg38 Arrowhead calls ("Arrowhead (hg38)") from 5 biosamples (LNCaP clone FGC, NCI-H460, RPMI7951, SJCRH30, SK-N-MC) for which ENCODE provides only domains lifted from hg19 ("Lifted from hg19"); the lifted domains carry no Arrowhead scores.
Contact domains were produced by the ENCODE uniform Hi-C processing pipeline (developed by the Aiden lab, built on Juicer), which calls domains with Arrowhead on GRCh38. For each biosample, one representative experiment was selected (preferring native hg38 Arrowhead calls, then an untreated baseline experiment over a perturbed one where both were available, then the deepest experiment by assay type and data volume), and that experiment's contact-domain files were pooled: each Juicer/Arrowhead paired-anchor BEDPE record (in which both anchors describe the same domain interval) was reduced to a single domain interval, the five Arrowhead scores were retained, and replicate calls whose endpoints fell within one 5 kb bin were merged, keeping the call with the higher corner score. The chosen ENCODE experiment accession is given in each subtrack's long label.
The raw data can be explored interactively with the Table Browser or the Data Integrator. For programmatic access, the track can be accessed using the Genome Browser's REST API. The underlying bigBed files can be downloaded from our download server.
ENCODE Project Consortium. An integrated encyclopedia of DNA elements in the human genome. Nature. 2012;489(7414):57-74. doi:10.1038/nature11247
Rao SS, Huntley MH, Durand NC, Stamenova EK, et al. A 3D map of the human genome at kilobase resolution reveals principles of chromatin looping. Cell. 2014;159(7):1665-80. doi:10.1016/j.cell.2014.11.021
Durand NC, Shamim MS, Machol I, Rao SS, Huntley MH, Lander ES, Aiden EL. Juicer Provides a One-Click System for Analyzing Loop-Resolution Hi-C Experiments. Cell Syst. 2016;3(1):95-8. doi:10.1016/j.cels.2016.07.002