Systematic inference and comparison of multi-scale chromatin sub-compartments connects spatial organization to cell phenotypes

Abstract

Chromatin compartmentalization reflects biological activity. However, inference of chromatin sub-compartments and compartment domains from chromosome conformation capture (Hi-C) experiments is limited by data resolution. As a result, these have been characterized only in a few cell types and systematic comparisons across multiple tissues and conditions are missing. Here, we present Calder, an algorithmic approach that enables the identification of multi-scale sub-compartments at variable data resolution. Calder allows to infer and compare chromatin sub-compartments and compartment domains in >100 cell lines. Our results reveal sub-compartments enriched for poised chromatin states and undergoing spatial repositioning during lineage differentiation and oncogenic transformation.

Systematic assessment of gene co-regulation within chromatin domains determines differentially active domains across human cancers

“…To match each cohort with Hi-C data derived from the closest possible tissue or cell line, we collected and analyzed 30 Hi-C datasets. Given the distinction between compartment domains and TADs [ 13 ], here we used the Calder algorithm [ 10 ] to infer compartment domains, and TopDom to infer TADs [ 24 ]. Overall, domains identified by these tools were often coincident [ 10 ] and, as we will show, DADo analyses based on either compartment domains or TADs led to largely similar results and conclusions.…”

“…Given the distinction between compartment domains and TADs [ 13 ], here we used the Calder algorithm [ 10 ] to infer compartment domains, and TopDom to infer TADs [ 24 ]. Overall, domains identified by these tools were often coincident [ 10 ] and, as we will show, DADo analyses based on either compartment domains or TADs led to largely similar results and conclusions. Given the consistency observed among these tools, the results discussed in the following will refer to chromatin domains identified by Calder, unless explicitly indicated.…”

“…Next, we computed and compared chromatin compartments (A and B), sub-compartments (n = 8), and compartment domain ranks using the Calder algorithm [ 10 ]. Specifically, Calder assigns a “rank” varying between 0 (most inactive sub-compartment) and 1 (most active sub-compartment) to each domain and genomic bin, and the rank difference between matching bins or domains can be used to assess sub-compartment repositioning [ 10 , 44 ]. Interestingly, we found that differentially active domains were 2-to-3 times more likely to change compartment (A to B or B to A) than other domains (Fig.…”

“…Different chromatin compaction and histone modifications are associated instead with the segregation of the chromatin in major compartments, in particular one characterized by high transcriptional activity (A compartment) and one enriched for heterochromatin and transcriptionally silenced regions (B compartment) [ 4 , 8 , 9 ]. However, chromatin epigenetic states encompass more than two states and recent studies have proposed different numbers of sub-compartments that better captured chromatin epigenetic features [ 4 , 10 ]. Stretches of DNA assigned to the same compartment or sub-compartment have been termed compartment domains [ 10 – 12 ].…”

“…However, chromatin epigenetic states encompass more than two states and recent studies have proposed different numbers of sub-compartments that better captured chromatin epigenetic features [ 4 , 10 ]. Stretches of DNA assigned to the same compartment or sub-compartment have been termed compartment domains [ 10 – 12 ]. Notably, although conceptually distinct, TADs and compartment domains often overlap and/or have coincident boundaries [ 13 ].…”