Abstract
BackgroundMechanisms underlying genome 3D organization and domain formation in the mammalian nucleus are not completely understood. Multiple processes such as transcriptional compartmentalization, DNA loop extrusion and interactions with the nuclear lamina dynamically act on chromatin at multiple levels. Here, we explore long-range interaction patterns between topologically associated domains (TADs) in several cell types.ResultsWe find that TAD long-range interactions are connected to many key features of chromatin organization, including open and closed compartments, compaction and loop extrusion processes. Domains that form large TAD cliques tend to be repressive across cell types, when comparing gene expression, LINE/SINE repeat content and chromatin subcompartments. Further, TADs in large cliques are larger in genomic size, less dense and depleted of convergent CTCF motifs, in contrast to smaller and denser TADs formed by a loop extrusion process.ConclusionsOur results shed light on the organizational principles that govern repressive and active domains in the human genome.
Highlights
Mechanisms underlying genome 3D organization and domain formation in the mammalian nucleus are not completely understood
Using Armatus [23], we identified a total of 5502–6008 topologically-associated domain (TAD) in each cell line (Additional file 1, Table S2), consistent with our previous findings in primary human adipose stem cells using the same algorithm [21]
We find a systematic enrichment of long interspersed element (LINE) coverage, and correspondingly a depletion of long interspersed element (SINE) coverage, for TADs in cliques compared to TADs outside cliques (Fig. 2 C)
Summary
Mechanisms underlying genome 3D organization and domain formation in the mammalian nucleus are not completely understood Multiple processes such as transcriptional compartmentalization, DNA loop extrusion and interactions with the nuclear lamina dynamically act on chromatin at multiple levels. Increasing resolution, decreasing the bin size of a Hi-C matrix, Based on analysis of the Drosophila genome, highresolution Hi-C data show that compartments of very small sizes can be computed from an eigenvector analysis similar to what has previously been applied on lowresolution Hi-C data [13] These compartments, termed compartment domains, correspond almost perfectly to Liyakat Ali et al BMC Genomics (2021) 22:499 transcription state transitions in the Drosophila genome [13]. The view of mammalian 3D genome organization is becoming increasingly complex, and further classification of the various types of chromatin domains has been suggested [14]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
