Journal
NATURE COMMUNICATIONS
Volume 9, Issue -, Pages -Publisher
NATURE PUBLISHING GROUP
DOI: 10.1038/s41467-018-05691-7
Keywords
-
Categories
Funding
- National Basic Research program of China [2014CB340302, 2014CB542002]
- National Nature Science Foundation of China [91540114, 31671342, 31741078, 61772503]
- Special Program for Applied Research on Super Computation of the NSFC-Guangdong Joint Fund
Ask authors/readers for more resources
Submegabase-size topologically associating domains (TAD) have been observed in high-throughput chromatin interaction data (Hi-C). However, accurate detection of TADs depends on ultra-deep sequencing and sophisticated normalization procedures. Here we propose a fast and normalization-free method to decode the domains of chromosomes (deDoc) that utilizes structural information theory. By treating Hi-C contact matrix as a representation of a graph, deDoc partitions the graph into segments with minimal structural entropy. We show that structural entropy can also be used to determine the proper bin size of the Hi-C data. By applying deDoc to pooled Hi-C data from 10 single cells, we detect megabase-size TAD-like domains. This result implies that the modular structure of the genome spatial organization may be fundamental to even a small cohort of single cells. Our algorithms may facilitate systematic investigations of chromosomal domains on a larger scale than hitherto have been possible.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available