Genome-wide mapping and analysis of chromosome architecture

Looping of chromatin fibres is an important mechanism for transcription regulation in animals. The past decade has witnessed an explosion of chromosome conformation capture (3C) technologies aimed at mapping such local or genome-wide chromatin architecture.Key recent methodological advancements for mapping chromatin conformation include improved methods for chromatin fragmentation, proximity ligation, single-cell analysis and targeted 3C. These improvements have propelled the field forward with optimized protocols that enhance the efficiency, scale and resolution of chromatin contact maps.Improved protocols and advances in ultra-high-throughput DNA-sequencing technology have facilitated the rapid accumulation of 3C data sets. The need to extract meaningful insight into hierarchical genome architecture has necessitated the development of novel computational algorithms and bioinformatics pipelines.Accounting for bias in Hi-C and Capture-HiC data is a critical first step towards appropriately analysing their data sets and reaching grounded conclusions. Current methods to account for bias use either explicit or implicit assumption models; however, it is recommended that researchers analyse their data using both approaches to ensure the biological relevance of their findings.Analyses of chromatin contact maps at various resolutions have revealed principles of hierarchical genome architecture, spanning from chromosome territories, compartments and topologically associating domains to contact domains, loops and other important contacts mediated by cis-regulatory elements. Numerous approaches exist for defining each of these features, and the selection of each method should be guided by a full understanding of the statistical model used by each approach.An exhaustive comparison of mapping technologies and analysis methods is sorely needed. To facilitate the evaluation of the 'accuracy' of each method, future efforts should focus on the development of new interaction data standards that consist of loci, the interaction tendencies of which have been rigorously characterized using genetic, biochemical and microscopy approaches.