Abstract:The development of Hi-C technology has generated terabytes of chromatin interaction data, which bring possibilities for insight study of chromatin structure. Several studies revealed that mammalian chromosomes are folded into topological associated domains (TADs), which are conserved across cell types. Accurate detection of topological associated domains is now a vital process for revealing the relationship between the structure and function of genome organization. Unfortunately, the current TAD detection methods require massive computing resources, careful parameter adjustment and/or encounter inconsistent results. In this paper, we propose a novel method, Spectral-Based TAD Detector (SBTD), and evaluate its performance with a set of widely accepted statistical methods. We treat the chromatin interaction matrix as a graph and first introduce cosine similarity as a measure of the interaction patterns between bins. The results show that SBTD identifies higher quality TADs than the popular methods (DomainCaller, TopDom and SpectralTAD) and the internal bins of TADs identified by SBTD have higher correlation. Besides, The TADs identified by SBTD show a highly similar histone modification signal enrichment pattern at the boundary as reported in the previous literature. Finally, the motif enrichment analysis shows that compared with the background region, the DNA motifs of known insulator proteins are significantly enriched in the TAD boundary region identified by our method, which proves the high performance of our proposed method. Overall, SBTD is much more effective than existing methods with only one easy-to-adjust parameter, cluster number, for which we provide optimization guidelines.

A comprehensive benchmarking with interpretation and operational guidance for the hierarchy of topologically associating domains

HiTAD: Detecting the Structural and Functional Hierarchies of Topologically Associating Domains from Chromatin Interactions

A Comparison of Topologically Associating Domain Callers Based on Hi-C Data

Comparison of computational methods for the identification of topologically associating domains

SuperTAD: robust detection of hierarchical topologically associated domains with optimized structural information

Deciphering the species-level structure of topologically associating domains

Identification of Topologically Associating Domains (TADs) and Long-distance Enhancer-gene/gene-gene Interactions with Hi-C and HiChIP

SBTD: A Novel Method for Detecting Topological Associated Domains from Hi-C Data

Structural Heterogeneity And Functional Diversity Of Topologically Associating Domains In Mammalian Genomes

CATAD: Exploring Topologically Associating Domains from an Insight of Core-Attachment Structure.

HTAD: a human-in-the-loop framework for supervised chromatin domain detection

Capturing the complexity of topologically associating domains through multi-feature optimization

An Empirical Mode Decomposition-Based Method to Identify Topologically Associated Domains from Chromatin Interactions

Uncovering topologically associating domains from three-dimensional genome maps with TADGATE

optimalTAD: annotation of topologically associating domains based on chromatin marks enrichment

On the existence and functionality of topologically associating domains

A Computational Method to Predict Topologically Associating Domain Boundaries Combining Histone Marks and Sequence Information

JOnTADS: a unified caller for TADs and stripes in Hi-C data

SuperTAD-Fast: Accelerating Topologically Associating Domains Detection Through Discretization

DiffDomain enables identification of structurally reorganized topologically associating domains

Deciphering hierarchical organization of topologically associated domains through change-point testing