Identification of differential topologically associating domains from low sequencing depth and pseudobulk chromatin contact maps.

HiDT accurately detects differential topologically associating domains from low sequencing depth data, outperforming existing methods at both TAD and subTAD levels.

• Why it matters: Understanding TAD reorganization is crucial for insights into development, disease, and gene regulation, but current methods struggle with low-depth and pseudobulk chromatin contact maps.
• What they did: The authors developed HiDT, a graph neural network-based algorithm with an attention-based, edge-enhanced layer, trained across various sequencing depths to identify differential TADs.
• The result: HiDT reveals TAD reorganization linked to oncogene dysregulation, transcriptional heterogeneity, and haplotype-specific variations, enabling robust analysis of chromatin structure-function relationships.