scDIAGRAM: detecting chromatin compartments from individual single-cell Hi-C matrix without imputation or reference...

Area of Science:

• Genomics
• Epigenetics
• Computational Biology

Background:

• Single-cell Hi-C (scHi-C) data reveals 3D genome organization but is sparse and noisy.
• Accurate detection of A/B compartments is vital for understanding chromatin structure and gene regulation.
• Existing methods face challenges with scHi-C data sparsity and noise.

Purpose of the Study:

• To present scDIAGRAM, a data-driven computational method for annotating A/B compartments in single cells.
• To overcome limitations of existing methods in analyzing sparse scHi-C data.
• To enable robust analysis of chromatin compartments at single-cell resolution.

Main Methods:

• scDIAGRAM utilizes direct statistical modeling and graph community detection on individual scHi-C matrices.
• The method infers chromatin compartments without imputation or external reference features.
• A/B labels are assigned using conventional genomic annotations.

Main Results:

• scDIAGRAM demonstrated accuracy and robustness on simulated and real scHi-C datasets.
• The method successfully identified A/B compartments in human cell lines.
• Application to mouse and human datasets revealed compartmental shifts linked to transcriptional variation.

Conclusions:

• scDIAGRAM provides a robust framework for analyzing A/B compartments in single-cell resolution.
• This method offers new insights into the functional roles of chromatin compartments across diverse biological contexts.
• scDIAGRAM advances the study of 3D genome organization and its impact on gene regulation.