ChromPolymerDB: a high-resolution database of single-cell 3D chromatin structures for functional genomics.

Area of Science:

• Genomics and Bioinformatics
• Molecular and Structural Biology
• Computational Biology

Background:

• The three-dimensional (3D) chromatin organization is crucial for gene regulation and genomic stability.
• Existing population-averaged assays (e.g., Hi-C) mask crucial single-cell conformational heterogeneity.
• Understanding single-cell 3D genome architecture is vital for deciphering gene expression regulation.

Purpose of the Study:

• To develop a computational framework for deconvoluting bulk Hi-C data into single-cell 3D chromatin conformations.
• To create a comprehensive, publicly accessible database (ChromPolymerDB) of high-resolution single-cell chromatin structures.
• To provide tools for exploring the relationship between 3D genome structure, gene expression, and other omics data.

Main Methods:

• Developed the sequential Bayesian inference framework (sBIF), a polymer physics-based modeling approach.
• Applied sBIF to deconvolve population-averaged Hi-C data to reconstruct individual 3D chromatin structures.
• Curated and organized reconstructed structures into the ChromPolymerDB database with an interactive web interface.

Main Results:

• Generated ChromPolymerDB, containing approximately 108 million reconstructed 5 kb-resolution single-cell structures.
• The database covers over 60,000 genomic loci across 50 human cell types and experimental conditions.
• The platform enables interactive 3D structural analysis, multi-omics integration, and comparative genomics.

Conclusions:

• ChromPolymerDB provides an unprecedented resource for studying single-cell genome architecture and its functional implications.
• The platform facilitates research into the links between 3D chromatin conformation, gene regulation, and cellular processes.
• This work advances comparative 3D genomics by enabling detailed structural analyses across diverse biological contexts.