Hi-BDiSCO: folding 3D mesoscale genome structures from Hi-C data using brownian dynamics

Area of Science:

• Genomics
• Computational Biology
• Biophysics

Background:

• Eukaryotic genome structure and dynamics are crucial for gene regulation.
• Chromosome conformation capture techniques like Hi-C provide genome-wide contact data.
• Existing 3D genome reconstruction methods often use low-resolution polymer models.

Purpose of the Study:

• To develop a novel method for reconstructing high-resolution 3D genome structures.
• To integrate chromatin simulation with 3D structure prediction from Hi-C and Micro-C data.
• To enable detailed analysis of chromatin organization at nucleosome resolution.

Main Methods:

• Utilized a Brownian Dynamics (BD) approach named Hi-BDiSCO.
• Employed a mesoscale-resolution chromatin model based on Discrete Surface Charge Optimization (DiSCO).
• Integrated reconstruction with chromatin simulations at nucleosome resolution using biophysical parameters.

Main Results:

• Successfully generated 3D genome structures from Hi-C and Micro-C data.
• Demonstrated the method's application on mouse genes (NXN, HOXC, HOXA, Fbn2) ranging from 50-100 kb.
• Achieved nucleosome-resolution structural predictions.

Conclusions:

• Hi-BDiSCO provides a powerful tool for high-resolution 3D genome structure reconstruction.
• Nucleosome-resolution structures facilitate deeper understanding of epigenomic regulation.
• The approach has significant potential for biomedical applications in human disease research.