Linking DNA-packing density distribution and TAD boundary locations

Area of Science:

• Genomics
• Biophysics
• Computational Biology

Background:

• DNA is organized into chromatin, which forms topologically associating domains (TADs) with critical regulatory boundaries.
• Understanding the factors that determine TAD boundary locations is crucial for genome regulation.

Purpose of the Study:

• To investigate how DNA-packing density distribution influences TAD boundary locations.
• To develop a predictive model for TAD formation based on polymer physics and DNA accessibility.

Main Methods:

• Developed a polymer-physics-based model of chromatin folding using DNA accessibility data to define DNA-packing density.
• Simulated stochastic folding of heteropolymers within a nucleus to generate conformation ensembles.
• Validated model predictions against Hi-C and FISH experimental data.

Main Results:

• The model successfully reproduced over 60% of human TAD boundaries and spatial distance matrices from FISH experiments.
• DNA accessibility data alone was sufficient to predict TAD dynamics during T cell differentiation.
• Regions with lower DNA-packing density were identified as preferential sites for domain boundary formation.
• The model explained the enrichment of TAD boundaries at CTCF binding sites by CTCF's influence on local DNA-packing density.

Conclusions:

• Established a strong correlation between TAD boundaries and regions of lower DNA-packing density.
• Provided insights into the mechanisms driving TAD formation and their cell-to-cell variability.
• Demonstrated the utility of biophysical modeling for understanding genome organization and regulation.