Modular combinatorial binding among human trans-acting factors reveals direct and indirect factor binding.

Area of Science:

• Computational biology
• Genomics
• Molecular biology

Background:

• Combinatorial binding of trans-acting factors (TFs) to DNA is crucial for gene regulation specificity.
• Existing methods struggle to analyze complex regulatory modules involving multiple TFs or assume single-module usage per region.

Purpose of the Study:

• To develop a computational approach for modeling the modular organization of TF combinatorial binding.
• To identify and characterize regulatory modules from in vivo binding data.

Main Methods:

• Utilized a topic model to learn compact and coherent regulatory modules from TF binding data.
• Analyzed TF binding patterns in K562 cells, identifying 49 interpretable modules from 115 TFs.

Main Results:

• Discovered that thousands of regulatory regions utilize multiple modules, a complexity missed by previous methods.
• Validated modules by their association with known protein-protein interactions and chromatin states.
• Identified context-specific co-binding patterns and predicted direct/indirect TF binding with high accuracy (95%).
• Revealed both shared and cell-type-specific regulatory modules across different cell types.

Conclusions:

• The study provides comprehensive cell-type-specific combinatorial binding maps.
• Results suggest a significant modular organization underlying combinatorial TF binding and gene regulation.