R3CPET

R3CPET infers protein complexes involved in maintaining chromatin interactions from ChIA-PET data using a non-parametric Bayesian approach.

Key Features:

• Non-Parametric Bayesian Approach: Implements a non-parametric Bayesian statistical framework for flexible modeling of cooperating proteins without predefined parameters.
• Integration of Multiple Data Sources: Integrates ChIA-PET data with transcription factor binding sites and known protein-protein interaction networks to improve prediction accuracy.
• Inference of Protein Complexes: Identifies probable sets of protein complexes responsible for specific chromatin interactions and predicts the genomic regions potentially regulated by these complexes.
• Biological Validation and Accuracy: Validates predictions against experimental data and simulation studies to assess biological relevance and accuracy.

Scientific Applications:

• Chromatin architecture analysis: Elucidates cooperating proteins that maintain chromatin interactions to inform studies of chromatin organization.
• Gene regulation: Links inferred protein complexes to genomic regions to support investigation of transcriptional regulatory mechanisms.
• Epigenetics: Provides insights into epigenetic regulation by identifying protein complexes at regulatory chromatin interactions.
• Disease mechanism studies: Supports analysis of how altered chromatin interactions and protein complexes contribute to disease mechanisms.

Methodology:

Applies non-parametric Bayesian inference to integrated ChIA-PET, transcription factor binding site, and protein-protein interaction network data, with predictions compared to experimental datasets and simulation studies.