ROCCO: a robust method for detection of open chromatin via convex optimization

Area of Science:

• Genomics
• Epigenetics
• Computational Biology

Background:

• Analyzing open chromatin regions across samples reveals gene regulatory patterns linked to phenotypes.
• Assays like ATAC-seq enable genome-wide open chromatin profiling but lack stable, broadly applicable annotations.
• Current methods independently call peaks per sample and heuristically combine them, challenging large cohort analysis and underutilizing spatial signal features.

Purpose of the Study:

• To develop a novel method for determining consensus open chromatin regions across multiple samples simultaneously.
• To improve the accuracy and efficiency of identifying regulatory elements from high-throughput sequencing data.

Main Methods:

• Proposed ROCCO, a novel method for simultaneous consensus open chromatin region identification.
• ROCCO utilizes robust summary statistics and solves a constrained optimization problem.
• The formulation accounts for both enrichment and spatial dependence of open chromatin signal data.

Main Results:

• ROCCO demonstrates superior empirical performance compared to current methodologies.
• The method exhibits attractive theoretical and conceptual properties for open chromatin analysis.
• Identified consensus regions provide more stable and broadly applicable genomic annotations.

Conclusions:

• ROCCO offers a significant advancement in analyzing open chromatin data from multiple samples.
• The method effectively reconciles sample-specific peak results and leverages spatial information.
• ROCCO provides a robust tool for identifying regulatory patterns associated with biological phenotypes.