MoCoLo: a testing framework for motif co-localization

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Genomic feature interactions are key to understanding biological processes.
• Existing methods for analyzing feature co-occurrence have limitations, including fixed window sizes and potential false positives.
• Robust methods for examining genomic feature co-localization and spatial interactions are needed.

Purpose of the Study:

• To develop a novel analytical method for examining feature interactions by assessing their co-localization.
• To introduce the concept of reciprocal co-occurrence and associated statistics and hypotheses.
• To provide a robust approach for inferring co-localization from conditional motif co-occurrence events.

Main Methods:

• Developed a new analytical method based on reciprocal co-occurrence of genomic features.
• Utilized conditional motif co-occurrence events and reverse conditional probabilities.
• Introduced a novel simulation approach to retain motif properties and account for confounders.
• Applied the method, named Motif Co-localization (MoCoLo), to analyze genomic data.

Main Results:

• MoCoLo confirmed the co-occurrence of histone markers in a breast cancer cell line.
• MoCoLo identified significant co-localization of oxidative DNA damage within non-B DNA-forming regions.
• The identified co-localization patterns differed significantly between various non-B DNA structures.

Conclusions:

• MoCoLo provides a robust method for testing spatial interactions between genomic features.
• The method accurately infers co-localization through reciprocal co-occurrence analysis.
• MoCoLo has potential utility in various genomic studies, including cancer research and DNA damage analysis.