EpiAlign: an alignment-based bioinformatic tool for comparing chromatin state sequences.

Area of Science:

• Genomics
• Epigenetics
• Bioinformatics

Background:

• Genome-wide epigenomic datasets offer insights into epigenetic modifications, chromatin structure, and gene expression.
• Existing sequence alignment tools do not address multi-track epigenomic signal comparison.
• A need exists for methods to detect common patterns in epigenomic data for functional or evolutionary insights.

Purpose of the Study:

• To introduce EpiAlign, a novel local alignment algorithm for comparing chromatin state sequences derived from multi-track epigenomic signals.
• To identify locally aligned chromatin regions and recurrent chromatin state patterns.

Main Methods:

• EpiAlign employs a dynamic programming approach.
• The algorithm uniquely accounts for variable lengths and frequencies of chromatin states.
• Validation was performed using extensive simulations and real data from the NIH Roadmap Epigenomics project.

Main Results:

• EpiAlign successfully extracts recurrent chromatin state patterns within single epigenomes, many exhibiting cell-type-specific features.
• The algorithm detects shared chromatin state patterns across multiple epigenomes.
• Demonstrated efficacy in grouping and distinguishing epigenomic samples based on chromatin patterns.

Conclusions:

• EpiAlign is an effective tool for analyzing multi-track epigenomic data.
• It facilitates the discovery of functional and evolutionary relationships through pattern detection.
• EpiAlign aids in the classification and comparison of epigenomic samples.