PanTools: representation, storage and exploration of pan-genomic data

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Next-generation sequencing generates numerous highly similar genome sequences.
• Genomics research is transitioning from single-genome to pan-genome analyses.
• Current pan-genomic approaches lack scalability, functionality, and usability.

Purpose of the Study:

• To introduce a novel, scalable, and user-friendly pan-genome representation and analysis tool.
• To facilitate comprehensive comparisons of multiple annotated genomes.
• To address limitations in existing pan-genomic methodologies.

Main Methods:

• Developed a generalized De Bruijn graph for pan-genome representation.
• Implemented an online algorithm for constructing the graph.
• Utilized a Neo4j graph database for scalability with large eukaryotic genomes.
• Created the PanTools software package for various pan-genome analyses.

Main Results:

• Demonstrated the scalability of the approach with large datasets (E. coli, yeast, Arabidopsis thaliana).
• PanTools provides functionalities for annotation, sequence retrieval, gene grouping, and genome reconstruction.
• The graph-based representation effectively captures genomic similarity and divergence.

Conclusions:

• The generalized De Bruijn graph and PanTools offer a robust solution for pan-genome analysis.
• This approach enhances the scalability and usability of comparative genomics.
• PanTools supports a wide range of analyses essential for modern genomics research.