Movi 2: fast and space-efficient queries on pangenomes

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Compressed indexes are vital for pangenomics to minimize reference bias and enhance efficiency.
• The previous Movi index, while fast, suffered from a large memory footprint.
• Existing compressed indexes often present trade-offs between space and query performance.

Purpose of the Study:

• To introduce Movi 2, a novel compressed index designed to drastically reduce the size and memory requirements of move-structure-based indexes.
• To develop new methods for optimizing compressed indexes for pangenomic data.
• To enable flexible trade-offs between query speed and memory efficiency.

Main Methods:

• Development of new algorithms to decrease the space footprint of move-structure indexes.
• Implementation of sampling strategies to balance query performance and memory usage.
• Evaluation of Movi 2 on large-scale pangenome collections, including the Human Pangenome Reference Consortium (HPRC) data.

Main Results:

• Movi 2 achieves a greater than fivefold reduction in space footprint compared to the original Movi index.
• The new methods significantly reduce the memory footprint of move-structure indexes.
• Movi 2 demonstrates superior performance in both speed and memory efficiency over existing methods, including r-index-based approaches.

Conclusions:

• Movi 2 offers a substantial improvement in space and memory efficiency for compressed pangenomic indexes.
• The developed methods provide a powerful tool for analyzing large and complex genomic datasets.
• Movi 2 represents a significant advancement in bioinformatics for handling large-scale pangenome collections.