BIGSI

BIGSI indexes presence and absence of genomic sequences to enable rapid querying and surveillance across large collections of bacterial and viral whole-genome sequence datasets.

Key Features:

• Efficient Data Indexing: Implements a bitsliced genomic signature index data structure that reduced storage requirements by four orders of magnitude and indexed 447,833 bacterial and viral whole-genome sequence datasets.
• Incremental Scalability: The index grows incrementally as new unprocessed or assembled sequence datasets are deposited into archives, supporting expansion to millions of datasets.
• Rapid Query Capabilities: Enables fast presence/absence queries for genes and variant alleles and has been applied to locate resistance genes MCR-1, MCR-2, and MCR-3, determine the host-range of 2,827 plasmids, and quantify antibiotic resistance within archived datasets.
• Computational Integration: Combines principles from microbial population genomics with computational techniques adapted from web search algorithms to produce a searchable genomic index.

Scientific Applications:

• Genomic epidemiology and surveillance: Enables rapid searching across extensive genomic archives to identify genetic markers associated with disease outbreaks.
• Antibiotic resistance monitoring: Facilitates detection and quantification of antibiotic resistance genes such as MCR-1, MCR-2, and MCR-3 across archived datasets.
• Plasmid host-range and gene tracking: Supports determination of plasmid host-range, demonstrated on 2,827 plasmids, and tracking of mobile genetic elements across datasets.

Methodology:

Combines insights from microbial population genomics with computational techniques adapted from web search algorithms to build a bitsliced genomic signature index that enables storage-efficient, incremental indexing and rapid presence/absence queries.