SANSparallel

SANSparallel performs rapid protein sequence database searches using the suffix array neighborhood search (SANS) method to identify homologous proteins for functional annotation and evolutionary analysis.

Key Features:

• Parallel client-server SANS: Re-implementation of the suffix array neighborhood search as a client-server architecture enabling parallel processing for accelerated searches.
• Low-latency responses: Produces near-immediate search results compared to servers that may require up to a minute for results.
• Alignment output formats and visualization: Produces lists, pairwise alignments, and stacked alignments and provides outputs compatible with Jalview for alignment visualization.
• Comprehensive database coverage: Supports searches across Uniprot, UniRef90, UniRef50, Swissprot, and the Protein Data Bank.
• Benchmarked sensitivity and speed: Reported to be as sensitive as BLAST for sequences with >50% identity and faster than UBLAST, DIAMOND, LAST, LAMBDA, RAPSEARCH2, and BLAT.

Scientific Applications:

• Protein functional annotation: Provides rapid identification of homologous sequences to support functional annotation pipelines.
• Annotation evidence exploration: Enables retrieval of homologous sequences and alignments that inform evidence supporting specific protein annotations.
• Evolutionary and homology analysis: Facilitates analysis of evolutionary relationships and functional predictions through fast homology searches.

Methodology:

The method uses the suffix array neighborhood search (SANS) algorithm re-engineered for parallel execution in a client-server architecture.