MTSv

MTSv performs alignment-based taxonomic classification of metagenomic and high-throughput sequencing reads to assign taxa with high confidence.

Key Features:

• FM-index assisted q-gram filtering: Uses an FM-index assisted q-gram filter to rapidly identify candidate reference sequences for alignment.
• SIMD-accelerated Smith-Waterman: Performs local sequence alignment using a SIMD (Single Instruction, Multiple Data) accelerated Smith-Waterman algorithm for high-quality alignments.
• Selective alignment strategy: Ceases further alignment attempts for a TaxID once an alignment of sufficient quality has been identified to reduce computational overhead.
• Full-alignment precision: Conducts full alignments to improve confidence and reduce false positives relative to exact k-mer matching approaches.
• Resource configurability: Can be tailored to operate under memory- or processor-constrained environments.

Scientific Applications:

• High-confidence taxonomic classification: Assigns taxonomy for metagenomic studies where accurate taxonomic resolution is required.
• Pathogen detection: Detects pathogens in clinical and environmental samples with minimal off-target assignments.
• Microbial community analysis: Summarizes microbial community composition by accurately assigning taxonomy to metagenomic reads.

Methodology:

MTSv applies an FM-index assisted q-gram filter to identify candidate sequences, performs SIMD-accelerated Smith-Waterman local alignments, stops further attempts per TaxID after a sufficient-quality alignment is found, performs full alignments instead of exact k-mer matching, and can be configured for memory or processor constraints.