TaxIt

TaxIt performs iterative strain-level identification of microbial organisms from tandem mass spectrometry (MS/MS) spectra for untargeted taxonomic classification.

Key Features:

• Iterative Workflow: Employs an iterative approach that first identifies species candidates using reference sequence data and then refines the search space for strain-level classification.
• Abundance Weighting Strategy: Implements an abundance weighting strategy that prioritizes more abundant peptides to mitigate ambiguities from proteome similarities and increase confidence in taxonomic assignments.
• Automated Strain Sequence Retrieval: Automatically retrieves strain-specific sequences required for low-level classification to expand and refine candidate sets.
• Use of Public Sequence Databases: Leverages public, continuously growing sequence resources such as NCBI to update and extend reference and strain sequence data.

Scientific Applications:

• Bacterial and Viral Strain Identification: Enables strain-level classification of samples of bacterial and viral origin using MS/MS-derived peptide evidence.
• Method Comparison and Benchmarking: Provides untargeted deeper taxonomic classification and has been compared with non-iterative approaches relying on unique peptides or abundance correction, showing consistent correct strain identification across tested examples (one tie).

Methodology:

Initial species identification using reference sequence data, automated retrieval of strain-specific sequences from public databases, and implementation of an abundance weighting strategy to resolve proteome-similarity ambiguities.