PepSplice

PepSplice identifies peptide variants in complex protein samples from tandem mass spectrometry (MS/MS) data by screening extensive peptide search spaces that include non-tryptic peptides, whole-genome sequences, posttranslational modifications, unannotated point mutations, and splice-site–derived peptides using cache-optimized algorithms.

Key Features:

• Cache-efficient algorithms: Employs cache-optimized search algorithms to maximize hardware performance and reduce computation time.
• Comprehensive search space: Simultaneously screens broad peptide search spaces including non-tryptic peptides, whole-genome–derived sequences, multiple posttranslational modifications, unannotated point mutations, and splice sites.
• Controlled search space: Enforces limits on combinations of variations that can co-occur on a single peptide to control combinatorial explosion and maintain computational feasibility.
• Hypergeometric scoring scheme: Applies a hypergeometric scoring method to evaluate and rank candidate peptide-spectrum matches within large datasets.

Scientific Applications:

• Proteomics research: High-throughput identification of peptide variants in complex MS/MS datasets for proteome characterization.
• Posttranslational modification analysis: Detection and evaluation of peptides bearing multiple posttranslational modifications.
• Mutation and splice-site discovery: Identification of unannotated point mutations and alternative splice-derived peptides from genomic or proteomic sources.

Methodology:

The methodology uses cache-optimized search algorithms, enforces limits on co-occurring variation combinations to constrain the search space, and employs a hypergeometric scoring scheme to evaluate matches on tandem mass spectrometry (MS/MS) spectra; it has been applied to datasets containing over 1.4 million spectra.