GFS

GFS maps peptide mass fingerprints to genomic sequences to enable annotation-independent protein identification.

Key Features:

• Genome-wide theoretical translation and digest scanning: Scans peptide mass fingerprints against the theoretical translation and proteolytic digest of an entire genome sequence.
• Annotation-independent mapping: Operates without predefined open reading frames (ORFs) or existing protein annotations.
• Windowed scoring system: Scores fixed-size windows along the genome using number of matching peptides, missed enzymatic cleavages, presence of in-frame stop codons, peptide adjacency, and duplicate matches.
• Statistical significance assessment: Assesses significance by comparing window scores to scores from windows matched with randomly produced mass data.
• Validation: Validated on Saccharomyces cerevisiae mitochondria and Escherichia coli with 86% concordance to peptident and mascot.
• Annotation error detection: Identifies proteins with incorrect or missing annotations and proteins affected by sequencing-induced framing mistakes.
• Recoding event detection: Detects proteins affected by recoding events such as frameshifting and stop-codon read-through.
• Distributed computation: Implemented as a client-server pair enabling distribution and concurrent analysis across cluster nodes.

Scientific Applications:

• Genome annotation: Locates peptide-supported coding regions independent of existing annotations.
• Proteomics with incomplete databases: Enables protein identification when protein databases lag behind genome sequencing and annotation.
• Annotation correction: Detects and supports correction of incorrect or missing protein annotations.
• Sequencing error analysis: Reveals proteins impacted by sequencing errors that cause framing mistakes.
• Recoding studies: Identifies candidates for frameshifting and stop-codon read-through investigations.

Methodology:

Scans peptide mass fingerprints against the theoretical translation and proteolytic digest of an entire genome; evaluates fixed-size genomic windows by matching peptide counts, missed enzymatic cleavages, in-frame stop codons, peptide adjacency, and duplicate matches; assesses statistical significance by comparing window scores to scores from windows matched with randomly produced mass data; implemented as a client-server pair for distributed concurrent analysis across cluster nodes.