Statistical analysis of electron transfer dissociation pairwise fragmentation patterns.

Area of Science:

• Proteomics
• Mass Spectrometry
• Analytical Chemistry

Background:

• Electron transfer dissociation (ETD) is an emerging peptide fragmentation technique.
• ETD produces c and z ions, differing from collision-induced dissociation (CID) which yields b and y ions.
• ETD's backbone cleavage selectivity has been debated, with initial assumptions suggesting lower selectivity than CID.

Purpose of the Study:

• To apply statistical data mining, specifically K-means clustering, to ETD spectra.
• To identify and characterize fragmentation patterns in ETD.
• To investigate factors influencing ETD fragmentation selectivity.

Main Methods:

• Utilized K-means clustering, a statistical data mining strategy, on ETD mass spectrometry data.
• Analyzed ETD datasets derived from peptides digested with three distinct proteases.
• Examined fragmentation patterns in relation to protease type, peptide charge states, and amino acid composition.

Main Results:

• Demonstrated that selective cleavages are indeed present in ETD.
• Found that fragmentation patterns are significantly influenced by the protease used, peptide charge states, and amino acid sequences.
• Observed a statistically strong signal for the c(n-1) ion (loss of the C-terminal residue) irrespective of the terminal amino acid, suggesting conformational influence.

Conclusions:

• ETD exhibits predictable fragmentation patterns, challenging earlier assumptions of low selectivity.
• The identified patterns provide a foundation for understanding ETD mechanisms.
• These findings can enhance spectral prediction and improve algorithms for peptide identification using ETD.