SPECLUST

SPECLUST clusters peptide peak masses from MALDI-MS mass spectra to identify and group protein isoforms separated by two-dimensional electrophoresis gels.

Key Features:

• Hierarchical clustering: Employs hierarchical clustering to match mass spectra against each other based on extracted peak mass lists.
• Dendrogram organization: Organizes clustered spectra as dendrograms in which protein isoforms group together.
• Cross-sample compatibility: Processes mass spectra from different sample sets, dates, and instruments to enable cross-experimental comparison.
• Shared peak mass identification: Highlights shared peak masses within clusters to create novel peak mass lists for peptide mass fingerprinting.
• Deconvolution of complex spectra: Uses information from other spectra within the same cluster to deconvolute spectra containing multiple proteins.
• Enhanced peak association: Associates a higher number of peptide peaks per protein compared to matching against calculated database masses, indicating the likely presence of modified peptides among shared peaks.

Scientific Applications:

• Protein isoform identification: Supports identification and characterization of protein isoforms from MALDI-MS data of two-dimensional electrophoresis-separated samples.
• Peptide mass fingerprinting support: Produces novel peak mass lists to improve peptide mass fingerprinting-based protein identification.
• Detection of modified peptides and PTMs: Reveals shared peaks that are candidates for modified peptides and post-translational modifications.
• Cross-experimental spectral comparison: Enables detection of spectral similarities across different experimental setups, dates, and instruments.

Methodology:

Hierarchical clustering of spectra based on extracted peak mass lists; generation of dendrograms to group similar spectra; identification of shared peak masses within clusters to produce novel peak mass lists for peptide mass fingerprinting; deconvolution of complex spectra using information from other spectra in the same cluster.