Computational protein profile similarity screening for quantitative mass spectrometry experiments

Area of Science:

• Proteomics
• Systems Biology
• Bioinformatics

Background:

• Accurate characterization of protein abundance profiles is crucial for understanding biological processes.
• Mass spectrometry-based quantitative proteomics provides peptide abundance data, requiring sophisticated analysis.
• Existing workflows often neglect the inherent correlation structure of quantitative proteomics data.

Purpose of the Study:

• To develop a novel statistical framework for analyzing quantitative proteomics data.
• To enable robust inference of protein-level similarity from peptide-level measurements.
• To identify proteins regulated by the Anaphase Promoting Complex/Cyclosome (APC/C).

Main Methods:

• Introduced a new distance measure for relative abundance profiles.
• Derived a statistical test for equality of protein abundance profiles.
• Developed a protein-level representation of peptide measurements.
• Utilized the true correlation structure of the data for analysis.

Main Results:

• The new method achieved a 50.9-fold enrichment of co-regulated protein candidates.
• Demonstrated a 2.5-fold improvement compared to an established protein correlation profiling method.
• Successfully identified candidate proteins post-transcriptionally controlled by the APC/C.

Conclusions:

• The developed workflow provides a more powerful and intuitive method for analyzing quantitative proteomics data.
• Accounting for the correlation structure significantly enhances the identification of co-regulated proteins.
• This approach offers a significant advancement in understanding protein regulation, particularly for cell cycle control by the APC/C.