Taxonomic-Level Protein Quantification in Metaproteomics Using a Biomass-Constrained Expectation-Maximization...

Area of Science:

• Microbiology
• Proteomics
• Bioinformatics

Background:

• Microbiome communities are crucial for ecosystem function and human health.
• Metaproteomics enables direct identification and quantification of microbial proteins.
• The shared peptide problem complicates accurate taxon-protein quantification in metaproteomics.

Purpose of the Study:

• To enhance the Microorganism Classification and Identification (MiCId) workflow by improving taxon-protein quantification.
• To address the shared peptide problem in mass-spectrometry-based metaproteomics.
• To enable more accurate representation of taxonomic-level proteomes.

Main Methods:

• Extended a modified expectation-maximization algorithm with taxonomic biomass constraints.
• Quantified taxon-protein pairs using clustered identified pairs.
• Evaluated performance using synthetic and clinical human stool microbiome datasets.

Main Results:

• Fold changes in simple synthetic datasets closely matched expected values.
• The algorithm accurately redistributed peptide counts among taxon-protein pairs sharing peptides in a 24-species dataset.
• MiCId demonstrated accurate and consistent results with previous findings in a clinical stool microbiome dataset.

Conclusions:

• The enhanced MiCId algorithm robustly quantifies taxon-protein pairs in complex microbial communities.
• Resolving the shared peptide problem advances the application of metaproteomics in microbiome research.
• The method enables accurate representation of taxonomic-level proteomes.