Prosit-XL: enhanced cross-linked peptide identification by fragment intensity prediction to study protein...

Area of Science:

• Proteomics
• Computational Biology
• Biochemistry

Background:

• Peptide spectrum matching (PSM) accuracy is crucial for confident peptide identification in mass spectrometry.
• Traditional scoring methods can be limited, necessitating improved approaches for complex analyses like cross-linked mass spectrometry (XL-MS).

Purpose of the Study:

• To introduce Prosit-XL, a novel tool for predicting fragment intensities in XL-MS.
• To develop a data-driven rescoring approach to improve false discovery rate estimation in XL-MS.
• To enhance the discovery of protein-protein interactions (PPIs) and map interactomes.

Main Methods:

• Developed Prosit-XL, a fragment intensity predictor for cleavable and non-cleavable cross-linkers.
• Implemented a data-driven rescoring strategy utilizing Prosit-XL predictions.
• Validated the approach on synthetic and proteome-scale datasets, including human cytomegalovirus virions.

Main Results:

• Prosit-XL demonstrated high accuracy and consistent performance across diverse datasets.
• The rescoring approach effectively limited the overestimation of the false discovery rate (FDR).
• Achieved up to a 3.4-fold improvement in PPI discovery and increased coverage of a viral interactome map.

Conclusions:

• Prosit-XL significantly enhances peptide identification and PPI discovery in XL-MS.
• The developed rescoring method improves the reliability of XL-MS data analysis.
• Prosit-XL facilitates deeper insights into complex biological systems and viral interactions.