MSLibrarian

MSLibrarian optimizes predicted spectral libraries for data-independent acquisition-mass spectrometry (DIA-MS) by calibrating in silico predicted spectra and retention times against spectrum-centric DIA data to improve peptide and protein identification in bottom-up proteomics.

Key Features:

• Optimized Predicted Spectral Libraries: Integrates spectrum-centric DIA data interpretation via the DIA-Umpire method to inform and calibrate in silico spectral library predictions and analysis parameters.
• Enhanced Coverage and Sensitivity: Optimizes intensity prediction parameters and calibrates retention time predictions for varying chromatographic setups to mitigate coverage and sensitivity losses from large or mismatched libraries.
• Library Scope and Sample Representativeness: Optimizes library scope and ensures sample representativeness to improve reliability of peptide and protein detection in complex samples.
• Benchmarking and Validation: Validated with a ground-truth-embedded species-mixed protein experiment showing up to 13% higher peptide identification and up to 8% improved protein-level accuracy while maintaining equivalent false discovery rate (FDR) control and validation criteria.

Scientific Applications:

• DIA-MS-based proteomic profiling: Improves deep, consistent single-shot profiling in bottom-up proteomics by aligning predicted libraries with experimental DIA data.
• Library-free or predicted-library workflows: Enables DIA workflows that reduce reliance on data-dependent acquisition (DDA)-derived spectral libraries for targeted quantification.

Methodology:

Integrates spectrum-centric DIA interpretation (DIA-Umpire) with in silico spectral predictions, calibrates intensity prediction parameters and retention time predictions for different chromatographic setups, optimizes library scope and sample representativeness, and benchmarks performance using ground-truth-embedded species-mixed protein experiments.