Locus-specific Retention Predictor (LsRP): A Peptide Retention Time Predictor Developed for Precision Proteomics
Wenyuan Lu1, Xiaohui Liu2, Shanshan Liu1
1Institutes of Biomedical Sciences and Department of Systems Biology for Medicine, School of Basic Medical, Fudan University, Shanghai, 200032, P. R. China.
Scientific Reports
|March 18, 2017
Summary
A new Locus-specific Retention Predictor (LsRP) precisely forecasts peptide retention times using amino acid locus data and Support Vector Regression. This method improves peptide identification and quantification in liquid chromatography-tandem mass spectrometry (LC-MS/MS) proteomics.
Area of Science:
- Proteomics
- Analytical Chemistry
- Bioinformatics
Background:
- Peptide retention time (RT) prediction is crucial for liquid chromatography-tandem mass spectrometry (LC-MS/MS) based proteomics.
- Accurate RT prediction aids in designing identification and quantification experiments due to liquid chromatography's reproducibility.
Purpose of the Study:
- To develop a Locus-specific Retention Predictor (LsRP) for precise peptide RT prediction.
- To enhance peptide identification and quantification in LC-MS/MS workflows.
Main Methods:
- Developed LsRP using amino acid locus information and the Support Vector Regression (SVR) algorithm.
- Converted peptide sequences into binary locus vectors (zeros and ones).
- Trained and evaluated the SVR model using locus vector data from LC-MS/MS datasets.
Main Results:
- LsRP achieved a prediction correlation coefficient of 0.95–0.99.
- LsRP outperformed two common peptide RT predictors.
- LsRP identified up to 30% more peptides within a 2-minute RT window.
Conclusions:
- LsRP offers precise peptide RT prediction, enhancing LC-MS/MS proteomics.
- A combined strategy of LsRP and calibration peptides presents new opportunities for precision proteomics.
- The locus vector approach provides a robust feature representation for peptide RT prediction.
