Abstract: Liquid chromatography-mass spectrometry (LC-MS) is the mainstream of high-throughput protein identification technology. Peptide retention time in reversed-phase liquid chromatography (RPLC) is mainly determined by the physicochemical properties of the peptide and the LC conditions (stationary phase and mobile phase). Retention time can be predicted by analyzing these properties and quantifying their effects on peptide chromatographic behavior. Prediction of peptide retention time in LC can be used to improve identification of peptides and post translational modifications (PTM). There are mainly two methods to predict retention time: i.e. retention coefficients and machine learning. The coefficient of determination between observed and predicted retention times can reach 0.93. With the development of LC-MS technology, retention time prediction will become an important tool to facilitate protein identification.

Key words: identification , machine learning, peptide, prediction, protein, retention coefficient, retention time, reversed-phase liquid chromatography-mass spectrometry (RPLC-MS)