Classification based Integration of Quantifications for LC-MS Analysis

Abstract

A classification based integration of quantification method for the Liquid Chromatography – Mass Spectrometry (LC-MS) analysis is described in this paper. Typically, one biological tissue may be sent to the LC-MS many times in practice to generate multiple LC-MS data. Due to the precise level or the profile of the search engine, these multiple individual quantitative results of the multiple LC-MS data may be partially identical. Here we proposed a method to integrate the quantitative results for the case where there are multiple individual measurements but the results are only partially identical. This proposed method applies a classifier to the peptides and treats the predicted probabilities of the classification as the weights to combine these multiple individual quantitative results into a better one. Experimental results show that in the task of quantitative LC-MS, the results generated by this integration method perform better than the ones produced by other individual measurements.