Model-Based Analysis of Quantitative Proteomics Data with Data Independent Acquisition Mass Spectrometry

Abstract

In shotgun proteomics, more abundant peptides are selected for MS/MS fragmentation leading to sequence assignment and their quantitative abundance is computed from peak area of the extracted ion chromatogram. This analysis framework is called data dependent acquisition (DDA). However, the bias towards abundant peptides limits reproducible extraction of peptide signals for a large proportion of the proteome. Recent advances in next generation mass spectrometers enabled implementation of an alternative approach called data independent acquisition (DIA), which improves data quality in terms of dynamic range, measurement precision, and more importantly, reproducible detection. In this chapter, we review the process of generating quantitative proteomics data with DIA, and present a computational tool mapDIA designed for data processing and statistical analysis of the DIA proteomics data. Using an example of renal cancer data set, we demonstrate that fragment intensity data from DIA provide a reliable repeated measure of peptide abundance after careful filtering, and direct modeling of the hierarchical data (protein → peptide → fragment) improves the detection of differentially expressed proteins compared to the analysis using protein intensity data derived by summation of fragment intensities.