Comparison Study of Data Processing Workflows for Label-Free DIA Quantitative Proteomics

Data processing tools for the data-independent acquisition method, rapidly gaining popularity in proteomics in recent years, are crucially important for this method's efficiency. Quite a number of them, serving both the identification and quantitation of proteins, have been developed over the past decade, yet the issue is still in its infancy, and the research on their comparison and integration is scarce. In this study, we collected four DIA search workflows─DIA-NN, DIA-Umpire, diaTracer, and MSFragger-DIA─as well as four protein quantitation utilities─directLFQ, AlphaPeptStats, Triqler, and simple t test─to evaluate the efficiencies of quantitation workflows integrating these tools in different combinations. The combinations were evaluated in terms of the number of protein identifications, differential expression sensitivity, fold change accuracy, and quantitative false discovery rate control, using two spike-in benchmark data sets. The study provides insights into the compatibility of the evaluated software pairs and practical recommendations for improving the data analysis pipeline for DIA proteomics.