Cross-Run Hybrid\nFeatures Improve the Identification\nof Data-Independent Acquisition Proteomics

The analysis of data-independent\nacquisition (DIA) mass\nspectrometry\ndata is crucial for comprehensive proteomics studies. However, traditional\nsingle-run methods often fall short in terms of identification depth\nand consistency. We present HFDiscrim, a specialized multirun DIA\nanalysis tool aimed at enhancing the depth and consistency of reliable\npeptide identifications of DIA analysis tools. HFDiscrim was extensively\nbenchmarked on multiple data sets, including the MCB data set, the\nccRCC data set, and a three-species benchmark mixture. Compared to\nPyProphet, HFDiscrim identified 22.04% more precursors, 19.1% more\npeptides, and 13.2% more proteins while maintaining a controllable\nfalse discovery rate. Furthermore, HFDiscrim demonstrated higher identification\nrates and improved reproducibility across multiple runs. HFDiscrim\nis publicly available at https://github.com/yachliu/HFDiscrim.