STAVER: A Standardized Benchmark Dataset-Based Algorithm for Effective Variation Reduction in Large-Scale DIA MS Data

Abstract

Abstract Mass spectrometry-based proteomics has emerged as a powerful tool for the comprehensive investigation of complex biological systems. Data-independent acquisition (DIA) mass spectrometry enables the simultaneous quantification of thousands of proteins, with multi- spectral library search strategies showing great promise for enhancing protein identification and quantification. However, the presence of poor-quality profiles can considerably impact the accuracy of quantitative results, leading to erroneous protein quantification. To address this challenge, we developed STAVER, a standardized benchmark dataset-based algorithm efficiently reduces variation in large-scale DIA MS data. By using the benchmark dataset to standardize mass spectrometry signals, STAVER effectively removes unwanted noise and enhances protein quantification accuracy, especially in the context of multi-spectral library searching. We validated the effectiveness of STAVER in several large-scale DIA datasets, demonstrating improved identification and quantification of thousands of proteins. STAVER represents an innovative and efficacious approach for removing unwanted noise information in large-scale DIA proteome data. It enables cross-study comparison and integration of DIA datasets across different platforms and laboratories, enhancing the consistency and reproducibility of clinical research findings. The complete package is accessible online at https://github.com/Ran485/STAVER.