High-Coverage Four-Dimensional Data-Independent Acquisition Proteomics and Phosphoproteomics Enabled by Deep Learning-Driven Multi-Dimensional Prediction

• DOI: 10.1101/2022.06.12.495786
• Publication Date: 2022-06-16T04:05:11Z
• Authors: Moran Chen; Pujia Zhu; Pengfei Wu; Yanhong Hao; Zhourui Zhang; Jian Sun; Wenjing Nie; Suming Chen
• Keywords: Phosphoproteomics; Identification

Abstract Four-dimensional (4D) data-independent acquisition (DIA)-based proteomics is an emerging technology that has been proven to have high precursor ion sampling efficiency and higher identification specificity. However, the current 4D DIA still dependent on building of project-specific experimental library which time-consuming limits coverage for identification/quantification. Herein, a workflow by using predicted multi-dimensional in silico was established. A deep learning model Deep4D could high-accurately predict CCS RT both unmodified phosphorylated peptides developed. By integrated containing millions peptides, we identified 25% more protein than libraries analysis HeLa cells. We further demonstrate introduction prediction can greatly complement directly obtained resulting greater increase proteins.