Xinpei’s paper Deep Learning Prediction Boosts Phosphoproteomics-Based Discoveries Through Improved...

Xinpei’s paper Deep Learning Prediction Boosts Phosphoproteomics-Based Discoveries Through Improved Phosphopeptide Identification has been published in Molecular & Cellular Proteomics. Shotgun phosphoproteomics enables high-throughput analysis of phosphopeptides in biological samples, but low phosphopeptide identification rates in data analysis limit its potential. This paper presents DeepRescore2, a computational workflow that leverages deep learning-based predictions of retention time and fragment ion intensity to enhance phosphopeptide identification and phosphosite localization. Benchmarking against existing workflows on a synthetic phosphopeptide dataset and application to real-world biological datasets demonstrate increased sensitivity, reduced missing values, and improved insights from phosphoproteomics-based biological analyses. DeepRescore2 is available at https://github.com/bzhanglab/DeepRescore2.