Integration of alternative fragmentation techniques into standard LC-MS workflows using a single deep learning model enhances proteome coverage

Bottom-up proteomics relies predominantly on collision-induced dissociation (CID) for peptide sequencing, which has achieved remarkable sensitivity and efficiency now enabling single-cell analysis. However, CID shows limitations in characterizing post-translational modifications and complex proteoforms. Here we have developed an integrated mass spectrometry platform enabling automated collision-, electron- and photon-based fragmentation techniques. Using multi-enzyme deep proteomics workflows, we generated comprehensive datasets to train a unified Prosit deep learning model predicting spectra across all dissociation methods. This publicly available model, now integrated into FragPipe's MSBooster module, increased protein identifications by >10% on average for both data-dependent and data-independent acquisition across all fragmentation techniques. We demonstrate that alternative approaches, particularly electron-induced and ultraviolet photodissociation, which generate richer, more informative spectra, achieve identification efficiency competitive with CID while providing superior sequence coverage. This work establishes a framework enabling routine application of advanced fragmentation techniques in standard proteomics pipelines.