Spectral libraries are useful resources in proteomic data analysis, but they can only identify previously known peptides existing within libraries. Recent advances in deep learning allow tandem mass spectra of peptides to be predicted from their amino acid sequences, which enables predicted spectral libraries to be compiled. Searching against such libraries has been shown to improve sensitivity in peptide identification over conventional sequence database searching. However, current prediction models lack support for longer peptides, and thus far predicted library searching was only demonstrated for backbone ion-only spectrum prediction methods. Here, we propose a deep learning-based full-spectrum prediction method to generate predicted spectral libraries for peptide identification. We...[ Read more ]

Spectral libraries are useful resources in proteomic data analysis, but they can only identify previously known peptides existing within libraries. Recent advances in deep learning allow tandem mass spectra of peptides to be predicted from their amino acid sequences, which enables predicted spectral libraries to be compiled. Searching against such libraries has been shown to improve sensitivity in peptide identification over conventional sequence database searching. However, current prediction models lack support for longer peptides, and thus far predicted library searching was only demonstrated for backbone ion-only spectrum prediction methods. Here, we propose a deep learning-based full-spectrum prediction method to generate predicted spectral libraries for peptide identification. We demonstrated the superiority of using full-spectrum libraries over backbone ion-only prediction approaches in spectral library searching. Furthermore, merging spectra from different prediction models, as a form of ensemble learning, can produce improved spectral libraries in terms of identification sensitivity. We also show that a hybrid library combining predicted and experimental spectra can lead to 20% more confident identifications over experimental library searching or sequence database searching. For carrying out target decoy search strategy in library search, we have experimented generating decoys via a shuffle-and-predict method. Results from our experiments suggest that predicted decoys can be a viable alternative to existing library decoy generation methods.