Uncovering Thousands of New Peptides with Sequence-Mask-Search Hybrid De Novo Peptide Sequencing Framework* (2019-10-07T00:00:00.000000Z)

TL;DR

SMSNet is developed, a deep learning-based de novo peptide sequencing framework that achieves >95% amino acid accuracy while retaining good identification coverage and is an invaluable tool for any future proteomics and peptidomics studies, including tumor neoantigen discovery, antibody sequencing, and proteome characterization of non-model organisms.

Abstract

De novo peptide sequencing is a promising approach for discovering new peptides. However, its performance is hindered by the fact that most MS/MS spectra do not contain complete amino acid sequence information. Here, we present a deep learning-based de novo sequencing model, SMSNet, together with a post-processing strategy that pinpoints misidentified residues and utilizes user-provided sequence database to revise the identifications. The results show that many new HLA antigens and phosphopeptides can be discovered with our strategy. Graphical Abstract Highlights Deep learning-based hybrid de novo sequencing with database search strategy. Accurate identifications via ability to revise confidence scores and amino acids. Discovery of >10,000 potential new HLA antigens and human phosphopeptides. A dataset of >26 million annotated HCD spectra from Q Exactive instruments. Typical analyses of mass spectrometry data only identify amino acid sequences that exist in reference databases. This restricts the possibility of discovering new peptides such as those that contain uncharacterized mutations or originate from unexpected processing of RNAs and proteins. De novo peptide sequencing approaches address this limitation but often suffer from low accuracy and require extensive validation by experts. Here, we develop SMSNet, a deep learning-based de novo peptide sequencing framework that achieves >95% amino acid accuracy while retaining good identification coverage. Applications of SMSNet on landmark proteomics and peptidomics studies reveal over 10,000 previously uncharacterized HLA antigens and phosphopeptides, and in conjunction with database-search methods, expand the coverage of peptide identification by almost 30%. The power to accurately identify new peptides of SMSNet would make it an invaluable tool for any future proteomics and peptidomics studies, including tumor neoantigen discovery, antibody sequencing, and proteome characterization of non-model organisms.

References38 items

Prosit: proteome-wide prediction of peptide tandem mass spectra by deep learning

Deep learning enables de novo peptide sequencing from data-independent-acquisition mass spectrometry

Noncoding regions are the main source of targetable tumor-specific antigens

UniProt: a worldwide hub of protein knowledge

A subset of HLA-I peptides are not genomically templated: Evidence for cis- and trans-spliced peptide ligands

High-throughput and high-sensitivity phosphoproteomics with the EasyPhos platform

PDV: an integrative proteomics data viewer

A Potential Golden Age to Come—Current Tools, Recent Use Cases, and Future Avenues for De Novo Sequencing in Proteomics

Estimating the Contribution of Proteasomal Spliced Peptides to the HLA-I Ligandome*

Focal Loss for Dense Object Detection

De novo peptide sequencing by deep learning

NetMHCpan-4.0: Improved Peptide–MHC Class I Interaction Predictions Integrating Eluted Ligand and Peptide Binding Affinity Data

Mass Spectrometry Profiling of HLA‐Associated Peptidomes in Mono‐allelic Cells Enables More Accurate Epitope Prediction

Building ProteomeTools based on a complete synthetic human proteome

A large fraction of HLA class I ligands are proteasome-generated spliced peptides

Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation

Layer Normalization

Deep Residual Learning for Image Recognition

Reference sequence (RefSeq) database at NCBI: current status, taxonomic expansion, and functional annotation

Novor: Real-Time Peptide de Novo Sequencing Software

HISAT: a fast spliced aligner with low memory requirements

PhosphoSitePlus, 2014: mutations, PTMs and recalibrations

Show and tell: A neural image caption generator

The immune epitope database (IEDB) 3.0

Sequence to Sequence Learning with Neural Networks

Neural Machine Translation by Jointly Learning to Align and Translate

Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation

A Cross-platform Toolkit for Mass Spectrometry and Proteomics

De Novo Sequencing and Homology Searching‡‡*

The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data.

MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification

PepNovo: de novo peptide sequencing via probabilistic network modeling.

WebLogo: a sequence logo generator.

Searching sequence databases via De novo peptide sequencing by tandem mass spectrometry

Error-tolerant identification of peptides in sequence databases by peptide sequence tags.

Neural Machine Translation

jointly learning to align and translate. arXiv preprint arXiv, 1409.0473 23

PEAKS: powerful software for peptide de novo sequencing by tandem mass spectrometry. Rapid communications in mass spectrometry