PepNet: A Fully Convolutional Neural Network for De novo Peptide Sequencing

preprint · Research Square · 2022

preprint · Research Square · 2022. Kaiyuan Liu et al. The de novo peptide sequencing, which does not rely on a comprehensive target sequence database, provided us…
Date 2022-02-09
Type preprint
Venue Research Square
Publisher Research Square
Contribution algorithm
DOI 10.21203/rs.3.rs-1341615/v1
Citations (OpenAlex) 5

Peer-reviewed version: Accurate de novo peptide sequencing using fully convolutional neural networks (2023-12-02, Nature Communications)

Abstract

The de novo peptide sequencing, which does not rely on a comprehensive target sequence database, provided us a way to identify novel peptides from tandem mass (MS/MS) spectra. However, current de novo sequencing algorithms suffer from lower accuracy and coverage, which hinders their applications in proteomics. In this paper, we present PepNet, a fully convolutional neural network (CNN) for high accuracy de novo peptide sequencing. It takes an MS/MS spectrum (represented as a high dimensional vector) as input, and outputs the optimal peptide sequence along with its confidence score. Our model was trained using a total of 30 million high-energy collisional dissociation (HCD) MS/MS spectra from multiple human peptide spectral libraries. The evaluation results show that PepNet significantly outperformed currently best-performing de novo sequencing algorithms (e.g. PointNovo and DeepNovo) at both peptide level accuracy and positional level accuracy. In addition, PepNet can sequence a large fraction of spectra that were not identified by database search engines, and thus could be used as a complementary tool of database search engines for peptide identification in proteomics.

Authors

  1. Kaiyuan Liu · Indiana University, Indiana University Bloomington
  2. Yuzhen Ye · Indiana University Bloomington
  3. Sujun Li · Indiana University Bloomington
  4. Haixu Tang · Indiana University, Indiana University Bloomington, University of California San Diego

Methods and tools

  • PepNet: Temporal convolutional network

Cited by (5)

Seen in the charts

Back to the full map

Back to top