DPST: De Novo Peptide Sequencing with Amino-Acid-Aware Transformers

preprint · arXiv · 2022

preprint · arXiv · 2022. Yan Yang et al. De novo peptide sequencing aims to recover amino acid sequences of a peptide from tandem mass spectrometry…
Date 2022-03-23
Type preprint
Venue arXiv
Publisher arXiv
Contribution algorithm
DOI 10.48550/arXiv.2203.13132
Citations (OpenAlex) 8

Abstract

De novo peptide sequencing aims to recover amino acid sequences of a peptide from tandem mass spectrometry (MS) data. Existing approaches for de novo analysis enumerate MS evidence for all amino acid classes during inference. It leads to over-trimming on receptive fields of MS data and restricts MS evidence associated with following undecoded amino acids. Our approach, DPST, circumvents these limitations with two key components: (1) A confidence value aggregation encoder to sketch spectrum representations according to amino-acid-based connectivity among MS; (2) A global-local fusion decoder to progressively assimilate contextualized spectrum representations with a predefined preconception of localized MS evidence and amino acid priors. Our components originate from a closed-form solution and selectively attend to informative amino-acid-aware MS representations. Through extensive empirical studies, we demonstrate the superiority of DPST, showing that it outperforms state-of-the-art approaches by a margin of 12% - 19% peptide accuracy.

Authors

  1. Yan Yang · Australian National University, CSIRO
  2. Zakir Hossain · Australian National University, CSIRO
  3. Khandaker Asif · CSIRO
  4. Liyuan Pan · Australian National University, CSIRO
  5. Shafin Rahman · North South University
  6. Eric Stone · Australian National University, CSIRO

Methods and tools

  • DPST: Amino-acid-aware transformer

Cites (5)

Cited by (11)

Seen in the charts

Back to the full map

Back to top