RT-GCTnovo: A Peptide De Novo Sequencing Model Incorporating Gated Multi-scale Features and Dynamic Mass Masks
peer-reviewed · 2025 18th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI) · 2025
| Date | 2025-10-25 |
| Type | peer-reviewed |
| Venue | 2025 18th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI) |
| Publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| Contribution | algorithm |
| DOI | 10.1109/cisp-bmei68103.2025.11259319 |
| Citations (OpenAlex) | 0 |
Abstract
De novo peptide sequencing infers amino acid sequences directly from tandem mass spectra without relying on protein databases. This paper present RT-GCTnovo, a model that integrates multi-scale feature extraction, gated fusion, and dynamic decoding to improve sequencing accuracy. The model employs Transformer encoders and convolutional neural networks to capture global and local spectral features, enhanced by retention time information as a new constraint. A sigmoid gating mechanism adaptively fuses these features, and a semiautoregressive decoder with dynamic mass masking iteratively refines predictions. Experimental results on multispecies datasets show that RT-GCTnovo outperforms DeepNovo, Casanovo, and TSARseqNovo, achieving 57.7 % peptide-level accuracy and 79.8% amino acid-level accuracy.
Methods and tools
- RT-GCTnovo: Combines Transformer encoders with convolutional layers to capture global and local spectral features, fuses them through a sigmoid gate, adds retention time as a constraint, and decodes semi-autoregressively with dynamic mass masks.
Cites (12)
- A transformer-based semi-autoregressive framework for high-speed and accurate de novo peptide sequencing (2025) crossref
- Bidirectional de novo peptide sequencing using a transformer model (2024) crossref
- Introducing π-HelixNovo for practical large-scale de novo peptide sequencing (2024) crossref
- De novo peptide sequencing with InstaNovo: Accurate, database-free peptide identification for large scale proteomics experiments (2023) crossref
- De novo mass spectrometry peptide sequencing with a transformer model (2022) crossref
- Computationally instrument-resolution-independent de novo peptide sequencing for high-resolution devices (2021) crossref
- Improved de novo peptide sequencing using LC retention time information (2018) crossref
- De novo peptide sequencing by deep learning (2017) crossref
- Novor: Real-Time Peptide de Novo Sequencing Software (2015) crossref
- PepNovo: de novo peptide sequencing via probabilistic network modeling (2005) crossref
- PEAKS: powerful software for peptide de novo sequencing by tandem mass spectrometry (2003) crossref
- Sequence database searches via de novo peptide sequencing by tandem mass spectrometry (1997) crossref