Problem
Research questions and friction points this paper is trying to address.
de novo peptide sequencing
tandem mass spectrometry
fragment imputation
mass-aware representation
decoder error
GyroNovo: Error-Guided Fragment Imputation with Mass-Aware Attention for \textit{De Novo} Peptide Sequencing
🤖 AI Summary
This study addresses the limited accuracy of de novo peptide sequencing in tandem mass spectrometry caused by missing fragments and spectral noise, as well as the oversight of decoding errors and mass discrepancies in existing methods. To this end, we propose GyroNovo, a novel framework that introduces an error-driven adaptive imputation mechanism leveraging decoder feedback to complete missing fragments, combined with a hard-sample augmentation strategy to enhance robustness. Furthermore, it incorporates a quality-aware attention mechanism based on rotary position encoding, explicitly modeling inter-peak mass differences as an inductive bias within the Transformer architecture. Evaluated on the NovoBench benchmark, GyroNovo improves peptide-level and amino acid-level accuracy by approximately 9% and 7%, respectively, over the strongest baseline.
📝 Abstract
De novo peptide sequencing from tandem mass spectra is essential for identifying peptides without relying on reference databases. Despite advances in deep learning, accurate sequencing remains challenging because experimental spectra are often sparse, noisy, and incomplete, leaving informative b- and y-ion fragments unobserved. Existing methods attempt to recover this missing evidence via latent-space imputation before autoregressive decoding. However, they typically treat imputation as a fixed reconstruction task, without considering which missing fragments are most relevant to decoder errors. Moreover, existing peak representations do not explicitly model mass differences between peaks, despite their fundamental importance.
We introduce GyroNovo, a framework with two main contributions. First, we use decoder errors observed during training to adapt the imputation objective, prioritizing fragments associated with frequent decoding errors. We further use the decoder error distribution to construct easy and hard augmented views of each spectrum, enabling the decoder to learn under varying degrees of spectral corruption and missing-fragment severity. Second, we introduce a mass-aware inductive bias into self-attention by using rotary embeddings to encode pairwise mass differences between spectral peaks. Together, these components align missing-fragment recovery with decoder behavior while explicitly incorporating the mass relationships that underlie peptide fragmentation. At inference time, GyroNovo retains a standard encoder-imputer-decoder architecture and requires neither additional inputs nor auxiliary search procedures. Experiments on NovoBench show gains of about 9 percentage points in peptide-level precision and 7 percentage points in amino-acid-level precision over the state-of-the-art baseline. Code: https://github.com/UBC-NLP/gyronovo.
Innovation
Methods, ideas, or system contributions that make the work stand out.
De novo peptide sequencing
Error-guided imputation
Mass-aware attention
Rotary embeddings
Tandem mass spectra
🔎 Similar Papers
No similar papers found.
