Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing

2025-05-23

AI Summary

To address the pervasive issue of missing fragment ions in mass spectrometry-based de novo peptide sequencing, this paper proposes a novel “complete-then-decode” paradigm operating in latent space. Unlike conventional signal-domain imputation, our method introduces a learnable peak-query mechanism grounded in theoretical fragment spectra, formulating missing fragment recovery as a set prediction problem solved via optimal bipartite matching—thereby avoiding spectral distortion inherent in raw-spectrum interpolation. We further integrate an autoregressive decoder to enable end-to-end peptide sequence inference. Evaluated on three standard benchmarks, our approach significantly outperforms existing state-of-the-art methods, achieving substantial gains in sequencing accuracy. The implementation is publicly available.

Technology Category

• Search and Optimization: Learning to Search
• Machine Learning: Deep Generative Models & Autoencoders
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

De novo peptide sequencing is a fundamental computational technique for ascertaining amino acid sequences of peptides directly from tandem mass spectrometry data, eliminating the need for reference databases. Cutting-edge models usually encode the observed mass spectra into latent representations from which peptides are predicted autoregressively. However, the issue of missing fragmentation, attributable to factors such as suboptimal fragmentation efficiency and instrumental constraints, presents a formidable challenge in practical applications. To tackle this obstacle, we propose a novel computational paradigm called underline{ extbf{L}}atent underline{ extbf{I}}mputation before underline{ extbf{P}}rediction (LIPNovo). LIPNovo is devised to compensate for missing fragmentation information within observed spectra before executing the final peptide prediction. Rather than generating raw missing data, LIPNovo performs imputation in the latent space, guided by the theoretical peak profile of the target peptide sequence. The imputation process is conceptualized as a set-prediction problem, utilizing a set of learnable peak queries to reason about the relationships among observed peaks and directly generate the latent representations of theoretical peaks through optimal bipartite matching. In this way, LIPNovo manages to supplement missing information during inference and thus boosts performance. Despite its simplicity, experiments on three benchmark datasets demonstrate that LIPNovo outperforms state-of-the-art methods by large margins. Code is available at href{https://github.com/usr922/LIPNovo}{https://github.com/usr922/LIPNovo}.

Problem

• Addresses missing fragmentation in peptide sequencing
• Proposes latent imputation before peptide prediction
• Enhances performance without raw data generation

Innovation

• Latent imputation compensates missing fragmentation information
• Set-prediction problem with learnable peak queries
• Optimal bipartite matching generates latent representations