NS-Pep: De novo Peptide Design with Non-Standard Amino Acids

2025-10-01

AI Summary

Current peptide design methods are constrained to canonical amino acids, leading to inadequate modeling and imbalanced distribution of non-standard amino acids (NSAAs). To address this, we propose the first unified sequence–structure co-design framework supporting NSAAs. Our method introduces three key innovations: (1) frequency-aware logit calibration to mitigate over-penalization of rare NSAAs; (2) a progressive side-chain–aware (PSP) module to enhance torsion angle and atomic coordinate prediction accuracy; and (3) an interaction-aware weighting (IAW) mechanism to improve binding-pocket residue modeling. Experiments demonstrate that our approach achieves 6.23% higher sequence recovery rate and 5.12% improved binding affinity over baselines. Peptide folding success increases by 17.76% relative to AlphaFold3, and—critically—the framework natively supports end-to-end folding of NSAA-containing peptides. This significantly expands the chemical space and modeling capability for peptide-based drug design.

Technology Category

• Multiagent Systems: Mechanism Design
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Neuro-Symbolic Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Peptide drugs incorporating non-standard amino acids (NSAAs) offer improved binding affinity and improved pharmacological properties. However, existing peptide design methods are limited to standard amino acids, leaving NSAA-aware design largely unexplored. We introduce NS-Pep, a unified framework for co-designing peptide sequences and structures with NSAAs. The main challenge is that NSAAs are extremely underrepresented-even the most frequent one, SEP, accounts for less than 0.4% of residues-resulting in a severe long-tailed distribution. To improve generalization to rare amino acids, we propose Residue Frequency-Guided Modification (RFGM), which mitigates over-penalization through frequency-aware logit calibration, supported by both theoretical and empirical analysis. Furthermore, we identify that insufficient side-chain modeling limits geometric representation of NSAAs. To address this, we introduce Progressive Side-chain Perception (PSP) for coarse-to-fine torsion and location prediction, and Interaction-Aware Weighting (IAW) to emphasize pocket-proximal residues. Moreover, NS-Pep generalizes naturally to the peptide folding task with NSAAs, addressing a major limitation of current tools. Experiments show that NS-Pep improves sequence recovery rate and binding affinity by 6.23% and 5.12%, respectively, and outperforms AlphaFold3 by 17.76% in peptide folding success rate.

Problem

• Designing peptides with non-standard amino acids for enhanced drug properties
• Addressing data scarcity and long-tailed distribution of rare amino acids
• Improving geometric representation through advanced side-chain modeling techniques

Innovation

• Co-designs peptide sequences with non-standard amino acids
• Uses frequency-guided modification for rare amino acids
• Employs progressive side-chain perception for geometric accuracy