DS-ProGen: A Dual-Structure Deep Language Model for Functional Protein Design

2025-05-18

AI Summary

Existing inverse protein folding (IPF) methods predominantly rely on either backbone coordinates or molecular surface features alone, failing to jointly model global geometric and local chemical constraints—thereby limiting sequence prediction accuracy. To address this, we propose the first multimodal modeling framework that integrates backbone 3D coordinates with molecular surface descriptors encoding both chemical and geometric properties. Our approach employs a dual-structure encoder within a deep language model, combining structure-aware embeddings with an autoregressive sequence generation paradigm. Evaluated on the PRIDE benchmark, our method achieves a state-of-the-art 61.47% sequence recovery rate. Moreover, it significantly improves prediction accuracy for binding sites of diverse biomolecular partners—including ligands, metal ions, and RNA—demonstrating enhanced capability in capturing functionally critical interactions. This work establishes a new paradigm for function-driven, high-precision protein design.

Technology Category

• Machine Learning: Multimodal Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Inverse Protein Folding (IPF) is a critical subtask in the field of protein design, aiming to engineer amino acid sequences capable of folding correctly into a specified three-dimensional (3D) conformation. Although substantial progress has been achieved in recent years, existing methods generally rely on either backbone coordinates or molecular surface features alone, which restricts their ability to fully capture the complex chemical and geometric constraints necessary for precise sequence prediction. To address this limitation, we present DS-ProGen, a dual-structure deep language model for functional protein design, which integrates both backbone geometry and surface-level representations. By incorporating backbone coordinates as well as surface chemical and geometric descriptors into a next-amino-acid prediction paradigm, DS-ProGen is able to generate functionally relevant and structurally stable sequences while satisfying both global and local conformational constraints. On the PRIDE dataset, DS-ProGen attains the current state-of-the-art recovery rate of 61.47%, demonstrating the synergistic advantage of multi-modal structural encoding in protein design. Furthermore, DS-ProGen excels in predicting interactions with a variety of biological partners, including ligands, ions, and RNA, confirming its robust functional retention capabilities.

Problem

• Improving amino acid sequence prediction for specified 3D protein structures
• Overcoming limitations of single-structure models in protein design
• Enhancing functional protein interactions with diverse biological partners

Innovation

• Dual-structure deep language model for protein design
• Integrates backbone geometry and surface-level representations
• Multi-modal structural encoding enhances functional retention