RIGA-Fold: A General Framework for Protein Inverse Folding via Recurrent Interaction and Geometric Awareness

2026-02-04

AI Summary

This work addresses the limitations of existing protein inverse folding methods, which often suffer from restricted receptive fields that neglect long-range dependencies and error accumulation during single-pass inference. To overcome these challenges, the authors propose RIGA-Fold, a novel framework featuring a geometric attention update module for SE(3)-invariant local encoding and a global context bridging mechanism that dynamically injects topological information. The architecture integrates a dual-stream design to fuse trainable geometric features with frozen evolutionary priors from ESM-2/ESM-IF, and employs an iterative denoising strategy based on prediction, recycling, and refinement. Evaluated on CATH 4.2, TS50, and TS500 benchmarks, RIGA-Fold significantly outperforms current state-of-the-art models, achieving notable advances in both sequence recovery accuracy and structural consistency.

Technology Category

• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

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

Abstract

Protein inverse folding, the task of predicting amino acid sequences for desired structures, is pivotal for de novo protein design. However, existing GNN-based methods typically suffer from restricted receptive fields that miss long-range dependencies and a"single-pass"inference paradigm that leads to error accumulation. To address these bottlenecks, we propose RIGA-Fold, a framework that synergizes Recurrent Interaction with Geometric Awareness. At the micro-level, we introduce a Geometric Attention Update (GAU) module where edge features explicitly serve as attention keys, ensuring strictly SE(3)-invariant local encoding. At the macro-level, we design an attention-based Global Context Bridge that acts as a soft gating mechanism to dynamically inject global topological information. Furthermore, to bridge the gap between structural and sequence modalities, we introduce an enhanced variant, RIGA-Fold*, which integrates trainable geometric features with frozen evolutionary priors from ESM-2 and ESM-IF via a dual-stream architecture. Finally, a biologically inspired ``predict-recycle-refine''strategy is implemented to iteratively denoise sequence distributions. Extensive experiments on CATH 4.2, TS50, and TS500 benchmarks demonstrate that our geometric framework is highly competitive, while RIGA-Fold* significantly outperforms state-of-the-art baselines in both sequence recovery and structural consistency.

Problem

• protein inverse folding
• long-range dependencies
• error accumulation
• de novo protein design
• sequence-structure mapping

Innovation

• Protein Inverse Folding
• Geometric Attention
• Recurrent Interaction
• SE(3) Invariance
• Dual-stream Architecture