ProtInvTree: Deliberate Protein Inverse Folding with Reward-guided Tree Search
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913869915488256 |
|---|---|
| author | Liu, Mengdi Cheng, Xiaoxue Gao, Zhangyang Chang, Hong Tan, Cheng Shan, Shiguang Chen, Xilin |
| author_facet | Liu, Mengdi Cheng, Xiaoxue Gao, Zhangyang Chang, Hong Tan, Cheng Shan, Shiguang Chen, Xilin |
| contents | Designing protein sequences that fold into a target 3D structure, known as protein inverse folding, is a fundamental challenge in protein engineering. While recent deep learning methods have achieved impressive performance by recovering native sequences, they often overlook the one-to-many nature of the problem: multiple diverse sequences can fold into the same structure. This motivates the need for a generative model capable of designing diverse sequences while preserving structural consistency. To address this trade-off, we introduce ProtInvTree, the first reward-guided tree-search framework for protein inverse folding. ProtInvTree reformulates sequence generation as a deliberate, step-wise decision-making process, enabling the exploration of multiple design paths and exploitation of promising candidates through self-evaluation, lookahead, and backtracking. We propose a two-stage focus-and-grounding action mechanism that decouples position selection and residue generation. To efficiently evaluate intermediate states, we introduce a jumpy denoising strategy that avoids full rollouts. Built upon pretrained protein language models, ProtInvTree supports flexible test-time scaling by expanding the search depth and breadth without retraining. Empirically, ProtInvTree outperforms state-of-the-art baselines across multiple benchmarks, generating structurally consistent yet diverse sequences, including those far from the native ground truth. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_00925 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ProtInvTree: Deliberate Protein Inverse Folding with Reward-guided Tree Search Liu, Mengdi Cheng, Xiaoxue Gao, Zhangyang Chang, Hong Tan, Cheng Shan, Shiguang Chen, Xilin Biomolecules Computer Vision and Pattern Recognition Machine Learning Designing protein sequences that fold into a target 3D structure, known as protein inverse folding, is a fundamental challenge in protein engineering. While recent deep learning methods have achieved impressive performance by recovering native sequences, they often overlook the one-to-many nature of the problem: multiple diverse sequences can fold into the same structure. This motivates the need for a generative model capable of designing diverse sequences while preserving structural consistency. To address this trade-off, we introduce ProtInvTree, the first reward-guided tree-search framework for protein inverse folding. ProtInvTree reformulates sequence generation as a deliberate, step-wise decision-making process, enabling the exploration of multiple design paths and exploitation of promising candidates through self-evaluation, lookahead, and backtracking. We propose a two-stage focus-and-grounding action mechanism that decouples position selection and residue generation. To efficiently evaluate intermediate states, we introduce a jumpy denoising strategy that avoids full rollouts. Built upon pretrained protein language models, ProtInvTree supports flexible test-time scaling by expanding the search depth and breadth without retraining. Empirically, ProtInvTree outperforms state-of-the-art baselines across multiple benchmarks, generating structurally consistent yet diverse sequences, including those far from the native ground truth. |
| title | ProtInvTree: Deliberate Protein Inverse Folding with Reward-guided Tree Search |
| topic | Biomolecules Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2506.00925 |