arXiv：cs.LG· Tianmeng Hu, Biao Luo, Ke Li·· 3 小时前
TANGO：RNA 拓扑与几何协同设计框架
TANGO: RNA Topology and Geometry Co-Design
AI 导读
TANGO 将 RNA 逆向设计建模为拓扑与几何协同设计问题，提出多目标多智能体强化学习框架，采用先学二级结构拓扑策略、再迁移到三级结构目标的分阶段训练，最后用前沿引导微调平衡二级结构保真度、三级结构保真度与序列多样性。实验显示其三级结构和二级结构性能分别比最强基线提升 18.3% 和 52.7%，并能生成多样且新颖的设计。
正文
Abstract:Designing RNA sequences that satisfy target tertiary-structure objectives is an important problem in RNA engineering, with broad applications in synthetic biology and therapeutics. Existing 3D RNA design methods have made substantial progress, but they largely treat the target fold as a geometry-only objective. This is incomplete for RNA, whose functional folds are specified not only by tertiary geometry but also by secondary base-pairing topology. We therefore formulate RNA inverse design as a topology-and-geometry co-design problem and propose TANGO, a multi-objective multi-agent reinforcement learning framework for this task. TANGO adopts a staged curriculum: it first learns a topology-aware policy from secondary-structure targets, then transfers this policy to tertiary-structure targets to train a general co-design model, and finally applies frontier-guided fine-tuning to refine target-specific trade-offs among secondary-structure fidelity, tertiary-structure fidelity, and sequence diversity. Experiments show that TANGO generates diverse and novel designs, improving tertiary- and secondary-structure performance by 18.3% and 52.7%, respectively, over the strongest baseline.
| Comments: | NeurIPS 2026 |
| Subjects: | Machine Learning (cs.LG); Multiagent Systems (cs.MA) |
| MSC classes: | I.2.6 |
| Cite as: | arXiv:2610.20277 [cs.LG] |
| (or arXiv:2610.20277v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.20277 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Tianmeng Hu [view email]
[v1]
Wed, 14 Oct 2026 14:37:51 UTC (3,783 KB)
来源：arXiv：cs.LG · arxiv.org