使用衍生自单体蛋白质的伪二聚体改善蛋白质和蛋白质相互作用
Improving protein and protein interactions using pseudo-dimers derived from monomeric proteins.
摘要(中文译文)
在二聚体复合物中准确预测蛋白质-蛋白质相互作用（protein-protein interactions, PPIs）仍然是计算生物学的一项基本挑战。尽管现有的PPIs预测模型，如AlphaFold-Multimer（AF-Multimer）和AlphaFold3（AF3），已取得令人瞩目的性能，但由于蛋白质二聚体结构的可用性有限——其收集既昂贵又耗费人力——这些模型仍受困于精度不尽如人意的问题。在此，我们提出一种简单而有效的预训练方法，称为拆分与合并代理（split and merge proxy, SMP），首次利用丰富的单体蛋白质来模拟各种PPIs任务。具体而言，SMP通过将单体数据拆分为两个亚基（称为伪受体和伪配体）来构建伪二聚体，并训练模型通过预测其伪相互作用（例如接触或对接）将其重新合并。这一代理任务使得无需额外成本即可进行大规模预训练。使用SMP预训练并随后在真实蛋白质二聚体数据集上微调的模型，在多个基准测试中表现出持续提升的准确性和泛化能力，超越了强基线。值得注意的是，在若干CASP15二聚体靶标上，SMP提供的结构预测比AF-Multimer和AF3更为准确。我们的发现凸显了SMP作为一种可扩展策略，能够利用单体数据推进蛋白质复合物建模，为单体与多聚体之间的联系提供了洞见。
Abstract
Accurately predicting protein-protein interactions (PPIs) in dimeric complexes remains a fundamental challenge in computational biology. Although existing PPIs prediction models, such as AlphaFold-Multimer (AF-Multimer) and AlphaFold3 (AF3), have achieved impressive performance, they still suffer from unsatisfactory accuracy due to the limited availability of protein dimer structures, whose collection is both expensive and labor-intensive. Here, we introduce a simple yet effective pre-training method, termed split and merge proxy (SMP), that leverages abundant monomeric proteins to simulate various PPIs tasks for the first time. Specifically, SMP constructs pseudo-dimers by splitting monomer data into two subunits, referred to as pseudo-receptors and pseudo-ligands, and trains models to merge them back by predicting their pseudo interactions (e.g., contact or docking). This proxy task enables large-scale pre-training without additional cost. Models pre-trained with SMP and subsequently fine-tuned on real protein dimer datasets demonstrate consistently improved accuracy and generalization across multiple benchmarks, surpassing strong baselines. Notably, SMP delivers more accurate structure predictions than both AF-Multimer and AF3 on several CASP15 dimer targets. Our findings highlight SMP as a scalable strategy for harnessing monomeric data to advance protein complex modeling, providing insights into the linkage between monomers and multimers.
如何引用
AMA
Hao Du, Xinzhe Zheng, Yuchen Ren, He Huang, Xinqi Gong, Wanli Ouyang, et al. Improving protein and protein interactions using pseudo-dimers derived from monomeric proteins.. Nature communications. 2026; doi:10.1038/s41467-026-73885-5.
APA
Hao Du, Xinzhe Zheng, Yuchen Ren, He Huang, Xinqi Gong, Wanli Ouyang, et al (2026). Improving protein and protein interactions using pseudo-dimers derived from monomeric proteins.. Nature communications. https://doi.org/10.1038/s41467-026-73885-5
GB/T 7714
Hao Du, Xinzhe Zheng, Yuchen Ren, He Huang, Xinqi Gong, Wanli Ouyang, et al. Improving protein and protein interactions using pseudo-dimers derived from monomeric proteins.[J]. Nature communications, 2026 doi:10.1038/s41467-026-73885-5.
在 PubMed增强版 阅读全文
翻译全文 · AI 文献问答 · PDF 原文下载 · 引用导出