美国密歇根大学Yang Zhang等研究人员合作利用DeepMSA2和海量元基因组学数据改进深度学习蛋白质单体和复合体结构预测。该项研究成果于2024年1月2日在线发表在《自然—方法学》杂志上。

利用基因组和元基因组序列数据库的迭代比对搜索，研究人员报告了用于统一蛋白质单链和多链多序列比对（MSA）构建的DeepMSA2管线。大规模基准测试表明，与目前最先进的方法相比，DeepMSA2 MSA能显著提高蛋白质三级和四级结构预测的准确性。与DeepMSA2集成的管线参与了最新的CASP15实验，并创建了复杂的结构模型，其质量大大高于AlphaFold2-Multimer服务器（v.2.2.0）。

详细的数据分析显示，DeepMSA2的主要优势在于其均衡的配准搜索和有效的模型选择，以及整合庞大的元基因组学数据库的能力。这些结果展示了通过高级MSA构建改进深度学习蛋白质结构预测的新途径，并进一步证明了优化基于深度学习的结构预测方法的输入信息，必须像设计预测器本身一样谨慎。

附：英文原文

Title: Improving deep learning protein monomer and complex structure prediction using DeepMSA2 with huge metagenomics data

Author: Zheng, Wei, Wuyun, Qiqige, Li, Yang, Zhang, Chengxin, Freddolino, P. Lydia, Zhang, Yang

Issue&Volume: 2024-01-02

Abstract: Leveraging iterative alignment search through genomic and metagenome sequence databases, we report the DeepMSA2 pipeline for uniform protein single- and multichain multiple-sequence alignment (MSA) construction. Large-scale benchmarks show that DeepMSA2 MSAs can remarkably increase the accuracy of protein tertiary and quaternary structure predictions compared with current state-of-the-art methods. An integrated pipeline with DeepMSA2 participated in the most recent CASP15 experiment and created complex structural models with considerably higher quality than the AlphaFold2-Multimer server (v.2.2.0). Detailed data analyses show that the major advantage of DeepMSA2 lies in its balanced alignment search and effective model selection, and in the power of integrating huge metagenomics databases. These results demonstrate a new avenue to improve deep learning protein structure prediction through advanced MSA construction and provide additional evidence that optimization of input information to deep learning-based structure prediction methods must be considered with as much care as the design of the predictor itself.

DOI: 10.1038/s41592-023-02130-4

Source: https://www.nature.com/articles/s41592-023-02130-4