基于基序的蛋白质序列与结构联合设计
Joint Design of Protein Sequence and Structure based on Motifs
AI总结：
针对蛋白质序列与结构协同设计不足的问题，本文提出 GeoPro，利用等变3D骨架编码器和几何引导序列解码器联合生成二者，并在金属蛋白数据上验证其可发现具备稳定折叠与功能潜力的新型蛋白。
摘要：
设计具有所需功能的新型蛋白质在生物学和化学中至关重要。然而，现有工作大多聚焦于蛋白质序列设计，使蛋白质序列与结构的协同设计仍未得到充分探索。本文提出 GeoPro，这是一种联合设计蛋白质骨架结构与序列的方法。我们的动机在于，蛋白质序列与其骨架结构相互约束，因此对二者进行联合设计不仅可以避免不折叠和错误折叠，还能产生更多具有所需功能的多样化候选蛋白质。为此，GeoPro 由用于三维（3D）骨架结构的等变编码器，以及受 3D 几何引导的蛋白质序列解码器驱动。在两个具有重要生物学意义的金属蛋白质数据集（包括 β-内酰胺酶和肌红蛋白）上的实验结果表明，所提出的 GeoPro 在大多数指标上优于若干强基线。值得注意的是，我们的方法发现了蛋白质数据库（PDB）和 UniProt 中不存在的新型 β-内酰胺酶和肌红蛋白。这些蛋白质表现出稳定的折叠以及令人联想到天然蛋白质的活性位点环境，证明了其具备生物学功能的出色潜力。
英文摘要：
Designing novel proteins with desired functions is crucial in biology and chemistry. However, most existing work focus on protein sequence design, leaving protein sequence and structure co-design underexplored. In this paper, we propose GeoPro, a method to design protein backbone structure and sequence jointly. Our motivation is that protein sequence and its backbone structure constrain each other, and thus joint design of both can not only avoid nonfolding and misfolding but also produce more diverse candidates with desired functions. To this end, GeoPro is powered by an equivariant encoder for three-dimensional (3D) backbone structure and a protein sequence decoder guided by 3D geometry. Experimental results on two biologically significant metalloprotein datasets, including $β$-lactamases and myoglobins, show that our proposed GeoPro outperforms several strong baselines on most metrics. Remarkably, our method discovers novel $β$-lactamases and myoglobins which are not present in protein data bank (PDB) and UniProt. These proteins exhibit stable folding and active site environments reminiscent of those of natural proteins, demonstrating their excellent potential to be biologically functional.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
