使用几何向量场网络进行从头蛋白质设计
De novo protein design using geometric vector field networks
AI总结：
本文提出向量场网络（VFN），通过在虚拟原子坐标间进行可学习的向量计算来建模蛋白质主链框架，在蛋白质扩散和逆折叠任务中均显著提升了设计性、多样性与序列恢复率，超越现有SoTA模型。
摘要：
像蛋白质扩散这样的创新使得从头蛋白质设计取得了显著进展，这是生命科学中的一个重要课题。这些方法通常依赖蛋白质结构编码器来对残基主链框架进行建模，而在这些框架中原子并不存在。大多数先前的编码器依赖于原子级特征，例如原子间的角度和距离，这些在此场景下是不可用的。迄今为止，仅提出了几种简单的编码器（如IPA）用于此场景，使得框架建模成为瓶颈。在本工作中，我们提出了向量场网络（VFN），它使网络层能够在框架锚定的虚拟原子坐标之间执行可学习的向量计算，从而实现更高的框架建模能力。该向量计算以类似于线性层的方式运作，每个输入通道接收3D虚拟原子坐标而非标量值。向量计算输出的多个特征向量随后通过注意力聚合来更新残基表示和虚拟原子坐标。值得注意的是，VFN在建模框架和原子方面同样出色，因为真实原子可被视为虚拟原子进行建模，这使得VFN成为一种潜在的通用编码器。在蛋白质扩散（框架建模）中，VFN展现出相对于IPA的显著性能优势，在设计性（67.04% vs. 53.58%）和多样性（66.54% vs. 51.98%）上均表现优异。在逆折叠（框架和原子建模）中，VFN在序列恢复率上超越了先前的SoTA模型PiFold（54.7% vs. 51.66%）。我们还提出了一种为VFN配备ESM模型的方法，该方法以大幅优势超越了先前基于ESM的SoTA模型LM-Design（62.67% vs. 55.65%）。
英文摘要：
Innovations like protein diffusion have enabled significant progress in de novo protein design, which is a vital topic in life science. These methods typically depend on protein structure encoders to model residue backbone frames, where atoms do not exist. Most prior encoders rely on atom-wise features, such as angles and distances between atoms, which are not available in this context. Thus far, only several simple encoders, such as IPA, have been proposed for this scenario, exposing the frame modeling as a bottleneck. In this work, we proffer the Vector Field Network (VFN), which enables network layers to perform learnable vector computations between coordinates of frame-anchored virtual atoms, thus achieving a higher capability for modeling frames. The vector computation operates in a manner similar to a linear layer, with each input channel receiving 3D virtual atom coordinates instead of scalar values. The multiple feature vectors output by the vector computation are then used to update the residue representations and virtual atom coordinates via attention aggregation. Remarkably, VFN also excels in modeling both frames and atoms, as the real atoms can be treated as the virtual atoms for modeling, positioning VFN as a potential universal encoder. In protein diffusion (frame modeling), VFN exhibits an impressive performance advantage over IPA, excelling in terms of both designability (67.04% vs. 53.58%) and diversity (66.54% vs. 51.98%). In inverse folding (frame and atom modeling), VFN outperforms the previous SoTA model, PiFold (54.7% vs. 51.66%), on sequence recovery rate. We also propose a method of equipping VFN with the ESM model, which significantly surpasses the previous ESM-based SoTA (62.67% vs. 55.65%), LM-Design, by a substantial margin.
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。
