发表机构
Institute for AI Industry Research (AIR), Tsinghua University; Department of Computer Science and Technology, Tsinghua University; PharMolix Inc.（清华大学人工智能产业研究院（AIR）; 清华大学计算机科学与技术系; PharMolix公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
AbGaze提出基于注意力几何表示学习的端到端抗体设计框架，编码距离、方向及表面法线并共享于多任务，显著提升氨基酸恢复率、结构精度、对接质量及亲和力。
AI 中文摘要
计算抗体设计需要能够捕捉抗原-抗体相互作用背后几何模式的表示，然而现有方法往往依赖标量距离或表面固有特征，使得跨分子几何在很大程度上保持隐式。我们提出了AbGaze，一种基于注意力几何表示学习的端到端抗体设计框架，该框架编码了相对于抗体残基局部坐标系的抗原表面的距离、空间方向和表面法线方向，并根据其界面上下文自适应地聚合这些几何相互作用。学习到的相互作用表示在多CDR协同设计、复合物结构预测和亲和力优化之间共享，局部坐标几何监督进一步约束该表示。AbGaze在所有三个任务上均优于先前方法：相对于第二好的方法，它在六个CDR上的平均氨基酸恢复率提高了7.1%，结构误差降低了14.9%，界面对接质量（DockQ）提高了6.6%，亲和力提升率（IMP）提高了32.5%。
英文摘要
Computational antibody design requires representations that capture the geometric patterns underlying antigen--antibody interactions, yet existing approaches often rely on scalar distances or surface-intrinsic features, leaving cross-molecular geometry largely implicit. We present AbGaze, an end-to-end antibody design framework based on attentive geometric representation learning, which encodes distance, spatial direction, and surface-normal orientation of antigen surfaces relative to antibody-residue local frames, and adaptively aggregates these geometric interactions according to their interfacial context. The learned interaction representation is shared across multi-CDR co-design, complex structure prediction, and affinity optimization, with local-frame geometric supervision further constraining the representation. AbGaze outperforms prior methods across all three tasks: relative to the second-best method, it improves amino-acid recovery by 7.1% and reduces structural error by 14.9% on average over the six CDRs, improves interface docking quality (DockQ) by 6.6%, and raises the affinity improvement rate (IMP) by 32.5%.
