Many protein design applications, such as binder or enzyme design, require scaffolding a structural motif with high precision. Generative modelling paradigms based on denoising diffusion processes emerged as a leading candidate to address this motif scaffolding problem and have shown early experimental success in some cases. In the diffusion paradigm, motif scaffolding is treated as a conditional generation task, and several conditional generation protocols were proposed or imported from the Computer Vision literature. However, most of these protocols are motivated heuristically, e.g. via analogies to Langevin dynamics, and lack a unifying framework, obscuring connections between the different approaches. In this work, we unify conditional training and conditional sampling procedures under one common framework based on the mathematically well-understood Doob's h-transform. This new perspective allows us to draw connections between existing methods and propose a new variation on existing conditional training protocols. We illustrate the effectiveness of this new protocol in both, image outpainting and motif scaffolding and find that it outperforms standard methods.

翻译：许多蛋白质设计应用，如结合剂或酶的设计，需要高精度地构建结构基序支架。基于去噪扩散过程的生成建模范式成为解决这一基序支架问题的主要候选方法，并在某些案例中显示出早期实验成功。在扩散范式中，基序支架被视为条件生成任务，已有多种条件生成协议被提出或从计算机视觉文献中引入。然而，这些协议大多基于启发式动机（例如通过类比朗之万动力学），且缺乏统一框架，导致不同方法之间的关联模糊不清。本研究基于数学上理解充分的Doob h-变换，将条件训练和条件采样过程统一在一个共同框架下。这一新视角使我们能够建立现有方法之间的联系，并在现有条件训练协议的基础上提出一种新变体。我们通过图像外推和基序支架两个实验验证了新协议的有效性，发现其性能优于标准方法。