用神经场建模分子密度，实现无结构假设的3D分子快速生成
Score-based 3D molecule generation with neural fields
- 基于连续原子密度场表示，结合行走-跳跃采样生成3D分子
- 在药物分子上表现优异，生成速度比同类方法快一个数量级
- 适合生成复杂大环肽，无需预设分子结构
我们提出一种基于连续原子密度场的3D分子新表示方法。基于此，我们设计了名为FuncMol的新模型，采用行走-跳跃采样策略，在连续空间中进行无条件3D分子生成。FuncMol通过条件神经场将分子场编码为潜在码，利用Langevin MCMC从高斯平滑分布中采样噪声码（行走），单步去噪后直接解码为分子场（跳跃）。该方法无需对分子结构做假设，可随分子规模良好扩展，相较于多数现有方法具有显著优势。在药物类分子上表现竞争力，且能轻松拓展至大环肽生成，采样速度至少提升一个数量级。代码已开源：https://github.com/prescient-design/funcmol。
原文摘要 · Abstract (English)
We introduce a new representation for 3D molecules based on their continuous atomic density fields. Using this representation, we propose a new model based on walk-jump sampling for unconditional 3D molecule generation in the continuous space using neural fields. Our model, FuncMol, encodes molecular fields into latent codes using a conditional neural field, samples noisy codes from a Gaussian-smoothed distribution with Langevin MCMC (walk), denoises these samples in a single step (jump), and finally decodes them into molecular fields. FuncMol performs all-atom generation of 3D molecules without assumptions on the molecular structure and scales well with the size of molecules, unlike most approaches. Our method achieves competitive results on drug-like molecules and easily scales to macro-cyclic peptides, with at least one order of magnitude faster sampling. The code is available at https://github.com/prescient-design/funcmol.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。