Torsional Diffusion for Molecular Conformer Generation
Bowen Jing, 1 Gabriele Corso,†† Jeffrey Chang,2 Regina Barzilay,1 Tommi Jaakkola1 1CSAIL, Massachusetts Institute of Technology 2Dept. of Physics, Harvard University Equal contribution. Correspondence to {bjing,
Abstract
Molecular conformer generation is a fundamental task in computational chemistry. Several machine learning approaches have been developed, but none have outperformed state-of-the-art cheminformatics methods. We propose torsional diffusion, a novel diffusion framework that operates on the space of torsion angles via a diffusion process on the hypertorus and an extrinsic-to-intrinsic score model. On a standard benchmark of drug-like molecules, torsional diffusion generates superior conformer ensembles compared to machine learning and cheminformatics methods in terms of both RMSD and chemical properties, and is orders of magnitude faster than previous diffusion-based models. Moreover, our model provides exact likelihoods, which we employ to build the first generalizable Boltzmann generator. Code is available at https://github.com/gcorso/torsional-diffusion.
中文速览
分子构象生成（conformer generation）是计算化学的核心任务，目标是预测一个小分子在三维空间中可能采取的低能量立体结构。扭转扩散（torsional diffusion）方法的关键洞见在于：分子构象的自由度主要来自可旋转键的扭转角，因此只需对扭转角空间（数学上构成一个超环面）建立扩散生成模型，而把键长、键角等其余自由度交由传统化学信息学方法处理，从而将采样空间的维度大幅压缩。为了在这个非欧几里得空间上学习打分函数，作者设计了一个"外在坐标到内在坐标"的打分网络，以三维点云作为输入，利用SE(3)等变图神经网络为每个可旋转键输出扭转角梯度，优雅地解决了不同分子扭转空间维度各异的问题。在标准药物分子基准数据集GEOM-DRUGS上，torsional diffusion在均方根偏差（RMSD）和化学性质两个维度上均超越了现有机器学习方法及顶尖商业软件OMEGA，且去噪步数比同类扩散模型少两个数量级；此外，该模型可提供精确的似然值，由此构建出首个可推广到未见分子的玻尔兹曼生成器（Boltzmann generator），对药物筛选等高通量应用具有重要意义。
原文 arXiv:2206.01729;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2206.01729v2