Efficient Exploration of Chemical Compound Space Using Active Learning for Prediction of Thermodynamic Properties of Alkane Molecules
Efficient Exploration of Chemical Compound Space Using Active Learning for Prediction of Thermodynamic Properties of Alkane Molecules
复制标题
DOI：
10.1021/acs.jcim.3c01430
复制
发表时间：
2023-10
影响因子：
5.6
通讯作者：
Yan Xiang;Yu-Hang Tang;Zheng Gong;Hongyi Liu;Liang Wu;Guang Lin;Huai Sun
中科院分区：
文献类型：
--
作者：
Yan Xiang;Yu-Hang Tang;Zheng Gong;Hongyi Liu;Liang Wu;Guang Lin;Huai Sun
我们引入了一种探索性主动学习（AL）算法，使用高斯过程回归和边缘化图内核（GPR-MGK）以最小的成本对化合物空间（CCS）进行采样。针对251,728个具有4-19个碳原子的烷烃分子，我们应用AL算法选择了一组多样化且有代表性的分子，然后对这些选定的分子进行了高通量分子模拟。为了展示 AL 算法的强大功能，我们使用模拟数据作为训练集构建了定向消息传递神经网络 (D-MPNN)，以预测 CCS 的液体密度、热容和汽化焓。验证表明，基于本工作中考虑的最小训练集（由 313 个分子或原始 CCS 的 0.124% 组成）构建的 D-MPNN 模型，预测的属性相对于计算数据为 R2 > 0.99，相对于实验数据为 R2 > 0.94。所提出的 AL 算法的优点是 GPR 的预测不确定性仅取决于分子结构，这使其与高通量数据生成兼容。
We introduce an exploratory active learning (AL) algorithm using Gaussian process regression and marginalized graph kernel (GPR-MGK) to sample chemical compound space (CCS) at minimal cost. Targeting 251,728 enumerated alkane molecules with 4-19 carbon atoms, we applied the AL algorithm to select a diverse and representative set of molecules and then conducted high-throughput molecular simulations on these selected molecules. To demonstrate the power of the AL algorithm, we built directed message-passing neural networks (D-MPNN) using simulation data as the training set to predict liquid densities, heat capacities, and vaporization enthalpies of the CCS. Validations show that D-MPNN models built on the smallest training set considered in this work, which consists of 313 molecules or 0.124% of the original CCS, predict the properties with R2 > 0.99 against the computational data and R2 > 0.94 against the experimental data. The advantage of the presented AL algorithm is that the predicted uncertainty of GPR depends on only the molecular structures, which renders it compatible with high-throughput data generation.