SepPCNET: Deeping Learning on a 3D Surface Electrostatic Potential Point Cloud for Enhanced Toxicity Classification and Its Application to Suspected Environmental Estrogens.
SepPCNET: Deeping Learning on a 3D Surface Electrostatic Potential Point Cloud for Enhanced Toxicity Classification and Its Application to Suspected Environmental Estrogens.
复制标题
DOI：
10.1021/acs.est.1c01228
复制
发表时间：
2021-07
影响因子：
11.4
通讯作者：
Liguo Wang;Lu Zhao;Xian Liu;Jianjie Fu;A. Zhang
中科院分区：
文献类型：
--
作者：
Liguo Wang;Lu Zhao;Xian Liu;Jianjie Fu;A. Zhang
深度学习（DL）提供了一个前所未有的机会，可以彻底改变基于大数据时代定量构效关系（QSAR）研究的毒性预测前景。然而，在已报道的DL-QSAR模型的结构描述仍然局限于二维水平。受点云几何数据结构的启发，提出了一种新的三维分子表面静电势点云模型（SepPC）.化学物质的每个表面点都被分配了其3D坐标和分子静电势。然后引入了一种新的DL架构SepPCNET，以直接消耗无序的SepPC数据进行毒性分类。SepPCNET模型在ToxCast程序的18种雌激素受体相关测定的电池中测试的1317种化学物质上进行训练。所获得的模型分别以82.8%和88.9%的准确率识别活性和非活性化学品，内部测试集的总准确率为88.3%，外部测试集的总准确率为92.5%，优于其他最新的机器学习模型，并成功识别异构体活性的差异。通过可视化临界点和提取活性化学物质的数据驱动点特征，还获得了对毒性机制的更多见解。
Deep learning (DL) offers an unprecedented opportunity to revolutionize the landscape of toxicity prediction based on quantitative structure-activity relationship (QSAR) studies in the big data era. However, the structural description in the reported DL-QSAR models is still restricted to the two-dimensional level. Inspired by point clouds, a type of geometric data structure, a novel three-dimensional (3D) molecular surface point cloud with electrostatic potential (SepPC) was proposed to describe chemical structures. Each surface point of a chemical is assigned its 3D coordinate and molecular electrostatic potential. A novel DL architecture SepPCNET was then introduced to directly consume unordered SepPC data for toxicity classification. The SepPCNET model was trained on 1317 chemicals tested in a battery of 18 estrogen receptor-related assays of the ToxCast program. The obtained model recognized the active and inactive chemicals at accuracies of 82.8 and 88.9%, respectively, with a total accuracy of 88.3% on the internal test set and 92.5% on the external test set, which outperformed other up-to-date machine learning models and succeeded in recognizing the difference in the activity of isomers. Additional insights into the toxicity mechanism were also gained by visualizing critical points and extracting data-driven point features of active chemicals.