Drug Metabolism and Pharmacokinetics Japan, Biopharmaceutical Assessment Core Function Unit, Medicine Development Center, Eisai Co., Ltd., Tokyo, Japan.
Department of Computer Science, Graduate School of Information Science and Engineering, Tokyo Institute of Technology, Tokyo, Japan.
Pharm Res. 2018 Aug 24;35(10):197. doi: 10.1007/s11095-018-2479-1.
The clearance pathways of drugs are critical elements for understanding the pharmacokinetics of drugs. We previously developed in silico systems to predict the five clearance pathway using a rectangular method and a support vector machine (SVM). In this study, we improved our classification system by increasing the number of clearance pathways available for our prediction (CYP1A2, CYP2C8, CYP2C19, and UDP-glucuronosyl transferases (UGTs)) and by accepting multiple major pathways.
Using the four default descriptors (charge, molecular weight, logD at pH 7.0, and unbound fraction in plasma), three kinds of SVM-based predictors based on traditional single-step approach or two-step focusing approaches with subset or partition clustering were developed. The two-step approach with subset clustering resulted in the highest prediction performance. The feature-selection of additional descriptors based on a greedy algorithm was employed to further improve the predictability.
The prediction accuracy for each pathway was increased to more than 0.83 with the exception of CYP2C19 and UGTs pathways, whose accuracies were below 0.7. Prediction performance of CYP1A2, CYP3A4 and renal excretion pathways were found to be acceptable using external dataset.
We successfully constructed a novel SVM-based predictor for the multiple major clearance pathways based on chemical structures.
药物的清除途径是理解药物药代动力学的关键因素。我们之前开发了基于矩方法和支持向量机（SVM）的计算系统，用于预测五种清除途径。在这项研究中，我们通过增加可供预测的清除途径数量（CYP1A2、CYP2C8、CYP2C19 和 UDP-葡萄糖醛酸转移酶（UGTs））并接受多种主要途径，改进了我们的分类系统。
使用四个默认描述符（电荷、分子量、pH7.0 时的logD 和血浆中未结合分数），开发了三种基于 SVM 的预测器，基于传统的单步方法或两步聚焦方法，具有子集或分区聚类。基于子集聚类的两步方法产生了最高的预测性能。还采用基于贪婪算法的附加描述符特征选择来进一步提高可预测性。
除 CYP2C19 和 UGTs 途径外，每个途径的预测准确性都提高到 0.83 以上，而 CYP2C19 和 UGTs 途径的准确性低于 0.7。使用外部数据集发现 CYP1A2、CYP3A4 和肾排泄途径的预测性能可以接受。
我们成功地基于化学结构构建了一种用于多种主要清除途径的新型 SVM 基预测器。
