ADME prediction for Breast Cancer Drugs in Computer-Aided Drug Design
Ziyue Fan, Shuyue Wang, Zhongsui Xie, Zhijun Li · 2022
ERα plays an important role in breast tumor development and is regarded as an important target for breast cancer treatment. In order to assist the drug design for breast cancer target ERα, a machine learning method named Adaptive Boosting Extremely Random Tree (ABERT) is proposed in this paper, which is applied to construct the prediction model for the absorption, distribution, metabolism and excretion (ADME) of candidate compounds. Moreover, traditional machine learning and deep learning models are used to compare and reveal the advantages of our model. Besides, an external data set is used to verify the accuracy and stability of our model. The results show that our model has high accuracy in predicting ADME properties and can help to study the pharmacokinetics of candidate drugs as well as the Computer-Aided Drug Design (CADD) in early drug discovery for breast cancer.