LAUSR

MOTIVATION In the process of discovery and optimization of lead compounds, it is difficult for non-expert pharmacologists to intuitively determine the contribution of substructure to a particular property of a molecule. RESULTS In this work, we develop a user-friendly web service, named interpretable-ADMET, which predict 59 ADMET-associated properties using 90 qualitative classification models and 28 quantitative regression models based on graph convolutional neural network (GCNN) and graph attention network (GAT) algorithms. In interpretable-ADMET, there are 250,729 entries associated with 59 kinds of absorption, distribution, metabolism, excretion and toxicity (ADMET) associated properties for 80,167 chemical compounds. In addition to making predictions, interpretable-ADMET provides interpretation models based on gradient-weighted class activation map (Grad-CAM) for identifying the substructure which is important to the particular property. Interpretable-ADMET also provides an optimize module to automatically generate a set of novel virtual candidates based on matched molecular pair (MMP) rules. We believe that interpretable-ADMET could serve as a useful tool for lead optimization in drug discovery. AVAILABILITY Interpretable-ADMET is available at http://cadd.pharmacy.nankai.edu.cn/interpretableadmet/. SUPPLEMENTARY INFORMATION Supplementary data are available at Bioinformatics online.