Properties such as absorption, distribution, metabolism, excretion, and toxicity (ADMET) are important in small molecule drug discovery and therapeutics. It is reported that many clinical trials fail due to the deficiencies in ADMET properties. While profiling ADMET in the early stage of drug discovery is desirable, experimental evaluation of ADMET properties is costly with limited available data.
Here, we apply an ensemble of features, including fingerprints and descriptors, and a tree-based machine learning model, extreme gradient boosting, for accurate ADMET prediction.
The models were trained on Therapeutics Data Commons ADMET Benchmark Group. For 22 tasks, our model is ranked first in 11 tasks and top 3 in 19 tasks. The codes are publicly available at GitHub.
If you find this useful, please cite: Tian, H., Ketkar, R. & Tao, P. ADMETboost: a web server for accurate ADMET prediction. J Mol Model 28, 408 (2022). https://doi.org/10.1007/s00894-022-05373-8.