Comprehensive strategies of machine-learning-based quantitative structure-activity relationship models

Early quantitative structure-activity relationship (QSAR) technologies have unsatisfactory versatility and accuracy in fields such as drug discovery because they are based on traditional machine learning and interpretive expert features. The development of Big Data and deep learning technologies significantly improve the processing of unstructured data and unleash the great potential of QSAR. Here we discuss the integration of wet experiments (which provide experimental data and reliable verification), molecular dynamics simulation (which provides mechanistic interpretation at the atomic/molecular levels), and machine learning (including deep learning) techniques to improve QSAR models. We first review the history of traditional QSAR and point out its problems. We then propose a better QSAR model characterized by a new iterative framework to integrate machine learning with disparate data input. Finally, we discuss the application of QSAR and machine learning to many practical research fields, including drug development and clinical trials.

Classification and QSAR models of leukotriene A4 hydrolase (LTA4H) inhibitors by machine learning methods

Leukotriene A4 hydrolase (LTA4H) is an important anti-inflammatory target which can convert leukotriene A4 (LTA4) into pro-inflammatory substance leukotriene B4 (LTB4). In this paper, we built 18 classification models for 463 LTA4H inhibitors by using support vector machine (SVM), random forest (RF) and K-Nearest Neighbour (KNN). The best classification model (Model 2A) was built from RF and MACCS fingerprints. The prediction accuracy of 88.96% and the Matthews correlation coefficient (MCC) of 0.74 had been achieved on the test set. We also divided the 463 LTA4H inhibitors into six subsets using K-Means. We found that the highly active LTA4H inhibitors mostly contained diphenylmethane or diphenyl ether as the scaffold and pyridine or piperidine as the side chain. In addition, six quantitative structure-activity relationship (QSAR) models for 172 LTA4H inhibitors were built by multiple linear regression (MLR) and SVM. The best QSAR model (Model 6A) was built by using SVM and CORINA Symphony descriptors. The coefficients of determination of the training set and the test set were equal to 0.81 and 0.79, respectively. Classification and QSAR models could be used for subsequent virtual screening, and the obtained fragments that were important for highly active inhibitors would be helpful for designing new LTA4H inhibitors.