Search

The n -octanol/buffer solution distribution coefficient at pH = 7.4 (log  D 7.4 ) is an indicator of lipophilicity, and it influences a wide variety of absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties and druggability of compounds. In log  D 7.4 prediction, graph neural networks (GNNs) can uncover subtle structure-property relationships (SPRs) by automatically extracting features from molecular graphs that facilitate the learning of SPRs, but their performances are often limited by the small size of available datasets. Herein, we present a transfer learning strategy called pretraining on computational data and then fine-tuning on experimental data (PCFE) to fully exploit the predictive potential of GNNs. PCFE works by pretraining a GNN model on 1.71 million computational log  D data (low-fidelity data) and then fine-tuning it on 19,155 experimental log  D 7.4 data (high-fidelity data). The experiments for three GNN architectures (graph convolutional network (GCN), graph attention network (GAT), and Attentive FP) demonstrated the effectiveness of PCFE in improving GNNs for log  D 7.4 predictions. Moreover, the optimal PCFE-trained GNN model (cx-Attentive FP, R test 2 = 0.909) outperformed four excellent descriptor-based models (random forest (RF), gradient boosting (GB), support vector machine (SVM), and extreme gradient boosting (XGBoost)). The robustness of the cx-Attentive FP model was also confirmed by evaluating the models with different training data sizes and dataset splitting strategies. Therefore, we developed a webserver and defined the applicability domain for this model. The webserver (http://tools.scbdd.com/chemlogd/) provides free log  D 7.4 prediction services. In addition, the important descriptors for log  D 7.4 were detected by the Shapley additive explanations (SHAP) method, and the most relevant substructures of log  D 7.4 were identified by the attention mechanism. Finally, the matched molecular pair analysis (MMPA) was performed to summarize the contributions of common chemical substituents to log  D 7.4 , including a variety of hydrocarbon groups, halogen groups, heteroatoms, and polar groups. In conclusion, we believe that the cx-Attentive FP model can serve as a reliable tool to predict log  D 7.4 and hope that pretraining on low-fidelity data can help GNNs make accurate predictions of other endpoints in drug discovery.

Hematotoxicity has been becoming a serious but overlooked toxicity in drug discovery. However, only a few in silico models have been reported for the prediction of hematotoxicity. In this study, we constructed a high-quality dataset comprising 759 hematotoxic compounds and 1623 nonhematotoxic compounds and then established a series of classification models based on a combination of seven machine learning (ML) algorithms and nine molecular representations. The results based on two data partitioning strategies and applicability domain (AD) analysis illustrate that the best prediction model based on Attentive FP yielded a balanced accuracy (BA) of 72.6%, an area under the receiver operating characteristic curve (AUC) value of 76.8% for the validation set, and a BA of 69.2%, an AUC of 75.9% for the test set. In addition, compared with existing filtering rules and models, our model achieved the highest BA value of 67.5% for the external validation set. Additionally, the shapley additive explanation (SHAP) and atom heatmap approaches were utilized to discover the important features and structural fragments related to hematotoxicity, which could offer helpful tips to detect undesired positive substances. Furthermore, matched molecular pair analysis (MMPA) and representative substructure derivation technique were employed to further characterize and investigate the transformation principles and distinctive structural features of hematotoxic chemicals. We believe that the novel graph-based deep learning algorithms and insightful interpretation presented in this study can be used as a trustworthy and effective tool to assess hematotoxicity in the development of new drugs.

Lipophilicity, as evaluated by the n -octanol/buffer solution distribution coefficient at pH = 7.4 (log  D 7.4 ), is a major determinant of various absorption, distribution, metabolism, elimination, and toxicology (ADMET) parameters of drug candidates. In this study, we developed several quantitative structure-property relationship (QSPR) models to predict log  D 7.4 based on a large and structurally diverse data set. Eight popular machine learning algorithms were employed to build the prediction models with 43 molecular descriptors selected by a wrapper feature selection method. The results demonstrated that XGBoost yielded better prediction performance than any other single model ( R T 2 = 0.906 and RMSE T = 0.395). Moreover, the consensus model from the top three models could continue to improve the prediction performance ( R T 2 = 0.922 and RMSE T = 0.359). The robustness, reliability, and generalization ability of the models were strictly evaluated by the Y-randomization test and applicability domain analysis. Moreover, the group contribution model based on 110 atom types and the local models for different ionization states were also established and compared to the global models. The results demonstrated that the descriptor-based consensus model is superior to the group contribution method, and the local models have no advantage over the global models. Finally, matched molecular pair (MMP) analysis and descriptor importance analysis were performed to extract transformation rules and give some explanations related to log  D 7.4 . In conclusion, we believe that the consensus model developed in this study can be used as a reliable and promising tool to evaluate log  D 7.4 in drug discovery.