Discovery of Dual FGFR4 and EGFR Inhibitors by Machine Learning and Biological Evaluation.

Kinase inhibitors are widely used in antitumor research, but there are still many problems such as drug resistance and off-target toxicity. A more suitable solution is to design a multitarget inhibitor with certain selectivity. Herein, computational and experimental studies were applied to the discovery of dual inhibitors against FGFR4 and EGFR. A quantitative structure-property relationship (QSPR) study was carried out to predict the FGFR4 and EGFR activity of a data set consisting of 843 and 5088 compounds, respectively. Four different machine learning methods including support vector machine (SVM), random forest (RF), gradient boost regression tree (GBRT), and XGBoost (XGB) were built using the most suitable features selected by the mutual information algorithm. As for FGFR4 and EGFR, SVM showed the best performance with R2test-FGFR4 = 0.80 and R2test-EGFR = 0.75, demonstrating excellent model stability, which was used to predict the activity of some compounds from an in-house database. Finally, compound 1 was selected, which exhibits inhibitory activity against FGFR4 (IC50 = 86.2 nM) and EGFR (IC50 = 83.9 nM) kinase, respectively. Furthermore, molecular docking and molecular dynamics simulations were performed to identify key amino acids for the interaction of compound 1 with FGFR4 and EGFR. In this paper, the machine-learning-based QSAR models were established and effectively applied to the discovery of dual-target inhibitors against FGFR4 and EGFR, demonstrating the great potential of machine learning strategies in dual inhibitor discovery.

In Silico Design and Analysis of a Kinase-Focused Combinatorial Library Considering Diversity and Quality.

A structurally diverse, high-quality, and kinase-focused database plays a critical role in finding hits or leads in kinase drug discovery. Here, we propose a workflow for designing a virtual kinase-focused combinatorial library using existing structures. Based on the analysis of known protein kinase inhibitors (PKIs), detailed fragment optimization, fragment selection, fragment linking, and a molecular filtering scheme were defined. Quick recognition of core fragments that can possibly form dual hydrogen bonds with the hinge region of the ATP-pocket was proposed. Furthermore, three diversity and four quality metrics were chosen for compound library analysis, which can be applied to databases with over 30 million structures. Compared with 13 commercial libraries, our protocol demonstrates a special advantage in terms of good skeleton diversity, acceptable fingerprint diversity, balanced scaffold distribution, and high quality, which can work well not only on existing PKIs, but also on four chosen commercial libraries. Overall, the strategy can greatly facilitate the expansion of a desirable chemical space for kinase drug discovery.

Protein-ligand interaction-guided discovery of novel VEGFR-2 inhibitors.

As an effective target in abnormal angiogenesis-related tumor treatment, VEGFR-2 has small-molecule inhibitors of various scaffolds being approved for treating diseases such as renal carcinoma, non-small cell lung cancer, etc. However, endogenous and acquired drug resistance are still considered to be the main contributors for the failure of VEGFR-2 clinical candidates. Therefore, development of novel VEGFR-2 inhibitors is still urgently needed in the market but also challenging. In this work, residues including Asp1046, Ile1025, HIS1026, Cys919 and Lys868 were identified as the most important residues for Hbonded interaction, while His1026, Asp1046, Glu885, Ile1025 and Leu840 exhibited critical role for the nonbonded interactions through a comprehensive analysis of protein-ligand interactions, which plays critical roles in the binding of compounds and targets. Guided by the analysis of binding interactions, a total of 10 novel VEGFR-2 inhibitors based on N-methyl-4-oxo-N-propyl-1,4-dihydroquinoline-2-carboxamide scaffold were discovered through fragment-based drug design and structure-based virtual screening, which expands the chemical space of current VEGFR-2 inhibitors. Biological activity evaluation showed that even though the enzymatic activity of these compounds against VEGFR-2 were inferior to that of the positive controls sorafenib and motesanib, compound I-10 showed moderate HepG2 cell inhibitory activity with an IC50 value of 33.65 µM and eight compounds exhibited moderate or higher HUVEC inhibitory activity in the range of 19.54-57.98 µM compared to the controls. Particularly, the HUVEC inhibitory activity of compound I-6 (IC50 = 19.54 µM) outperformed motesanib and can be used as starting points for further optimization and development for cancer treatment.Communicated by Ramaswamy H. Sarma.

Discovery of potent apoptosis signal-regulating kinase 1 inhibitors via integrated computational strategy and biological evaluation.

Apoptosis signal-regulating Kinase 1 (ASK1) has been confirmed as a potential therapeutic target for the treatment of non-alcoholic steatohepatitis (NASH) disorder and the discovery of ASK1 inhibitors has attracted increasing attention. In this work, a series of in silico methods including pharmacophore screening, docking binding site analysis, protein-ligand interaction fingerprint (PLIF) similarity investigation and molecular docking were applied to find the potential hits from commercial compound databases. Five compounds with potential inhibitory activity were purchased and submitted to biological activity validation. Thus, one hit compound was discovered with micromolar IC50 value (10.59 µM) against ASK1. Results demonstrated that the integration of computation methods and biological test was quite reliable for the discovery of potent ASK1 inhibitors and the strategy could be extended to other similar targets of interest.

Investigation of Crystal Structures in Structure-Based Virtual Screening for Protein Kinase Inhibitors.

Protein kinases are important drug targets in several therapeutic areas ,and structure-based virtual screening (SBVS) is an important strategy in discovering lead compounds for kinase targets. However, there are multiple crystal structures available for each target, and determining which one is the most favorable is a key step in molecular docking for SBVS due to the ligand induce-fit effect. This work aimed to find the most desirable crystal structures for molecular docking by a comprehensive analysis of the protein kinase database which covers 190 different kinases from all eight main kinase families. Through an integrated self-docking and cross-docking evaluation, 86 targets were eventually evaluated on a total of 2608 crystal structures. Results showed that molecular docking has great capability in reproducing conformation of crystallized ligands and for each target, the most favorable crystal structure was selected, and the AGC family outperformed the other family targets based on RMSD comparison. In addition, RMSD values, GlideScore, and corresponding bioactivity data were compared and demonstrated certain relationships. This work provides great convenience for researchers to directly select the optimal crystal structure in SBVS-based kinase drug design and further validates the effectiveness of molecular docking in drug discovery.

Investigation of Machine Intelligence in Compound Cell Activity Classification.

Machine intelligence has been greatly developed in the past decades and has been widely used in many fields. In the recent years, many reports have shown its satisfactory effect in drug discovery. In this study, machine intelligence methods were explored to assist the cell activity prediction. Multiple machine intelligence methods including support vector machine, decision tree, random forest, extra trees, gradient boosting machine, convolutional neural network, long short-term memory network, and gated recurrent unit network were employed to separate compounds based on their cell activity. Different from some reported classification models, compounds were expressed as a string by the simplified molecular input line entry system and directly used as input rather than any chemical descriptors, which mimicked natural language processing. Both the single cell strain and whole data set under the balanced and imbalanced data distributions were discussed, respectively. Different activity cutoffs were set for the single (Z-score = 3) and the whole (Z-score = 5 and 6) data set. Nine metrics were used to evaluate the models including accuracy, precision, recall, f1-score, area under the receiver operating characteristic curve score, Cohen's κ, Brier score, Matthews correlation coefficient, and balanced accuracy. The results show that the gradient boosting machine is competent at balanced data distribution, and convolutional neural network is qualified for the imbalanced one. The results demonstrate that both classic machine learning methods and deep learning methods have potential in classification of compound cell activity.

In Silico Prediction of Human Intravenous Pharmacokinetic Parameters with Improved Accuracy.

Human pharmacokinetics is of great significance in the selection of drug candidates, and in silico estimation of pharmacokinetic parameters in the early stage of drug development has become the trend of drug research owing to its time- and cost-saving advantages. Herein, quantitative structure-property relationship studies were carried out to predict four human pharmacokinetic parameters including volume of distribution at steady state (VDss), clearance (CL), terminal half-life (t1/2), and fraction unbound in plasma (fu), using a data set consisting of 1352 drugs. A series of regression models were built using the most suitable features selected by Boruta algorithm and four machine learning methods including support vector machine (SVM), random forest (RF), gradient boosting machine (GBM), and XGBoost (XGB). For VDss, SVM showed the best performance with R2test = 0.870 and RMSEtest = 0.208. For the other three pharmacokinetic parameters, the RF models produced the superior prediction accuracy (for CL, R2test = 0.875 and RMSEtest = 0.103; for t1/2, R2test = 0.832 and RMSEtest = 0.154; for fu, R2test = 0.818 and RMSEtest = 0.291). Assessed by 10-fold cross validation, leave-one-out cross validation, Y-randomization test and applicability domain evaluation, these models demonstrated excellent stability and predictive ability. Compared with other published models for human pharmacokinetic parameters estimation, it was further confirmed that our models obtained better predictive ability and could be used in the selection of preclinical candidates.

Prediction of hERG K+ channel blockage using deep neural networks.

Human ether-a-go-go-related gene (hERG) K+ channel blockage may cause severe cardiac side-effects and has become a serious issue in safety evaluation of drug candidates. Therefore, improving the ability to avoid undesirable hERG activity in the early stage of drug discovery is of significant importance. The purpose of this study was to build predictive models of hERG activity by deep neural networks. For each combination of sampling methods and descriptors, deep neural networks with different architectures were implemented to build classification models. The optimal model M15 with three hidden layers, undersampling method, and 2D descriptors yielded the prediction accuracy of 0.78 and F1 score of 0.75 on the test set as well as accuracy of 0.77 and F1 score of 0.34 on the external validation set, outperforming the other 35 models including 9 random forest models. Particularly, the optimal model M15 achieved the highest F1 score and the second highest accuracy when compared with other five methods from four groups using different machine learning algorithms with the same external validation set. It can be believed that this model has powerful capability on prediction of hERG toxicity, which is of great benefit for developing novel drug candidates.

A combined drug discovery strategy based on machine learning and molecular docking.

Data mining methods based on machine learning play an increasingly important role in drug design and discovery. In the current work, eight machine learning methods including decision trees, k-Nearest neighbor, support vector machines, random forests, extremely randomized trees, AdaBoost, gradient boosting trees, and XGBoost were evaluated comprehensively through a case study of ACC inhibitor data sets. Internal and external data sets were employed for cross-validation of the eight machine learning methods. Results showed that the extremely randomized trees model performed best and was adopted as the first step of virtual screening. Together with structure-based virtual screening in the second step, this combined strategy obtained desirable results. This work indicates that the combination of machine learning methods with traditional structure-based virtual screening can effectively strengthen the ability in finding potential hits from large compound database for a given target.

In silico Prediction of Inhibitory Constant of Thrombin Inhibitors Using Machine Learning.

BACKGROUND: Thrombin is the central protease of the vertebrate blood coagulation cascade, which is closely related to cardiovascular diseases. The inhibitory constant Ki is the most significant property of thrombin inhibitors. METHOD: This study was carried out to predict Ki values of thrombin inhibitors based on a large data set by using machine learning methods. Taking advantage of finding non-intuitive regularities on high-dimensional datasets, machine learning can be used to build effective predictive models. A total of 6554 descriptors for each compound were collected and an efficient descriptor selection method was chosen to find the appropriate descriptors. Four different methods including multiple linear regression (MLR), K Nearest Neighbors (KNN), Gradient Boosting Regression Tree (GBRT) and Support Vector Machine (SVM) were implemented to build prediction models with these selected descriptors. RESULTS: The SVM model was the best one among these methods with R2=0.84, MSE=0.55 for the training set and R2=0.83, MSE=0.56 for the test set. Several validation methods such as yrandomization test and applicability domain evaluation, were adopted to assess the robustness and generalization ability of the model. The final model shows excellent stability and predictive ability and can be employed for rapid estimation of the inhibitory constant, which is full of help for designing novel thrombin inhibitors.

ADME properties evaluation in drug discovery: in silico prediction of blood-brain partitioning.

The absorption, distribution, metabolism and excretion properties are important for drugs, and prediction of these properties in advance will save the cost of drug discovery substantially. The ability to penetrate the blood-brain barrier is critical for drugs targeting central nervous system, which is represented by the ratio of its concentration in brain and in blood. Herein, a quantitative structure-property relationship study was carried out to predict blood-brain partitioning coefficient (logBB) of a data set consisting of 287 compounds. Four different methods including support vector machine, multivariate linear regression, multivariate adaptive regression splines and random forest were employed to build prediction models with 116 molecular descriptors selected by Boruta algorithm. The RF model had best performance in training set ([Formula: see text] = 0.938), test set ([Formula: see text] = 0.840) and tenfold cross-validation ([Formula: see text] = 0.788). Finally, we found that the polar surface area and octanol-water partition coefficient have the greatest influence on blood-brain partitioning. Results suggest that the proposed model is a useful and practical tool to predict the logBB values of drug candidates.