No abstract

Drug–target interactions (DTIs) play a crucial role in target-based drug discovery and development. Computational prediction of DTIs can effectively complement experimental wet-lab techniques for the identification of DTIs, which are typically time- and resource-consuming. However, the performances of the current DTI prediction approaches suffer from a problem of low precision and high false-positive rate. In this study, we aim to develop a novel DTI prediction method for improving the prediction performance based on a cascade deep forest (CDF) model, named DTI-CDF, with multiple similarity-based features between drugs and the similarity-based features between target proteins extracted from the heterogeneous graph, which contains known DTIs. In the experiments, we built five replicates of 10-fold cross-validation under three different experimental settings of data sets, namely, corresponding DTI values of certain drugs (SD), targets (ST), or drug-target pairs (SP) in the training sets are missed but existed in the test sets. The experimental results demonstrate that our proposed approach DTI-CDF achieves a significantly higher performance than that of the traditional ensemble learning-based methods such as random forest and XGBoost, deep neural network, and the state-of-the-art methods such as DDR. Furthermore, there are 1352 newly predicted DTIs which are proved to be correct by KEGG and DrugBank databases. The data sets and source code are freely available at https://github.com//a96123155/DTI-CDF.

Drug discovery is an academical and commercial process of global importance. Accurate identification of drug-target interactions (DTIs) can significantly facilitate the drug discovery process. Compared to the costly, labor-intensive and time-consuming experimental methods, machine learning (ML) plays an ever-increasingly important role in effective, efficient and high-throughput identification of DTIs. However, upstream feature extraction methods require tremendous human resources and expert insights, which limits the application of ML approaches. Inspired by the unsupervised representation learning methods like Word2vec, we here proposed SPVec, a novel way to automatically represent raw data such as SMILES strings and protein sequences into continuous, information-rich and lower-dimensional vectors, so as to avoid the sparseness and bit collisions from the cumbersomely manually extracted features. Visualization of SPVec nicely illustrated that the similar compounds or proteins occupy similar vector space, which indicated that SPVec not only encodes compound substructures or protein sequences efficiently, but also implicitly reveals some important biophysical and biochemical patterns. Compared with manually-designed features like MACCS fingerprints and amino acid composition (AAC), SPVec showed better performance with several state-of-art machine learning classifiers such as Gradient Boosting Decision Tree, Random Forest and Deep Neural Network on BindingDB. The performance and robustness of SPVec were also confirmed on independent test sets obtained from DrugBank database. Also, based on the whole DrugBank dataset, we predicted the possibilities of all unlabeled DTIs, where two of the top five predicted novel DTIs were supported by external evidences. These results indicated that SPVec can provide an effective and efficient way to discover reliable DTIs, which would be beneficial for drug reprofiling.

A drug may be metabolized by multiple cytochrome P450 (CYP450) isoforms. Predicting the metabolic fate of drugs is very important to prevent drug− drug interactions in the development of novel pharmaceuticals. Prediction of CYP450 enzyme−substrate selectivity is formulized as a multilabel learning task in this study. First, we compared the performance of feature combinations based on four different categories of features, which are physiochemical property descriptors, mol2vec descriptors, extended connectivity fingerprints, and molecular access system key fingerprints on modeling. After identifying the best combination of features, we applied seven different multilabel models, which are multilabel k-nearest neighbor (ML-kNN), multilabel twin support vector machine, and five network-based label space division (NLSD)-based methods (NLSD-MLP, NLSD-XGB, NLSD-EXT, NLSD-RF, and NLSD-SVM). All of the six models (ML-kNN, NLSD-MLP, NLSD-XGB, NLSD-EXT, NLSD-RF, and NLSD-SVM) in this paper exhibit better performances than the previous work. Besides, NLSD-XGB achieves the best performance with the average top-1 prediction success of 91.1%, the average top-2 prediction success of 96.2%, and the average top-3 prediction success of 98.2%. When compared with the previous work, NLSD-XGB shows a significant improvement over 11% on top-1 in the 10 times repeated 5-fold cross-validation test and over 14% on top-1 in the 10 times repeated hold-out method. To the best of our knowledge, the network-based label space division model is first introduced in drug metabolism and performs well in this task.

Identifying drug-target interactions (DTIs) is an important step for drug discovery and drug repositioning. To reduce the experimental cost, a large number of computational approaches have been proposed for this task. The machine learning-based models, especially binary classification models, have been developed to predict whether a drug-target pair interacts or not. However, there is still much room for improvement in the performance of current methods. Multi-label learning can overcome some difficulties caused by single-label learning in order to improve the predictive performance. The key challenge faced by multi-label learning is the exponential-sized output space, and considering label correlations can help to overcome this challenge. In this paper, we facilitate multi-label classification by introducing community detection methods for DTI prediction, named DTI-MLCD. Moreover, we updated the gold standard data set by adding 15,000 more positive DTI samples in comparison to the data set, which has widely been used by most of previously published DTI prediction methods since 2008. The proposed DTI-MLCD is applied to both data sets, demonstrating its superiority over other machine learning methods and several existing methods. The data sets and source code of this study are freely available at https://github.com/a96123155/DTI-MLCD.

We construct a protein–protein interaction (PPI) targeted drug-likeness dataset and propose a deep molecular generative framework to generate novel drug-likeness molecules from the features of the seed compounds. This framework gains inspiration from published molecular generative models, uses the key features associated with PPI inhibitors as input and develops deep molecular generative models for de novo molecular design of PPI inhibitors. For the first time, quantitative estimation index for compounds targeting PPI was applied to the evaluation of the molecular generation model for de novo design of PPI-targeted compounds. Our results estimated that the generated molecules had better PPI-targeted drug-likeness and drug-likeness. Additionally, our model also exhibits comparable performance to other several state-of-the-art molecule generation models. The generated molecules share chemical space with iPPI-DB inhibitors as demonstrated by chemical space analysis. The peptide characterization-oriented design of PPI inhibitors and the ligand-based design of PPI inhibitors are explored. Finally, we recommend that this framework will be an important step forward for the de novo design of PPI-targeted therapeutics.

One of the main problems with the joint use of multiple drugs is that it may cause adverse drug interactions and side effects that damage the body. Therefore, it is important to predict potential drug interactions. However, most of the available prediction methods can only predict whether two drugs interact or not, whereas few methods can predict interaction events between two drugs. Accurately predicting interaction events of two drugs is more useful for researchers to study the mechanism of the interaction of two drugs. In the present study, we propose a novel method, MDF-SA-DDI, which predicts drug–drug interaction (DDI) events based on multi-source drug fusion, multi-source feature fusion and transformer self-attention mechanism. MDF-SA-DDI is mainly composed of two parts: multi-source drug fusion and multi-source feature fusion. First, we combine two drugs in four different ways and input the combined drug feature representation into four different drug fusion networks (Siamese network, convolutional neural network and two auto-encoders) to obtain the latent feature vectors of the drug pairs, in which the two auto-encoders have the same structure, and their main difference is the number of neurons in the input layer of the two auto-encoders. Then, we use transformer blocks that include self-attention mechanism to perform latent feature fusion. We conducted experiments on three different tasks with two datasets. On the small dataset, the area under the precision–recall-curve (AUPR) and F1 scores of our method on task 1 reached 0.9737 and 0.8878, respectively, which were better than the state-of-the-art method. On the large dataset, the AUPR and F1 scores of our method on task 1 reached 0.9773 and 0.9117, respectively. In task 2 and task 3 of two datasets, our method also achieved the same or better performance as the state-of-the-art method. More importantly, the case studies on five DDI events are conducted and achieved satisfactory performance. The source codes and data are available at https://github.com/ShenggengLin/MDF-SA-DDI.

Accurate identification of the miRNA-disease associations (MDAs) helps to understand the etiology and mechanisms of various diseases. However, the experimental methods are costly and time-consuming. Thus, it is urgent to develop computational methods towards the prediction of MDAs. Based on the graph theory, the MDA prediction is regarded as a node classification task in the present study. To solve this task, we propose a novel method MDA-GCNFTG, which predicts MDAs based on Graph Convolutional Networks (GCNs) via graph sampling through the Feature and Topology Graph to improve the training efficiency and accuracy. This method models both the potential connections of feature space and the structural relationships of MDA data. The nodes of the graphs are represented by the disease semantic similarity, miRNA functional similarity and Gaussian interaction profile kernel similarity. Moreover, we considered six tasks simultaneously on the MDA prediction problem at the first time, which ensure that under both balanced and unbalanced sample distribution, MDA-GCNFTG can predict not only new MDAs but also new diseases without known related miRNAs and new miRNAs without known related diseases. The results of 5-fold cross-validation show that the MDA-GCNFTG method has achieved satisfactory performance on all six tasks and is significantly superior to the classic machine learning methods and the state-of-the-art MDA prediction methods. Moreover, the effectiveness of GCNs via the graph sampling strategy and the feature and topology graph in MDA-GCNFTG has also been demonstrated. More importantly, case studies for two diseases and three miRNAs are conducted and achieved satisfactory performance.