Determining the target genes that interact with drugs—drug–target interactions—plays an important role in drug discovery. Identification of drug–target interactions through biological experiments is time consuming, laborious, and costly. Therefore, using computational approaches to predict candidate targets is a good way to reduce the cost of wet-lab experiments. However, the known interactions (positive samples) and the unknown interactions (negative samples) display a serious class imbalance, which has an adverse effect on the accuracy of the prediction results. To mitigate the impact of class imbalance and completely exploit the negative samples, we proposed a new method, named DTIGBDT, based on gradient boosting decision trees, for predicting candidate drug–target interactions. We constructed a drug–target heterogeneous network that contains the drug similarities based on the chemical structures of drugs, the target similarities based on target sequences, and the known drug–target interactions. The topological information of the network was captured by random walks to update the similarities between drugs or targets. The paths between drugs and targets could be divided into multiple categories, and the features of each category of paths were extracted. We constructed a prediction model based on gradient boosting decision trees. The model establishes multiple decision trees with the extracted features and obtains the interaction scores between drugs and targets. DTIGBDT is a method of ensemble learning, and it effectively reduces the impact of class imbalance. The experimental results indicate that DTIGBDT outperforms several state-of-the-art methods for drug–target interaction prediction. In addition, case studies on Quetiapine, Clozapine, Olanzapine, Aripiprazole, and Ziprasidone demonstrate the ability of DTIGBDT to discover potential drug–target interactions.

The computational prediction of novel drug-target interactions (DTIs) may hasten drug repositioning while reducing costs. Most previous methods integrated multiple kinds of connections between drugs and targets by constructing shallow models. These methods failed to learn low-dimension feature vectors for drugs and targets and ignored their distribution. We proposed a graph convolutional network and generative adversarial network-based method, GANDTI, to predict drug-target interactions. We constructed a drug-target heterogeneous network to integrate various connections between drugs and targets. A graph convolutional autoencoder was established to learn the network embedding of the drug and target nodes in a low-dimensional feature space. This encoder deeply integrated the various connections and attributes of drug and target nodes. A generative adversarial network was introduced to regularize the feature vectors into a Gaussian distribution. Severe class imbalance exists between known and unknown DTIs. We built a classifier based on the ensemble learning model, LightGBM, to evaluate the interaction propensities of drugs and targets. This classifier completely exploited all unknown DTIs and counteracted the negative effects of class imbalance. The experimental results indicated that GANDTI performs better than several state-of-the-art methods. Additionally, case studies of five drugs demonstrated the ability of GANDTI to discover potential drugs' targets.

Identifying novel indications for approved drugs can accelerate drug development and reduce research costs. Most previous studies used shallow models for prioritizing the potential drug-related diseases and failed to deeply integrate the paths between drugs and diseases which may contain additional association information. A deep-learning-based method for predicting drug–disease associations by integrating useful information is needed. We proposed a novel method based on a convolutional neural network (CNN) and bidirectional long short-term memory (BiLSTM)—CBPred—for predicting drug-related diseases. Our method deeply integrates similarities and associations between drugs and diseases, and paths among drug-disease pairs. The CNN-based framework focuses on learning the original representation of a drug-disease pair from their similarities and associations. As the drug-disease association possibility also depends on the multiple paths between them, the BiLSTM-based framework mainly learns the path representation of the drug-disease pair. In addition, considering that different paths have discriminate contributions to the association prediction, an attention mechanism at path level is constructed. Our method, CBPred, showed better performance and retrieved more real associations in the front of the results, which is more important for biologists. Case studies further confirmed that CBPred can discover potential drug-disease associations.

Exploring the potential drug–target interactions (DTIs) is a key step in drug discovery and repurposing. In recent years, predicting the probable DTIs through computational methods has gradually become a research hot spot. However, most of the previous studies failed to judiciously take into account the consistency between the chemical properties of drug and its functions. The changes of these relationships may lead to a severely negative effect on the prediction of DTIs. We propose an autoencoder-based method, AEFS, under spatial consistency constraints to predict DTIs. A heterogeneous network is established to integrate the information of drugs, proteins and diseases. The original drug features are projected to an embedding (protein) space by a multi-layer encoder, and further projected into label (disease) space by a decoder. In this process, the clinical information of drugs is introduced to assist the DTI prediction. By maintaining the distribution of drug correlation in the original feature, embedding and label space, AEFS keeps the consistency between chemical properties and functions of drugs. Experimental comparisons indicate that AEFS is more robust for imbalanced data and of significantly superior performance in DTI prediction. Case studies further confirm its ability to mine the latent DTIs. The code of AEFS is available at https://github.com/JackieSun818/AEFS. Supplementary data are available at Bioinformatics online.

Although several traditional Chinese medicine (TCM)-related databases have emerged, they focus on researching single medicinal materials, which is far from sufficient for clinical research and application. In comparison, compound prescriptions are more informative and meaningful in TCM, for they embody the information on the compatibility of TCM besides the relatively isolated information about single medicinal materials. The compatibility information is essential in TCM because it conveys not only what components are involved to treat special diseases but also how to combine these single medical materials. We established a database of Chinese patent medicine and compound prescription (CPMCP). It demonstrates the prescription information of Chinese patent medicines (CPMs) and ancient Chinese medicine prescriptions (CMPs). CPMCP reports their comprehensive and standardized information such as the components, indications and contraindications. It is worth mentioning that we organized relevant experts and spent lots of time manually mapping the functions of compound prescriptions in ancient Chinese to the standardized TCM symptom vocabularies, obtaining a total of 71 414 associations between compound prescriptions and TCM symptoms. In this way, CPMCP established the associations between TCM and modern medicine (MM) according to the associations between TCM symptoms and MM symptoms. In addition, to further exhibit the compatibility mechanism of compound prescriptions, CPMCP summarizes a set of common drug combination principles by analyzing the existing prescriptions. We believe that CPMCP can promote the modernization of TCM and make greater contributions to MM. Database URLhttp://cpmcp.top

Exploring drug-protein interactions (DPIs) through computational methods can effectively reduce the workload and the cost of DPI identification. Previous works try to predict DPIs by integrating and analyzing the unique features of drugs and proteins. They cannot adequately analyze the consistency between the drug features and the protein features due to their different semantics. However, the consistency of their features, such as the correlation originating from their sharing diseases, may reveal some potential DPIs. Here we propose a deep neural network-based co-coding method (DNNCC for short) to predict novel DPIs. DNNCC projects the original features of drugs and proteins to a common embedding space through a co-coding strategy. In this way, the embedding features of drugs and proteins have the same semantics. Therefore, the prediction module can discover the unknown DPIs by exploring the feature consistency between drugs and proteins. The experimental results indicate that the performance of DNNCC is significantly superior to five state-of-the-art DPI prediction methods under several evaluation metrics. The superiority of integrating and analyzing the common features of drugs and proteins is proved by the ablation experiments. The novel DPIs predicted by DNNCC verify that DNNCC is a powerful prior tool that can effectively discover potential DPIs.