Motivation An essential part of drug discovery is the accurate prediction of the binding affinity of new compound-protein pairs. Most of the standard computational methods assume that compounds or proteins of the test data are observed during the training phase. However, in real-world situations, the test and training data are sampled from different domains with different distributions. To cope with this challenge, we propose a deep learning-based approach that consists of three steps. In the first step, the training encoder network learns a novel representation of compounds and proteins. To this end, we combine convolutional layers and LSTM layers so that the occurrence patterns of local substructures through a protein and a compound sequence are learned. Also, to encode the interaction strength of the protein and compound substructures, we propose a two-sided attention mechanism. In the second phase, to deal with the different distributions of the training and test domains, a feature encoder network is learned for the test domain by utilizing an adversarial domain adaptation approach. In the third phase, the learned test encoder network is applied to new compound-protein pairs to predict their binding affinity. Results To evaluate the proposed approach, we applied it to KIBA, Davis, and BindingDB datasets. The results show that the proposed method learns a more reliable model for the test domain in more challenging situations. Availability https://github.com/LBBSoft/DeepCDA

No abstract

No abstract

Multi-target drugs against particular multiple targets get better protection, resistance profiles and curative influence by cooperative rules of a key beneficial target with resistance behavior and compensatory elements. Computational techniques can assist us in the efforts to design novel drugs (ligands) with a preferred bioactivity outline and alternative bioactive molecules at an early stage. A number of in silico methods have been explored extensively in order to facilitate the investigation of individual target agents and to propose a selective drug. A different, progressively more significant field which is used to predict the bioactivity of chemical compounds is the data mining method. Some of the previously mentioned methods have been investigated for multi-target drug design (MTDD) to find drug leads interact simultaneously with multiple targets. Several cheminformatics methods and structure-based approaches try to extract information from units working cooperatively in a biomolecular system to fulfill their task. To dominate the difficulties of the experimental specification of ligand-target structures, rational methods, namely molecular docking, SAR and QSAR are vital substitutes to obtain knowledge for each structure in atomic insight. These procedures are logically successful for the prediction of binding affinity and have shown promising potential in facilitating MTDD. Here, we review some of the important features of the multi-target therapeutics discoveries using the computational approach, highlighting the SAR, QSAR, docking and pharmacophore methods to discover interactions between drug-target that could be leveraged for curative benefits. A summary of each, followed by examples of its applications in drug design has been provided. Computational efficiency of each method has been represented according to its main strengths and limitations.

Solar-driven hydrogen production can alleviate ever-growing energy crisis, but developing affordable and efficient photocatalysts is challenging. We here report a noble-metal-free WO3@ZnIn2S4 S-scheme heterojunction photocatalyst with ZnIn2S4 nanosheets vertically growing...