Predicting Drug-Target Affinity Based on Recurrent Neural Networks and Graph Convolutional Neural Networks.

BACKGROUND: Drug development requires a lot of money and time, and the outcome of the challenge is unknown. So, there is an urgent need for researchers to find a new approach that can reduce costs. Therefore, the identification of drug-target interactions (DTIs) has been a critical step in the early stages of drug discovery. These computational methods aim to narrow the search space for novel DTIs and to elucidate the functional background of drugs. Most of the methods developed so far use binary classification to predict the presence or absence of interactions between the drug and the target. However, it is more informative but also more challenging to predict the strength of the binding between a drug and its target. If the strength is not strong enough, such a DTI may not be useful. Hence, the development of methods to predict drug-target affinity (DTA) is of significant importance Method: We have improved the GraphDTA model from a dual-channel model to a triple-channel model. We interpreted the target/protein sequences as time series and extracted their features using the LSTM network. For the drug, we considered both the molecular structure and the local chemical background, retaining the four variant networks used in GraphDTA to extract the topological features of the drug and capturing the local chemical background of the atoms in the drug by using BiGRU. Thus, we obtained the latent features of the target and two latent features of the drug. The connection of these three feature vectors is then inputted into a 2 layer FC network, and a valuable binding affinity is the output. RESULT: We used the Davis and Kiba datasets, using 80% of the data for training and 20% of the data for validation. Our model showed better performance when compared with the experimental results of GraphDTA Conclusion: In this paper, we altered the GraphDTA model to predict drug-target affinity. It represents the drug as a graph and extracts the two-dimensional drug information using a graph convolutional neural network. Simultaneously, the drug and protein targets are represented as a word vector, and the convolutional neural network is used to extract the time-series information of the drug and the target. We demonstrate that our improved method has better performance than the original method. In particular, our model has better performance in the evaluation of benchmark databases.

Efficacy and safety of tocilizumab in COVID-19 patients.

In this retrospective study we assessed the efficacy and safety of tocilizumab in patients with critical or severe coronavirus disease 2019 (COVID-19). We enrolled 181 patients admitted to Huoshenshan Hospital (Wuhan, China) with confirmed COVID-19 between January 2020 and February 2020. Ninety-two patients were treated with tocilizumab, and 89 patients were treated conventionally. We analyzed the clinical manifestations, changes in CT scan images, and laboratory tests before and after tocilizumab treatment, and compared these results with the conventionally treated group. A significant reduction in the level of C-reactive protein was observed 1 week after tocilizumab administration. In some cases this meant the end of the IL-6-related cytokine storm. In addition, tocilizumab relieved fever, cough, and shortness of breath with no reported adverse drug reactions. These findings suggest tocilizumab improves clinical outcomes and is effective for treatment of patients with critical or severe COVID-19. However, future clinical trials are needed to better understand the impact of tocilizumab interference with IL-6 and provide a therapeutic strategy for treatment of COVID-19.

Identifying potential treatments of COVID-19 from Traditional Chinese Medicine (TCM) by using a data-driven approach.

ETHNOPHARMACOLOGICAL RELEVANCE: Traditional Chinese Medicine (TCM) has been widely used as an approach worldwide. Chinese Medicines (CMs) had been used to treat and prevent viral infection pneumonia diseases for thousands of years and had accumulated a large number of clinical experiences and effective prescriptions. AIM OF THE STUDY: This research aimed to systematically excavate the classical prescriptions of Chinese Medicine (CM), which have been used to prevent and treat Pestilence (Wenbing, Wenyi, Shiyi or Yibing) for long history in China, to obtain the potential prescriptions and ingredients to alternatively treat COVID-19. MATERIALS AND METHODS: We developed the screening system based on data mining, molecular docking and network pharmacology. Data mining and association network were used to mine the high-frequency herbs and formulas from ancient prescriptions. Virtual screening for the effective components of high frequency CMs and compatibility Chinese Medicine was explored by a molecular docking approach. Furthermore, network pharmacology method was used to preliminarily uncover the molecule mechanism. RESULTS: 574 prescriptions were obtained from 96,606 classical prescriptions with the key words to treat "Warm diseases (Wenbing)", "Pestilence (Wenyi or Yibing)" or "Epidemic diseases (Shiyi)". Meanwhile, 40 kinds of CMs, 36 CMs-pairs, 6 triple-CMs-groups existed with high frequency among the 574 prescriptions. Additionally, the key targets of SARS-COV-2, namely 3CL hydrolase (Mpro) and angiotensin-converting enzyme 2(ACE2), were used to dock the main ingredients from the 40 kinds by the LigandFitDock method. A total of 66 compounds components with higher frequency were docked with the COVID-19 targets, which were distributed in 26 kinds of CMs, among which Gancao (Glycyrrhizae Radix Et Rhizoma), HuangQin (Scutellariae Radix), Dahuang (Rhei Radix Et Rhizome) and Chaihu (Bupleuri Radix) contain more potential compounds. Network pharmacology results showed that Gancao (Glycyrrhizae Radix Et Rhizoma) and HuangQin (Scutellariae Radix) CMs-pairs could also interact with the targets involving in immune and inflammation diseases. CONCLUSIONS: These results we obtained probably provided potential candidate CMs formulas or active ingredients to overcome COVID-19. Prospectively, animal experiment and rigorous clinic studies are needed to confirm the potential preventive and treat effect of these CMs and compounds.