Abstract: Lung cancer is the leading cause of cancer deaths globally, and lung adenocarcinoma (LUAD) is the most common type of lung cancer. Gene dysregulation plays an essential role in the development of LUAD. Drug repositioning based on associations between drug target genes and LUAD target genes are useful to discover potential new drugs for the treatment of LUAD, while also reducing the monetary and time costs of new drug discovery and development. Here, we developed a pipeline based on machine learning to predict potential LUAD-related target genes through established graph attention networks (GATs). We then predicted potential drugs for the treatment of LUAD through gene coincidence-based and gene network distance-based methods. Using data from 535 LUAD tissue samples and 59 precancerous tissue samples from The Cancer Genome Atlas, 48,597 genes were identified and used for the prediction model building of the GAT. The GAT model achieved good predictive performance, with an area under the receiver operating characteristic curve of 0.90. 1,597 potential LUAD-related genes were identified from the GAT model. These LUAD-related genes were then used for drug repositioning. The gene overlap and network distance with the target genes were calculated for 3,070 drugs and 672 preclinical compounds approved by the US Food and Drug Administration. At which, bromoethylamine was predicted as a novel potential preclinical compound for the treatment of LUAD, and cimetidine and benzbromarone were predicted as potential therapeutic drugs for LUAD. The pipeline established in this study presents new approach for developing targeted therapies for LUAD.

Abstract: Background: Leonurus japonicus Houtt has an obvious efficacy on cardiovascular diseases. As the most representative component in the herb, leonurine has attracted increasing attention for its potential in myocardial ischemia. However, its protective mechanism against myocardial ischemia remains incompletely elucidated. Objectives: The present study aimed to reveal the potential mechanism of leonurine in acute myocardial ischemia using a strategy combining metabolomics and network pharmacology. Methods: First, a metabolomics method was proposed to identify the differential metabolites of plasma in rats. Then, network pharmacology was performed to screen candidate targets of leonurine against acute myocardial ischemia. A compound-reaction-enzyme-gene network was thus constructed with the differential metabolites and targets. Finally, molecular docking was carried out to predict the binding capability of leonurine with key targets. Results: A total of 32 differential metabolites were identified in rat plasma, and 16 hub genes were detected through network pharmacology. According to the results of compound-reaction-enzyme-gene network and molecular docking, what was screened included six key targets (GSR, CYP2C9, BCHE, GSTP1, TGM2, and PLA2G2A) and seven differential metabolites (glycerylphosphorylcholine, lysophosphatidylcholine, choline phosphate, linoleic acid, 13-HpODE, tryptophan and glutamate) with four important metabolic pathways involved: glycerophospholopid metabolism, linoleic acid metabolism, tryptophan metabolism and glutamate metabolism. Among them, glycerophospholipid and tryptophan metabolism were shown to be important, since the regulation of leonurine on these two pathways was also observed in our previous metabolomics study conducted on clinical hyperlipidemia patients. Conclusion: This is the first study of its kind to reveal the underlying mechanism of leonurine against acute myocardial ischemia through a strategy combining metabolomics and network pharmacology, which provides a valuable reference for the research on its future application.

Abstract: Artificial intelligence represents a powerful computational tool to analyze large amounts of data with and without supervision. Such powerful techniques have significantly impacted the advancement of multiple sectors from social media to data security, the automotive industry, gaming, finances, and healthcare. Only recently, however, artificial intelligence has been making significant strides in healthcare and biomedical research. Despite such advancements and the expectation of potential breakthroughs arising from implementing artificial intelligence in these fields, there is limited knowledge transfer and training for most healthcare professionals on using these computational techniques. While there is a wide array of publications on artificial intelligence applied to scientific research, most are too technical (not aimed at the non-initiated), too broad (with an overwhelming amount of data), or too focused on a particular application. This chapter presents an overview of artificial intelligence and its derivatives, giving a historical perspective, a succinct technical explanation of the underlying basis, and some examples of its applications. It finishes with a brief discussion on the challenges for implementing AI to be fully accepted in the scientific community.

Abstract: Atherosclerosis is a chronic inflammatory disease characterized by the formation of plaque and endothelial dysfunction. Under pro-inflammatory conditions, endothelial cells adopt a mesenchymal phenotype by a process called endothelial-to-mesenchymal transition (EndMT) which plays an important role in the pathogenesis of atherosclerosis. Dan-Shen-Yin (DSY) is a well-known traditional Chinese medicine used in the treatment of cardiovascular disease. However, the molecular mechanism whereby DSY mitigates atherosclerosis remains unknown. Therefore, we employed a network pharmacology-based strategy in this study to determine the therapeutic targets of DSY, and in vitro experiments to understand the molecular pharmacology mechanism. The targets of the active ingredients of DSY related to EndMT and atherosclerosis were obtained and used to construct a protein-protein interaction (PPI) network followed by network topology and functional enrichment analysis. Network pharmacology analysis revealed that the PI3K/AKT pathway was the principal signaling pathway of DSY against EndMT in atherosclerosis. Molecular docking simulations indicated strong binding capabilities of DSY's bioactive ingredients toward PI3K/AKT pathway molecules. Experimentally, DSY could efficiently modify expression of signature EndMT genes and decrease expression of PI3K/AKT pathway signals including integrin alphaV, integrin beta1, PI3K, and AKT1 in TGF-beta2-treated HUVECs. LASP1, which is upstream of the PI3K/AKT pathway, had strong binding affinity to the majority of DSY's bioactive ingredients, was induced by EndMT-promoting stimuli involving IL-1beta, TGF-beta2, and hypoxia, and was downregulated by DSY. Knock-down of LASP1 attenuated the expression of integrin alphaV, integrin beta1, PI3K, AKT1 and EndMT-related genes induced by TGF-beta2, and minimized the effect of DSY. Thus, our study showed that DSY potentially exerted anti-EndMT activity through the LASP1/PI3K/AKT pathway, providing a possible new therapeutic intervention for atherosclerosis.

Abstract: Background: Shenlingbaizhu powder (SLBZP), one of the classic Earth-cultivating and gold-generating prescriptions of traditional Chinese medicine, is widely used to treat various diseases. However, the pharmacological mechanisms of SLBZP on bronchial asthma (BA) and allergic colitis (AC) remain to be elucidated. Methods: Network pharmacology and molecular docking technology were used to explore the potential mechanism of SLBZP in treating BA and AC with the simultaneous treatment of different diseases. The potential active compounds of SLBZP and their corresponding targets were obtained from BATMAN-TCM, ETCM, SymMap TCM@TAIWAN, and TCMSP databases. BA and AC disease targets were collected through DisGeNET, TTD, GeneCards, PharmGKB, OMIM, NCBI, The Human Phenotype Ontology, and DrugBank databases. Common targets for drugs and diseases were screened by using the bioinformatics and evolutionary genomics platform. The analyses and visualizations of Gene Ontology (GO) function and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment of common targets were carried out by R software. The key targets were screened by using the plug-in "cytoHubba" of Cytoscape software, and the "active compound-key target" network was constructed. Molecular docking analysis was performed using AutoDock software. The miRTarBase database was used to predict microRNAs (miRNAs) targeting key targets, and the key target-miRNA network was constructed. Result: Through screening, 246 active compounds and 281 corresponding targets were obtained. Common targets were mainly enriched in 2933 biological processes and 182 signal pathways to play the role of treating BA and AC. There were 131 active compounds related to key targets. The results of molecular docking showed that the important active compounds in SLBZP had good binding ability with the key targets. The key target-miRNA network showed that 94 miRNAs were predicted. Conclusion: SLBZP has played the role of treating different diseases with the same treatment on BA and AC through the characteristics of multicompound, multitarget, and multipathway of traditional Chinese medicine, which provides a theoretical basis for explaining the mechanism and clinical application of SLBZP treating different diseases with the same treatment in BA and AC.

Abstract: The therapeutic challenges pertaining to leishmaniasis due to reported chemoresistance and toxicity necessitate the need to explore novel pathways to identify plausible inhibitory molecules. Leishmania donovani 24-sterol methyltransferase (LdSMT) is vital for the synthesis of ergosterols, the main constituents of Leishmania cellular membranes. So far, mammals have not been shown to possess SMT or ergosterols, making the pathway a prime candidate for drug discovery. The structural model of LdSMT was elucidated using homology modeling to identify potential novel 24-SMT inhibitors via virtual screening, scaffold hopping, and de-novo fragment-based design. Altogether, six potential novel inhibitors were identified with binding energies ranging from -7.0 to -8.4 kcal/mol with e-LEA3D using 22,26-azasterol and S1-S4 obtained from scaffold hopping via the ChEMBL, DrugBank, PubChem, ChemSpider, and ZINC15 databases. These ligands showed comparable binding energy to 22,26-azasterol (-7.6 kcal/mol), the main inhibitor of LdSMT. Moreover, all the compounds had plausible ligand efficiency-dependent lipophilicity (LELP) scores above 3. The binding mechanism identified Tyr92 to be critical for binding, and this was corroborated via molecular dynamics simulations and molecular mechanics Poisson-Boltzmann surface area (MM-PBSA) calculations. The ligand A1 was predicted to possess antileishmanial properties with a probability of activity (Pa) of 0.362 and a probability of inactivity (Pi) of 0.066, while A5 and A6 possessed dermatological properties with Pa values of 0.205 and 0.249 and Pi values of 0.162 and 0.120, respectively. Structural similarity search via DrugBank identified vabicaserin, daledalin, zanapezil, imipramine, and cefradine with antileishmanial properties suggesting that the de-novo compounds could be explored as potential antileishmanial agents.

Abstract: The application of network pharmacology has greatly promoted the scientific interpretation of disease treatment mechanism of traditional Chinese medicine (TCM). However, the data required by network pharmacology analysis were scattered in different resources. In the present work, by integrating and reorganizing the data from multiple resources, we developed the intelligent network pharmacology platform unique for traditional Chinese medicine, called INPUT (http://cbcb.cdutcm.edu.cn/INPUT/), for automatically performing network pharmacology analysis. Besides the curated data collected from multiple resources, a series of bioinformatics tools for network pharmacology analysis were also embedded in INPUT, which makes it become the first automatic platform able to explore the disease treatment mechanisms of TCM. With the built-in tools, researchers can also analyze their own in-house data and obtain the results of pivotal ingredients, GO and KEGG pathway, protein-protein interactions, etc. In addition, as a proof-of-principle, INPUT was applied to decipher the antidepressant mechanism of a commonly used prescription. In summary, INPUT is a powerful platform for network pharmacology analysis and will facilitate the researches on drug discovery.

Abstract: Antibiotic resistance is a major public health concern. Antibiotic combinations, offering better efficacy at lower doses, are a useful way to handle this problem. However, it is difficult for us to find effective antibiotic combinations in the vast chemical space. Herein, we propose a graph learning framework to predict synergistic antibiotic combinations. In this model, a network proximity method combined with network propagation was used to quantify the relationships of drug pairs, and we found that synergistic antibiotic combinations tend to have smaller network proximity. Therefore, network proximity can be used for building an affinity matrix. Subsequently, the affinity matrix was fed into a graph regularization model to predict potential synergistic antibiotic combinations. Compared with existing methods, our model shows a better performance in the prediction of synergistic antibiotic combinations and interpretability.

Abstract: Biomedical knowledge is represented in structured databases and published in biomedical literature, and different computational approaches have been developed to exploit each type of information in predictive models. However, the information in structured databases and literature is often complementary. We developed a machine learning method that combines information from literature and databases to predict drug targets and indications. To effectively utilize information in published literature, we integrate knowledge graphs and published literature using named entity recognition and normalization before applying a machine learning model that utilizes the combination of graph and literature. We then use supervised machine learning to show the effects of combining features from biomedical knowledge and published literature on the prediction of drug targets and drug indications. We demonstrate that our approach using datasets for drug-target interactions and drug indications is scalable to large graphs and can be used to improve the ranking of targets and indications by exploiting features from either structure or unstructured information alone.

Abstract: Traditional Chinese medicine (TCM) plays an important role in the treatment of complex diseases, especially cardiovascular diseases. However, it is hard to identify their modes of action on account of their multiple components. The present study aims to evaluate the effects of Dan-Shen-Yin (DSY) granules on hypoxia-induced pulmonary hypertension (HPH), and then to decipher the molecular mechanisms of DSY. Systematic pharmacology was employed to identify the targets of DSY on HPH. Furthermore, core genes were identified by constructing a protein-protein interaction (PPI) network and analyzed by Gene Ontology (GO) and Kyoto Encyclopedia of Genes (KEGG) analysis. Related genes and pathways were verified using a hypoxia-induced mouse model and hypoxia-treated pulmonary artery cells. Based on network pharmacology, 147 potential targets of DSY on HPH were found, constructing a PPI network, and 13 hub genes were predicted. The results showed that the effect of DSY may be closely associated with AKT serine/threonine kinase 1 (AKT1), signal transducer and activator of transcription 3 (STAT3), and HIF-1 signaling pathways, as well as biological processes such as cell proliferation. Consistent with network pharmacology analysis, experiments in vivo demonstrated that DSY could prevent the development of HPH in a hypoxia-induced mouse model and alleviate pulmonary vascular remodeling. In addition, inhibition of STAT3/HIF-1alpha/VEGF and FAK/AKT signaling pathways might serve as mechanisms. Taken together, the network pharmacology analysis suggested that DSY exhibited therapeutic effects through multiple targets in the treatment of HPH. The inferences were initially confirmed by subsequent in vivo and in vitro studies. This study provides a novel perspective for studying the relevance of TCM and disease processes and illustrates the advantage of this approach and the multitargeted anti-HPH effect of DSY.