Network-Based Methods for Prediction of Drug-Target Interactions

Promiscuity in drug discovery on the verge of the structural revolution: recent advances and future chances

INTRODUCTION

Promiscuity denotes the ability of ligands and targets to specifically interact with multiple binding partners. Despite negative aspects like side effects, promiscuity is receiving increasing attention in drug discovery as it can enhance drug efficacy and provides a molecular basis for drug repositioning. The three-dimensional structure of ligand-target complexes delivers exclusive insights into the molecular mechanisms of promiscuity and structure-based methods enable the identification of promiscuous interactions. With the recent breakthrough in protein structure prediction, novel possibilities open up to reveal unknown connections in ligand-target interaction networks.

AREAS COVERED

This review highlights the significance of structure in the identification and characterization of promiscuity and evaluates the potential of protein structure prediction to advance our knowledge of drug-target interaction networks. It discusses the definition and relevance of promiscuity in drug discovery and explores different approaches to detecting promiscuous ligands and targets.

EXPERT OPINION

Examination of structural data is essential for understanding and quantifying promiscuity. The recent advancements in structure prediction have resulted in an abundance of targets that are well-suited for structure-based methods like docking. In silico approaches may eventually completely transform our understanding of drug-target networks by complementing the millions of predicted protein structures with billions of predicted drug-target interactions.

Expanding potential targets of herbal chemicals by node2vec based on herb-drug interactions

BACKGROUND

The identification of chemical-target interaction is key to pharmaceutical research and development, but the unclear materials basis and complex mechanisms of traditional medicine (TM) make it difficult, especially for low-content chemicals which are hard to test in experiments. In this research, we aim to apply the node2vec algorithm in the context of drug-herb interactions for expanding potential targets and taking advantage of molecular docking and experiments for verification.

METHODS

Regarding the widely reported risks between cardiovascular drugs and herbs, Salvia miltiorrhiza (Danshen, DS) and Ligusticum chuanxiong (Chuanxiong, CX), which are widely used in the treatment of cardiovascular disease (CVD), and approved drugs for CVD form the new dataset as an example. Three data groups DS-drug, CX-drug, and DS-CX-drug were applied to serve as the context of drug-herb interactions for link prediction. Three types of datasets were set under three groups, containing information from chemical-target connection (CTC), chemical-chemical connection (CCC) and protein-protein interaction (PPI) in increasing steps. Five algorithms, including node2vec, were applied as comparisons. Molecular docking and pharmacological experiments were used for verification.

RESULTS

Node2vec represented the best performance with average AUROC and AP values of 0.91 on the datasets "CTC, CCC, PPI". Targets of 32 herbal chemicals were identified within 43 predicted edges of herbal chemicals and drug targets. Among them, 11 potential chemical-drug target interactions showed better binding affinity by molecular docking. Further pharmacological experiments indicated caffeic acid increased the thermal stability of the protein GGT1 and ligustilide and low-content chemical neocryptotanshinone induced mRNA change of FGF2 and MTNR1A, respectively.

CONCLUSIONS

The analytical framework and methods established in the study provide an important reference for researchers in discovering herb-drug interactions, alerting clinical risks, and understanding complex mechanisms of TM.

Investigation of Anti-Alzheimer's Mechanisms of Sarsasapogenin Derivatives by Network-Based Combining Structure-Based Methods

Alzheimer's disease (AD), a neurodegenerative disease with no cure, affects millions of people worldwide and has become one of the biggest healthcare challenges. Some investigated compounds play anti-AD roles at the cellular or the animal level, but their molecular mechanisms remain unclear. In this study, we designed a strategy combining network-based and structure-based methods together to identify targets for anti-AD sarsasapogenin derivatives (AAs). First, we collected drug-target interactions (DTIs) data from public databases, constructed a global DTI network, and generated drug-substructure associations. After network construction, network-based models were built for DTI prediction. The best bSDTNBI-FCFP_4 model was further used to predict DTIs for AAs. Second, a structure-based molecular docking method was employed for rescreening the prediction results to obtain more credible target proteins. Finally, in vitro experiments were conducted for validation of the predicted targets, and Nrf2 showed significant evidence as the target of anti-AD compound AA13. Moreover, we analyzed the potential mechanisms of AA13 for the treatment of AD. Generally, our combined strategy could be applied to other novel drugs or compounds and become a useful tool in identification of new targets and elucidation of disease mechanisms. Our model was deployed on our NetInfer web server (http://lmmd.ecust.edu.cn/netinfer/).

Pharmacological Mechanism of Aucklandiae Radix against Gastric Ulcer Based on Network Pharmacology and In Vivo Experiment

Background and Objectives: Aucklandiae Radix is a well-known medicinal herb that is often used to treat gastric ulcer, but its molecular mechanism of anti-ulcer action is poorly understood. This research aimed to reveal the potential active components, core targets, and mechanisms of Aucklandiae Radix in treating gastric ulcer by combining network pharmacology and animal experimentation. Materials and Methods: First, a network pharmacology strategy was used to predict the main components, candidate targets, and potential signaling pathways. Molecular docking was then used to confirm the binding affinity between the main components and primary targets. Finally, rats were treated with indomethacin 30 mg/kg to establish a gastric ulcer model. Aucklandiae Radix extract (0.15, 0.3, and 0.6 g/kg) was pre-treated in rats by oral gavage for 14 days, and the protective effect and candidate targets of network pharmacology were validated through morphological observation, pathological staining, and biochemical index detection. Results: A total of eight potential active components and 331 predicted targets were screened from Aucklandiae Radix, 37 of which were common targets with gastric ulcer. According to the component–target network and protein-protein interaction (PPI) network, stigmasterol, mairin, sitosterol, and dehydrocostus lactone were identified as the key components, and RAC-alpha serine/threonine-protein kinase (AKT1), prostaglandin-endoperoxide synthase 2 (PTGS2), interleukin 1 beta (IL1B), caspase-3 (CASP3), and CASP8 were selected as the core targets. Gene ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) enrichment results revealed the pharmacological mechanism of Aucklandiae Radix against gastric ulcer related to many biological processes and pathways, including antibacterial, anti-inflammatory, prostaglandin receptor response, and apoptosis. Molecular docking verification showed that the key components and core targets had good binding affinities. In the in vivo experiments, Aucklandiae Radix notably relieved the gastric ulcer by reducing the levels of tumor necrosis factor (TNF)-α, interleukin (IL)-1β, and myeloperoxidase (MPO) while improving the gastric histopathological features. Conclusion: The overall findings suggest that Aucklandiae Radix treats gastric ulcer with a multi-component, multi-target, and multi-mechanism model.

DTiGNN: Learning drug-target embedding from a heterogeneous biological network based on a two-level attention-based graph neural network

MOTIVATION

In vitro experiment-based drug-target interaction (DTI) exploration demands more human, financial and data resources. In silico approaches have been recommended for predicting DTIs to reduce time and cost. During the drug development process, one can analyze the therapeutic effect of the drug for a particular disease by identifying how the drug binds to the target for treating that disease. Hence, DTI plays a major role in drug discovery. Many computational methods have been developed for DTI prediction. However, the existing methods have limitations in terms of capturing the interactions via multiple semantics between drug and target nodes in a heterogeneous biological network (HBN).

METHODS

In this paper, we propose a DTiGNN framework for identifying unknown drug-target pairs. The DTiGNN first calculates the similarity between the drug and target from multiple perspectives. Then, the features of drugs and targets from each perspective are learned separately by using a novel method termed an information entropy-based random walk. Next, all of the learned features from different perspectives are integrated into a single drug and target similarity network by using a multi-view convolutional neural network. Using the integrated similarity networks, drug interactions, drug-disease associations, protein interactions and protein-disease association, the HBN is constructed. Next, a novel embedding algorithm called a meta-graph guided graph neural network is used to learn the embedding of drugs and targets. Then, a convolutional neural network is employed to infer new DTIs after balancing the sample using oversampling techniques.

RESULTS

The DTiGNN is applied to various datasets, and the result shows better performance in terms of the area under receiver operating characteristic curve (AUC) and area under precision-recall curve (AUPR), with scores of 0.98 and 0.99, respectively. There are 23,739 newly predicted DTI pairs in total.

Target identification by structure-based computational approaches: Recent advances and perspectives

The use of computational techniques in the early stages of drug discovery has recently experienced a boost, especially in the target identification step. Finding the biological partner(s) for new or existing synthetic and/or natural compounds by "wet" approaches may be challenging; therefore, preliminary in silico screening is even more recommended. After a brief overview of some of the most known target identification techniques, recent advances in structure-based computational approaches for target identification are reported in this digest, focusing on Inverse Virtual Screening and its recent applications. Moreover, future perspectives concerning the use of such methodologies, coupled or not with other approaches, are analyzed.

Computational Approaches to Enzyme Inhibition by Marine Natural Products in the Search for New Drugs

The exploration of biologically relevant chemical space for the discovery of small bioactive molecules present in marine organisms has led not only to important advances in certain therapeutic areas, but also to a better understanding of many life processes. The still largely untapped reservoir of countless metabolites that play biological roles in marine invertebrates and microorganisms opens new avenues and poses new challenges for research. Computational technologies provide the means to (i) organize chemical and biological information in easily searchable and hyperlinked databases and knowledgebases; (ii) carry out cheminformatic analyses on natural products; (iii) mine microbial genomes for known and cryptic biosynthetic pathways; (iv) explore global networks that connect active compounds to their targets (often including enzymes); (v) solve structures of ligands, targets, and their respective complexes using X-ray crystallography and NMR techniques, thus enabling virtual screening and structure-based drug design; and (vi) build molecular models to simulate ligand binding and understand mechanisms of action in atomic detail. Marine natural products are viewed today not only as potential drugs, but also as an invaluable source of chemical inspiration for the development of novel chemotypes to be used in chemical biology and medicinal chemistry research.

Using chemical and biological data to predict drug toxicity

Various sources of information can be used to better understand and predict compound activity and safety-related endpoints, including biological data such as gene expression and cell morphology. In this review, we first introduce types of chemical, in vitro and in vivo information that can be used to describe compounds and adverse effects. We then explore how compound descriptors based on chemical structure or biological perturbation response can be used to predict safety-related endpoints, and how especially biological data can help us to better understand adverse effects mechanistically. Overall, the described applications demonstrate how large-scale biological information presents new opportunities to anticipate and understand the biological effects of compounds, and how this can support predictive toxicology and drug discovery projects.

NMTF-DTI: A Nonnegative Matrix Tri-factorization Approach With Multiple Kernel Fusion for Drug-Target Interaction Prediction

Prediction of drug-target interactions (DTIs) plays a significant role in drug development and drug discovery. Although this task requires a large investment in terms of time and cost, especially when it is performed experimentally, the results are not necessarily significant. Computational DTI prediction is a shortcut to reduce the risks of experimental methods. In this study, we propose an effective approach of nonnegative matrix tri-factorization, referred to as NMTF-DTI, to predict the interaction scores between drugs and targets. NMTF-DTI utilizes multiple kernels (similarity measures) for drugs and targets and Laplacian regularization to boost the prediction performance. The performance of NMTF-DTI is evaluated via cross-validation and is compared with existing DTI prediction methods in terms of the area under the receiver operating characteristic (ROC) curve (AUC) and the area under the precision and recall curve (AUPR). We evaluate our method on four gold standard datasets, comparing to other state-of-the-art methods. Cross-validation and a separate, manually created dataset are used to set parameters. The results show that NMTF-DTI outperforms other competing methods. Moreover, the results of a case study also confirm the superiority of NMTF-DTI.

Systems Pharmacology and Pharmacokinetics Strategy to Decode Bioactive Ingredients and Molecular Mechanisms from Zingiber officinale as Phyto-therapeutics against Neurological Diseases

BACKGROUND

The bioactive constituents from Zingiber officinale (Z. officinale) have shown a positive effect on neurodegenerative diseases like Alzheimer's disease (AD), which manifests as progressive memory loss and cognitive impairment.

OBJECTIVE

This study investigates the binding ability and the pharmaco-therapeutic potential of Z. officinale with AD disease targets by molecular docking and molecular dynamic (MD) simulation approaches.

METHODS

By coupling enormous available phytochemical data and advanced computational technologies, the possible molecular mechanism of action of these bioactive compounds was deciphered by evaluating phytochemicals, target fishing, and network biological analysis.

RESULTS

As a result, 175 bioactive compounds and 264 human target proteins were identified. The gene ontology and Kyoto Encyclopaedia of Genes and Genomes pathway enrichment analysis and molecular docking were used to predict the basis of vital bioactive compounds and biomolecular mechanisms involved in the treatment of AD. Amongst selected bioactive compounds, 10- Gingerdione and 1-dehydro-[8]-gingerdione exhibited significant anti-neurological properties against AD targeting amyloid precursor protein with docking energy of -6.0 and -5.6, respectively.

CONCLUSION

This study suggests that 10-Gingerdione and 1-dehydro-[8]-gingerdione strongly modulates the anti-neurological activity and are associated with pathological features like amyloid-β plaques and hyperphosphorylated tau protein are found to be critically regulated by these two target proteins. This comprehensive analysis provides a clue for further investigation of these natural compounds' inhibitory activity in drug discovery for AD treatment.

The curse and blessing of abundance-the evolution of drug interaction databases and their impact on drug network analysis

BACKGROUND

Widespread bioinformatics applications such as drug repositioning or drug-drug interaction prediction rely on the recent advances in machine learning, complex network science, and comprehensive drug datasets comprising the latest research results in molecular biology, biochemistry, or pharmacology. The problem is that there is much uncertainty in these drug datasets-we know the drug-drug or drug-target interactions reported in the research papers, but we cannot know if the not reported interactions are absent or yet to be discovered. This uncertainty hampers the accuracy of such bioinformatics applications.

RESULTS

We use complex network statistics tools and simulations of randomly inserted previously unaccounted interactions in drug-drug and drug-target interaction networks-built with data from DrugBank versions released over the plast decade-to investigate whether the abundance of new research data (included in the latest dataset versions) mitigates the uncertainty issue. Our results show that the drug-drug interaction networks built with the latest dataset versions become very dense and, therefore, almost impossible to analyze with conventional complex network methods. On the other hand, for the latest drug database versions, drug-target networks still include much uncertainty; however, the robustness of complex network analysis methods slightly improves.

CONCLUSIONS

Our big data analysis results pinpoint future research directions to improve the quality and practicality of drug databases for bioinformatics applications: benchmarking for drug-target interaction prediction and drug-drug interaction severity standardization.

Acteoside and ursolic acid synergistically protects H2O2-induced neurotrosis by regulation of AKT/mTOR signalling: from network pharmacology to experimental validation

CONTEXT

Ursolic acid (UA) and acteoside (ATS) are important active components that have been used to treat Alzheimer's disease (AD) because of their neuroprotective effects, but the exact mechanism is still unclear.

OBJECTIVE

Network pharmacology was used to explore the mechanism of UA + ATS in treating AD, and cell experiments were used to verify the mechanism.

MATERIALS AND METHODS

UA + ATS targets and AD-related genes were retrieved from TCMSP, STITCH, SwissTargetPrediction, GeneCards, DisGeNET and GEO. Key targets were obtained by constructing protein interaction network through STRING. The neuroprotective effects of UA + ATS were verified in H₂O₂-treated PC12 cells. The subsequent experiments were divided into Normal, Model (H₂O₂ pre-treatment for 4 h), Control (H₂O₂+ solvent pre-treatment), UA (5 μM), ATS (40 μM), UA (5 μM) + ATS (40 μM). Then apoptosis, mitochondrial membrane potential, caspase-3 activity, ATG5, Beclin-1 protein expression and Akt, mTOR phosphorylation levels were detected.

RESULTS

The key targets of UA + ATS-AD network were mainly enriched in Akt/mTOR pathway. Cell experiments showed that UA (ED₅₀: 5 μM) + ATS (ED₅₀: 40 μM) could protect H₂O₂-induced (IC₅₀: 250 μM) nerve damage by enhancing cells viability, combating apoptosis, restoring MMP, reducing the activation of caspase-3, lessening the phosphorylation of Akt and mTOR, and increasing the expression of ATG5 and Beclin-1.

CONCLUSIONS

ATS and UA regulates multiple targets, bioprocesses and signal pathways against AD pathogenesis. ATS and UA synergistically protects H₂O₂-induced neurotrosis by regulation of AKT/mTOR signalling.

To explore the active constituents of Sedum aizoon L in the treatment of coronary heart disease based on network pharmacology and molecular docking methodology

Background

There is a lack of effective drugs for the treatment of coronary heart disease (CHD). Sedum aizoon L (SL) has multiple effects, and there is no report on CHD in SL at present. The aim of this study is to explore the mechanisms of action of SL in the treatment of CHD based on network pharmacology and molecular docking technology.

Methods

The targets and active ingredients of SL were screened using the Traditional Chinese Medicine Systems Pharmacology (TCMSP) database, and CHD-related targets were obtained by searching GeneCards and DisGeNet databases. The intersection of LS active ingredient targets and CHD targets was used to construct a "drug-ingredient-disease-target" network using the Cytoscape software. The STRING database was used to construct a protein-protein interaction (PPI) network, and Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed. Key targets and core active ingredients were selected and molecular docking was performed using the AutoDock software.

Results

According to the predicted results, a total of 134 corresponding target genes for LS, 12 active components, 1,704 CHD-related targets, and 52 intersecting targets were obtained. GO function and KEGG pathway analysis showed that the key targets were involved with signal transducer and activator of transcription 3 (STAT3), tumor protein p53 (TP53), and vascular endothelial growth factor A (VEGFA). The molecular docking results showed that the key targets bound to the important active ingredients in a stable conformation. The core active ingredients of LS in the treatment of CHD were determined to be ursolic acid, myricetin, and beta-sitosterol.

Conclusions

SL may act on targets such as STAT3, TP53, and VEGFA through tumor necrosis factor (TNF) signaling pathway, interleukin 17A (IL-17A) signaling pathway, AGE-RAGE signaling pathway in diabetic complications, and other related pathways, thereby playing a role in preventing and treating CHD.

Discovering drug-target interaction knowledge from biomedical literature

MOTIVATION

The interaction between drugs and targets (DTI) in human body plays a crucial role in biomedical science and applications. As millions of papers come out every year in the biomedical domain, automatically discovering DTI knowledge from biomedical literature, which are usually triplets about drugs, targets and their interaction, becomes an urgent demand in the industry. Existing methods of discovering biological knowledge are mainly extractive approaches that often require detailed annotations (e.g. all mentions of biological entities, relations between every two entity mentions, etc.). However, it is difficult and costly to obtain sufficient annotations due to the requirement of expert knowledge from biomedical domains.

RESULTS

To overcome these difficulties, we explore an end-to-end solution for this task by using generative approaches. We regard the DTI triplets as a sequence and use a Transformer-based model to directly generate them without using the detailed annotations of entities and relations. Further, we propose a semi-supervised method, which leverages the aforementioned end-to-end model to filter unlabeled literature and label them. Experimental results show that our method significantly outperforms extractive baselines on DTI discovery. We also create a dataset, KD-DTI, to advance this task and release it to the community.

AVAILABILITY AND IMPLEMENTATION

Our code and data are available at https://github.com/bert-nmt/BERT-DTI.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Computational approaches for network-based integrative multi-omics analysis

Advances in omics technologies allow for holistic studies into biological systems. These studies rely on integrative data analysis techniques to obtain a comprehensive view of the dynamics of cellular processes, and molecular mechanisms. Network-based integrative approaches have revolutionized multi-omics analysis by providing the framework to represent interactions between multiple different omics-layers in a graph, which may faithfully reflect the molecular wiring in a cell. Here we review network-based multi-omics/multi-modal integrative analytical approaches. We classify these approaches according to the type of omics data supported, the methods and/or algorithms implemented, their node and/or edge weighting components, and their ability to identify key nodes and subnetworks. We show how these approaches can be used to identify biomarkers, disease subtypes, crosstalk, causality, and molecular drivers of physiological and pathological mechanisms. We provide insight into the most appropriate methods and tools for research questions as showcased around the aetiology and treatment of COVID-19 that can be informed by multi-omics data integration. We conclude with an overview of challenges associated with multi-omics network-based analysis, such as reproducibility, heterogeneity, (biological) interpretability of the results, and we highlight some future directions for network-based integration.

A novel method for drug-target interaction prediction based on graph transformers model

BACKGROUND

Drug-target interactions (DTIs) prediction becomes more and more important for accelerating drug research and drug repositioning. Drug-target interaction network is a typical model for DTIs prediction. As many different types of relationships exist between drug and target, drug-target interaction network can be used for modeling drug-target interaction relationship. Recent works on drug-target interaction network are mostly concentrate on drug node or target node and neglecting the relationships between drug-target.

RESULTS

We propose a novel prediction method for modeling the relationship between drug and target independently. Firstly, we use different level relationships of drugs and targets to construct feature of drug-target interaction. Then, we use line graph to model drug-target interaction. After that, we introduce graph transformer network to predict drug-target interaction.

CONCLUSIONS

This method introduces a line graph to model the relationship between drug and target. After transforming drug-target interactions from links to nodes, a graph transformer network is used to accomplish the task of predicting drug-target interactions.

Methodology of network pharmacology for research on Chinese herbal medicine against COVID-19: A review

Traditional Chinese medicine, as a complementary and alternative medicine, has been practiced for thousands of years in China and possesses remarkable clinical efficacy. Thus, systematic analysis and examination of the mechanistic links between Chinese herbal medicine (CHM) and the complex human body can benefit contemporary understandings by carrying out qualitative and quantitative analysis. With increasing attention, the approach of network pharmacology has begun to unveil the mystery of CHM by constructing the heterogeneous network relationship of "herb-compound-target-pathway," which corresponds to the holistic mechanisms of CHM. By integrating computational techniques into network pharmacology, the efficiency and accuracy of active compound screening and target fishing have been improved at an unprecedented pace. This review dissects the core innovations to the network pharmacology approach that were developed in the years since 2015 and highlights how this tool has been applied to understanding the coronavirus disease 2019 and refining the clinical use of CHM to combat it.

GCHN-DTI: Predicting drug-target interactions by graph convolution on heterogeneous networks

Determining the interaction of drug and target plays a key role in the process of drug development and discovery. The calculation methods can predict new interactions and speed up the process of drug development. In recent studies, the network-based approaches have been proposed to predict drug-target interactions. However, these methods cannot fully utilize the node information from heterogeneous networks. Therefore, we propose a method based on heterogeneous graph convolutional neural network for drug-target interaction prediction, GCHN-DTI (Predicting drug-target interactions by graph convolution on heterogeneous net-works), to predict potential DTIs. GCHN-DTI integrates network information from drug-target interactions, drug-drug interactions, drug-similarities, target-target interactions, and target-similarities. Then, the graph convolution operation is used in the heterogeneous network to obtain the node embedding of the drugs and the targets. Furthermore, we incorporate an attention mechanism between graph convolutional layers to combine node embedding from each layer. Finally, the drug-target interaction score is predicted based on the node embedding of the drugs and the targets. Our model uses fewer network types and achieves higher prediction performance. In addition, the prediction performance of the model will be significantly improved on the dataset with a higher proportion of positive samples. The experimental evaluations show that GCHN-DTI outperforms several state-of-the-art prediction methods.

A pharmacometric approach to evaluate drugs for potential repurposing as COVID-19 therapeutics

INTRODUCTION

Developing and evaluating novel compounds for treatment or prophylaxis of emerging infectious diseases is costly and time-consuming. Repurposing of already available marketed compounds is an appealing option as they already have an established safety profile. This approach could substantially reduce cost and time required to make effective treatments available to fight the COVID-19 pandemic. However, this approach is challenging since many drug candidates show efficacy in in vitro experiments, but fail to deliver effect when evaluated in clinical trials. Better approaches to evaluate in vitro data are needed, in order to prioritize drugs for repurposing.

AREAS COVERED

This article evaluates potential drugs that might be of interest for repurposing in the treatment of patients with COVID-19 disease. A pharmacometric simulation-based approach was developed to evaluate in vitro activity data in combination with expected clinical drug exposure, in order to evaluate the likelihood of achieving effective concentrations in patients.

EXPERT OPINION

The presented pharmacometric approach bridges in vitro activity data to clinically expected drug exposures, and could therefore be a useful compliment to other methods in order to prioritize repurposed drugs for evaluation in prospective randomized controlled clinical trials.

Network Pharmacology and Molecular Docking Study of Yupingfeng Powder in the Treatment of Allergic Diseases

Objective

To explore the potential mechanisms of Yupingfeng Powder (YPFP) in the treatment of allergic diseases by using network pharmacology and molecular docking technology.

Methods

The active components and targets of YPFP were screened by the TCMSP database. The targets associated with atopic dermatitis, asthma, allergic rhinitis, and food allergy were obtained from GeneCards and OMIM databases, respectively. The intersection of the above disease-related targets was identified as allergy-related targets. Then, allergy-related targets and YPFP-related targets were crossed to obtain the potential targets of YPFP for allergy treatment. A protein-protein-interaction (PPI) network and a drug-target-disease topology network were constructed to screen hub targets and key ingredients. Next, GO and KEGG pathway enrichment analyses were performed separately on the potential targets and hub targets to identify the biological processes and signaling pathways involved. Finally, molecular docking was conducted to verify the binding affinity between key ingredients and hub targets.

Results

In this study, 45 active ingredients were identified from YPFP, and 48 allergy-related targets were predicted by network pharmacology. IL6, TNF, IL1B, PTGS2, CXCL8, JUN, CCL2, IL10, IFNG, and IL4 were screened as hub targets by the PPI network. However, quercetin, kaempferol, wogonin, formononetin, and 7-O-methylisomucronulatol were identified as key ingredients by the drug-target-disease topological network. GO and KEGG pathway enrichment analysis indicated that the therapeutic effect of YPFP on allergy involved multiple biological processes and signaling pathways, including positive regulation of fever generation, positive regulation of neuroinflammatory response, vascular endothelial growth factor production, negative regulation of cytokine production involved in immune response, positive regulation of mononuclear cell migration, type 2 immune response, and negative regulation of lipid storage. Molecular docking verified that all the key ingredients had good binding affinity with hub targets.

Conclusion

This study revealed the key ingredients, hub targets, and potential mechanisms of YPFP antiallergy, and these data can provide some theoretical basis for subsequent allergy treatment and drug development.

Drug-Target Network Study Reveals the Core Target-Protein Interactions of Various COVID-19 Treatments

The coronavirus disease 2019 (COVID-19) pandemic has caused a dramatic loss of human life and devastated the worldwide economy. Numerous efforts have been made to mitigate COVID-19 symptoms and reduce the death rate. We conducted literature mining of more than 250 thousand published works and curated the 174 most widely used COVID-19 medications. Overlaid with the human protein-protein interaction (PPI) network, we used Steiner tree analysis to extract a core subnetwork that grew from the pharmacological targets of ten credible drugs ascertained by the CTD database. The resultant core subnetwork consisted of 34 interconnected genes, which were associated with 36 drugs. Immune cell membrane receptors, the downstream cellular signaling cascade, and severe COVID-19 symptom risk were significantly enriched for the core subnetwork genes. The lung mast cell was most enriched for the target genes among 1355 human tissue-cell types. Human bronchoalveolar lavage fluid COVID-19 single-cell RNA-Seq data highlighted the fact that T cells and macrophages have the most overlapping genes from the core subnetwork. Overall, we constructed an actionable human target-protein module that mainly involved anti-inflammatory/antiviral entry functions and highly overlapped with COVID-19-severity-related genes. Our findings could serve as a knowledge base for guiding drug discovery or drug repurposing to confront the fast-evolving SARS-CoV-2 virus and other severe infectious diseases.

From Traditional Ethnopharmacology to Modern Natural Drug Discovery: A Methodology Discussion and Specific Examples

Ethnopharmacology, through the description of the beneficial effects of plants, has provided an early framework for the therapeutic use of natural compounds. Natural products, either in their native form or after crude extraction of their active ingredients, have long been used by different populations and explored as invaluable sources for drug design. The transition from traditional ethnopharmacology to drug discovery has followed a straightforward path, assisted by the evolution of isolation and characterization methods, the increase in computational power, and the development of specific chemoinformatic methods. The deriving extensive exploitation of the natural product chemical space has led to the discovery of novel compounds with pharmaceutical properties, although this was not followed by an analogous increase in novel drugs. In this work, we discuss the evolution of ideas and methods, from traditional ethnopharmacology to in silico drug discovery, applied to natural products. We point out that, in the past, the starting point was the plant itself, identified by sustained ethnopharmacological research, with the active compound deriving after extensive analysis and testing. In contrast, in recent years, the active substance has been pinpointed by computational methods (in silico docking and molecular dynamics, network pharmacology), followed by the identification of the plant(s) containing the active ingredient, identified by existing or putative ethnopharmacological information. We further stress the potential pitfalls of recent in silico methods and discuss the absolute need for in vitro and in vivo validation as an absolute requirement. Finally, we present our contribution to natural products' drug discovery by discussing specific examples, applying the whole continuum of this rapidly evolving field. In detail, we report the isolation of novel antiviral compounds, based on natural products active against influenza and SARS-CoV-2 and novel substances active on a specific GPCR, OXER1.

BETA: a comprehensive benchmark for computational drug-target prediction

Internal validation is the most popular evaluation strategy used for drug-target predictive models. The simple random shuffling in the cross-validation, however, is not always ideal to handle large, diverse and copious datasets as it could potentially introduce bias. Hence, these predictive models cannot be comprehensively evaluated to provide insight into their general performance on a variety of use-cases (e.g. permutations of different levels of connectiveness and categories in drug and target space, as well as validations based on different data sources). In this work, we introduce a benchmark, BETA, that aims to address this gap by (i) providing an extensive multipartite network consisting of 0.97 million biomedical concepts and 8.5 million associations, in addition to 62 million drug-drug and protein-protein similarities and (ii) presenting evaluation strategies that reflect seven cases (i.e. general, screening with different connectivity, target and drug screening based on categories, searching for specific drugs and targets and drug repurposing for specific diseases), a total of seven Tests (consisting of 344 Tasks in total) across multiple sampling and validation strategies. Six state-of-the-art methods covering two broad input data types (chemical structure- and gene sequence-based and network-based) were tested across all the developed Tasks. The best-worst performing cases have been analyzed to demonstrate the ability of the proposed benchmark to identify limitations of the tested methods for running over the benchmark tasks. The results highlight BETA as a benchmark in the selection of computational strategies for drug repurposing and target discovery.

In silico Methods for Identification of Potential Therapeutic Targets

At the initial stage of drug discovery, identifying novel targets with maximal efficacy and minimal side effects can improve the success rate and portfolio value of drug discovery projects while simultaneously reducing cycle time and cost. However, harnessing the full potential of big data to narrow the range of plausible targets through existing computational methods remains a key issue in this field. This paper reviews two categories of in silico methods—comparative genomics and network-based methods—for finding potential therapeutic targets among cellular functions based on understanding their related biological processes. In addition to describing the principles, databases, software, and applications, we discuss some recent studies and prospects of the methods. While comparative genomics is mostly applied to infectious diseases, network-based methods can be applied to infectious and non-infectious diseases. Nonetheless, the methods often complement each other in their advantages and disadvantages. The information reported here guides toward improving the application of big data-driven computational methods for therapeutic target discovery.

Graphical abstract

Identification of Bioactive Compounds and Potential Mechanisms of Kuntai Capsule in the Treatment of Polycystic Ovary Syndrome by Integrating Network Pharmacology and Bioinformatics

Objective

This study elucidates the potential therapeutic targets and molecular mechanisms of KTC in the treatment of PCOS.

Materials and Methods

Using the Traditional Chinese Medicine System Pharmacology Database and Analysis Platform (TCMSP), the active ingredients and potential targets of KTC were obtained. The Gene Expression Omnibus (GEO) database was used to find differentially expressed genes (DEGs) related to PCOS. Search the CTD, DisGeNet, genecards, NCBI, OMIM, and PharmGKB databases for therapeutic targets related to PCOS. The intersection of potential targets, DEGs, and therapeutic targets was submitted to perform bioinformatics analysis by R language. Finally, the analyses' core targets and their corresponding active ingredients were molecularly docked.

Results

88 potential therapeutic targets of KTC for PCOS were discovered by intersecting the potential targets, DEGs, and therapeutic targets. According to bioinformatics analysis, the mechanisms of KTC treatment for PCOS could be linked to IL-17 signaling route, p53 signaling pathway, HIF-1 signaling pathway, etc. The minimal binding energies of the 5 core targets and their corresponding ingredients were all less than -6.5. Further research found that quercetin may replace KTC in the treatment of PCOS. Discussion and Conclusions. We explored the active ingredients and molecular mechanisms of KTC in the treatment of PCOS and found that quercetin may be the core ingredient of KTC in the treatment of PCOS.

Synergy between machine learning and natural products cheminformatics: Application to the lead discovery of anthraquinone derivatives

Cheminformatics utilizing machine learning (ML) techniques have opened up a new horizon in drug discovery. This is owing to vast chemical space expansion with rocketing numbers of expected hits and lead compounds that match druggable macromolecular targets, in particular from natural compounds. Due to the natural products' (NP) structural complexity, uniqueness, and diversity, they could occupy a bigger space in pharmaceuticals, allowing the industry to pursue more selective leads in the nanomolar range of binding affinity. ML is an essential part of each step of the drug design pipeline, such as target prediction, compound library preparation, and lead optimization. Notably, molecular mechanic and dynamic simulations, induced docking, and free energy perturbations are essential in predicting best binding poses, binding free energy values, and molecular mechanics force fields. Those applications have leveraged from artificial intelligence (AI), which decreases the computational costs required for such costly simulations. This review aimed to describe chemical space and compound libraries related to NPs. High-throughput screening utilized for fractionating NPs and high-throughput virtual screening and their strategies, and significance, are reviewed. Particular emphasis was given to AI approaches, ML tools, algorithms, and techniques, especially in drug discovery of macrocyclic compounds and approaches in computer-aided and ML-based drug discovery. Anthraquinone derivatives were discussed as a source of new lead compounds that can be developed using ML tools for diverse medicinal uses such as cancer, infectious diseases, and metabolic disorders. Furthermore, the power of principal component analysis in understanding relevant protein conformations, and molecular modeling of protein-ligand interaction were also presented. Apart from being a concise reference for cheminformatics, this review is a useful text to understand the application of ML-based algorithms to molecular dynamics simulation and in silico absorption, distribution, metabolism, excretion, and toxicity prediction.

A Network Pharmacology Approach for Uncovering the Antitumor Effects and Potential Mechanisms of the Sijunzi Decoction for the Treatment of Gastric Cancer

Background

Sijunzi decoction (SJZD), a classic Chinese formula, has been clinically used for the treatment of gastrointestinal disorders. However, few studies have uncovered its antitumor effects and its potential mechanisms against gastric cancer (GC). Therefore, this work aimed to identify the active compounds and putative targets of the SJZD and to further explore the potential mechanisms involved in the treatment of GC.

Materials and Methods

The active compounds and potential targets of the SJZD and related genes for GC treatment were collected from a public database. Traditional Chinese medicine (TCM)-compound-target-disease networks, Venn diagrams, protein–protein interactions (PPIs), gene ontology (GO), and the Kyoto Encyclopedia of Genes and Genomes (KEGG) were used to obtain the bioactive compounds, key targets, and potential pathways. Next, the human gastric adenocarcinoma cell line NUGC-4 was inoculated subcutaneously into the right flank of NCG mice to build a tumor-bearing mouse model to further verify the findings.

Results

There were 117 compounds in the SJZD in total. The SJZD and GC had 161 and 3288 potential targets, respectively, among which 123 targets overlapped. The network analysis showed that quercetin, kaempferol formononetin, ginsenoside, atractylenolide III, etc., were bioactive molecules. The tumor necrosis factor (TNF), interleukin-6 (IL-6), cellular tumor antigen p53 (TP53), transcription factor AP-1 (JUN), and vascular endothelial growth factor A (VEGFA) were potential targets. A KEGG pathway enrichment analysis revealed 110 pathways involved in the pathways for cancer, including the PI3K-AKT signaling pathway. Validation experiments showed that the SJZD inhibited tumor growth and induced apoptosis in tumor cells. In addition, the SJZD downregulated expressions of VEGFA, iNOS, COX-2, and Bax/Bcl2 and inhibited the expressions of p-PI3K and p-AKT.

Conclusion

The SJZD treats GC by inhibiting blood vessel hyperplasia and inducing cell apoptosis by regulating the PI3K/AKT pathway.

Drug Repurposing for Newly Emerged Diseases via Network-Based Inference on A Gene-Disease-Drug Network

Identification of disease-drug associations is an effective strategy for drug repurposing, especially in searching old drugs for newly emerged diseases like COVID-19. In this study, we put forward a network-based method named NEDNBI to predict disease-drug associations based on a gene-disease-drug tripartite network, which could be applied in drug repurposing. The novelty of our method lies in the fact that no negative data are required, and new disease could be added into the disease-drug network with gene as the bridge. The comprehensive evaluation results showed that the proposed method had good performance, with AUC value 0.948 ± 0.009 for 10-fold cross validation. In a case study, 8 of the 20 predicted old drugs have been tested clinically for the treatment of COVID-19, which illustrated the usefulness of our method in drug repurposing. The source code and data of the method are available at https://github.com/Qli97/NEDNBI.

Affinity2Vec: drug-target binding affinity prediction through representation learning, graph mining, and machine learning

Drug-target interaction (DTI) prediction plays a crucial role in drug repositioning and virtual drug screening. Most DTI prediction methods cast the problem as a binary classification task to predict if interactions exist or as a regression task to predict continuous values that indicate a drug's ability to bind to a specific target. The regression-based methods provide insight beyond the binary relationship. However, most of these methods require the three-dimensional (3D) structural information of targets which are still not generally available to the targets. Despite this bottleneck, only a few methods address the drug-target binding affinity (DTBA) problem from a non-structure-based approach to avoid the 3D structure limitations. Here we propose Affinity2Vec, as a novel regression-based method that formulates the entire task as a graph-based problem. To develop this method, we constructed a weighted heterogeneous graph that integrates data from several sources, including drug-drug similarity, target-target similarity, and drug-target binding affinities. Affinity2Vec further combines several computational techniques from feature representation learning, graph mining, and machine learning to generate or extract features, build the model, and predict the binding affinity between the drug and the target with no 3D structural data. We conducted extensive experiments to evaluate and demonstrate the robustness and efficiency of the proposed method on benchmark datasets used in state-of-the-art non-structured-based drug-target binding affinity studies. Affinity2Vec showed superior and competitive results compared to the state-of-the-art methods based on several evaluation metrics, including mean squared error, rm2, concordance index, and area under the precision-recall curve.

A Novel Deep Neural Network Technique for Drug–Target Interaction

Drug discovery (DD) is a time-consuming and expensive process. Thus, the industry employs strategies such as drug repositioning and drug repurposing, which allows the application of already approved drugs to treat a different disease, as occurred in the first months of 2020, during the COVID-19 pandemic. The prediction of drug–target interactions is an essential part of the DD process because it can accelerate it and reduce the required costs. DTI prediction performed in silico have used approaches based on molecular docking simulations, including similarity-based and network- and graph-based ones. This paper presents MPS2IT-DTI, a DTI prediction model obtained from research conducted in the following steps: the definition of a new method for encoding molecule and protein sequences onto images; the definition of a deep-learning approach based on a convolutional neural network in order to create a new method for DTI prediction. Training results conducted with the Davis and KIBA datasets show that MPS2IT-DTI is viable compared to other state-of-the-art (SOTA) approaches in terms of performance and complexity of the neural network model. With the Davis dataset, we obtained 0.876 for the concordance index and 0.276 for the MSE; with the KIBA dataset, we obtained 0.836 and 0.226 for the concordance index and the MSE, respectively. Moreover, the MPS2IT-DTI model represents molecule and protein sequences as images, instead of treating them as an NLP task, and as such, does not employ an embedding layer, which is present in other models.

Network-based drug repurposing: A critical review

New drug development for a disease is a tedious time taking, complex and expensive process. Even if it is done, still the chances for success of newly developed drugs are very low. Modern reports state that repurposing the pre-existing drugs will have more efficient functioning than newly developed drugs. This repurposing process will save time, reduce expenses and provide more success rate. The only limitation for this repurposing is getting a desired pharmacological and characteristic parameter of various drugs from vast data available about a huge number of drugs, their effects, and target mechanisms. This drawback can be avoided by introducing computational methods of analysis. This includes various network analysis types that use various biological processes and relationships with various drugs to make data interpretation a simple process. Some of the data sets now available in standard and simplified forms include gene expression, drug-target interactions, protein networks, electronic health records, clinical trial results, and drug adverse event reports. Integrating various data sets and interpretation methods gives way for a more efficient and easy way to repurpose an exact drug for desired target and effect. In this review, we are going to discuss briefly various computational biological network analysis methods like gene regulatory networks, metabolic networks, protein-protein interaction networks, drug-target interaction networks, drug-disease association networks, drug-drug interaction networks, drug-side effects networks, integrated network-based methods, semantic link networks, and isoform-isoform networks. Along with these, we have also briefly presented limitations, predicting methods, data sets used of various biological networks used of the drug for drug repurposing.

Network Pharmacology and Molecular Docking Analysis on Pharmacological Mechanisms of Astragalus membranaceus in the Treatment of Gastric Ulcer

BACKGROUND

Astragalus membranaceus (AM, family: Leguminosae) exerts significant therapeutic effect on gastric ulcer (GU); however, there are scarce studies on its molecular mechanism against GU. This study aims to explore the key ingredients, key targets, and potential mechanisms of AM in the treatment of GU by utilizing network pharmacology and molecular docking.

METHODS

Several public databases were used to predict the targets of AM and GU, respectively, and the drug and disease targets were intersected to obtain the common targets. Next, the key ingredients and key targets were identified by constructing ingredient-target network and protein-protein-interaction (PPI) network. Gene Ontology biological processes (GOBP) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were carried out on the common targets in order to ascertain the biological processes and signaling pathways involved. Finally, molecular docking was conducted to verify the binding affinity between the key ingredients and key targets.

RESULTS

A total of 552 predicted targets were obtained from 23 screened active ingredients, of which 203 targets were the common targets with GU. Quercetin, kaempferol, and isorhamnetin were identified as the key ingredients by constructing ingredient-target network, and TP53, AKT1, VEGFA, IL6, TNF, CASP3, and EGFR were selected as the key targets by constructing PPI network. GOBP and KEGG pathway enrichment analysis suggested that the therapeutic effect of AM on GU involved multiple biological processes and signaling pathways related to inflammation, oxidative stress, apoptosis, cell proliferation, and angiogenesis. Molecular docking validation demonstrated that all key ingredients had good binding affinity with the key targets.

CONCLUSION

This study revealed the key ingredients, key targets, and potential mechanisms of AM against GU, and these data may provide some crucial references for subsequent research and development of drugs for treating GU.

Elucidation of Potential Targets of San-Miao-San in the Treatment of Osteoarthritis Based on Network Pharmacology and Molecular Docking Analysis

Background

To examine the potential therapeutic targets of Chinese medicine formula San-Miao-San (SMS) in the treatment of osteoarthritis (OA), we analyzed the active compounds of SMS and key targets of OA and investigated the interacting pathways using network pharmacological approaches and molecular docking analysis.

Methods

The active compounds of SMS and OA-related targets were searched and screened by TCMSP, DrugBank, Genecards, OMIM, DisGeNet, TTD, and PharmGKB databases. Venn analysis and PPI were performed for evaluating the interaction of the targets. The topological analysis and molecular docking were used to confirm the subnetworks and binding affinity between active compounds and key targets, respectively. The GO and KEGG functional enrichment analysis for all targets of each subnetwork were conducted.

Results

A total of 57 active compounds and 203 targets of SMS were identified by the TCMSP and DrugBank database, while 1791 OA-related targets were collected from the Genecards, OMIM, DisGeNet, TTD, and PharmGKB databases. By Venn analysis, 108 intersection targets between SMS targets and OA targets were obtained. Most of these intersecting targets involve quercetin, kaempferol, and wogonin. Moreover, intersecting targets identified by PPI analysis were introduced into Cytoscape plug-in CytoNCA for topological analysis. Hence, nine key targets of SMS for OA treatment were obtained. Furthermore, the potential binding conformations between active compounds and key targets were found through molecular docking analysis. According to the DAVID enrichment analysis, the main biological processes of SMS in the treatment of OA include oxidative stress, response to reactive oxygen species, and apoptotic signaling pathways. Finally, we found wogonin, the key compound in SMS, might play a pivotal role on Toll-like receptor, IL-17, TNF, osteoclast differentiation, and apoptosis signaling pathways through interacting with four key targets.

Conclusions

Therefore, this study elucidated the potential active compounds and key targets of SMS in the treatment of OA based on network pharmacology.

wSDTNBI: a novel network-based inference method for virtual screening

In recent years, the rapid development of network-based methods for the prediction of drug-target interactions (DTIs) provides an opportunity for the emergence of a new type of virtual screening (VS), namely, network-based VS. Herein, we reported a novel network-based inference method named wSDTNBI. Compared with previous network-based methods that use unweighted DTI networks, wSDTNBI uses weighted DTI networks whose edge weights are correlated with binding affinities. A two-pronged approach based on weighted DTI and drug-substructure association networks was employed to calculate prediction scores. To show the practical value of wSDTNBI, we performed network-based VS on retinoid-related orphan receptor γt (RORγt), and purchased 72 compounds for experimental validation. Seven of the purchased compounds were confirmed to be novel RORγt inverse agonists by in vitro experiments, including ursonic acid and oleanonic acid with IC50 values of 10 nM and 0.28 μM, respectively. Moreover, the direct contact between ursonic acid and RORγt was confirmed using the X-ray crystal structure, and in vivo experiments demonstrated that ursonic acid and oleanonic acid have therapeutic effects on multiple sclerosis. These results indicate that wSDTNBI might be a powerful tool for network-based VS in drug discovery.

Translational Informatics for Natural Products as Antidepressant Agents

Depression, a neurological disorder, is a universally common and debilitating illness where social and economic issues could also become one of its etiologic factors. From a global perspective, it is the fourth leading cause of long-term disability in human beings. For centuries, natural products have proven their true potential to combat various diseases and disorders, including depression and its associated ailments. Translational informatics applies informatics models at molecular, imaging, individual, and population levels to promote the translation of basic research to clinical applications. The present review summarizes natural-antidepressant-based translational informatics studies and addresses challenges and opportunities for future research in the field.

ADENet: a novel network-based inference method for prediction of drug adverse events

Identification of adverse drug events (ADEs) is crucial to reduce human health risks and improve drug safety assessment. With an increasing number of biological and medical data, computational methods such as network-based methods were proposed for ADE prediction with high efficiency and low cost. However, previous network-based methods rely on the topological information of known drug-ADE networks, and hence cannot make predictions for novel compounds without any known ADE. In this study, we introduced chemical substructures to bridge the gap between the drug-ADE network and novel compounds, and developed a novel network-based method named ADENet, which can predict potential ADEs for not only drugs within the drug-ADE network, but also novel compounds outside the network. To show the performance of ADENet, we collected drug-ADE associations from a comprehensive database named MetaADEDB and constructed a series of network-based prediction models. These models obtained high area under the receiver operating characteristic curve values ranging from 0.871 to 0.947 in 10-fold cross-validation. The best model further showed high performance in external validation, which outperformed a previous network-based and a recent deep learning-based method. Using several approved drugs as case studies, we found that 32-54% of the predicted ADEs can be validated by the literature, indicating the practical value of ADENet. Moreover, ADENet is freely available at our web server named NetInfer (http://lmmd.ecust.edu.cn/netinfer). In summary, our method would provide a promising tool for ADE prediction and drug safety assessment in drug discovery and development.

Biclique extension as an effective approach to identify missing links in metabolic compound-protein interaction networks

Motivation

Metabolic networks are complex systems of chemical reactions proceeding via physical interactions between metabolites and proteins. We aimed to predict previously unknown compound-protein interactions (CPI) in metabolic networks by applying biclique extension, a network-structure-based prediction method.

Results

We developed a workflow, named BiPredict, to predict CPIs based on biclique extension and applied it to Escherichia coli and human using their respective known CPI networks as input. Depending on the chosen biclique size and using a STITCH-derived E.coli CPI network as input, a sensitivity of 39% and an associated precision of 59% was reached. For the larger human STITCH network, a sensitivity of 78% with a false-positive rate of <5% and precision of 75% was obtained. High performance was also achieved when using KEGG metabolic-reaction networks as input. Prediction performance significantly exceeded that of randomized controls and compared favorably to state-of-the-art deep-learning methods. Regarding metabolic process involvement, TCA-cycle and ribosomal processes were found enriched among predicted interactions. BiPredict can be used for network curation, may help increase the efficiency of experimental testing of CPIs, and can readily be applied to other species.

Availability and implementation

BiPredict and related datasets are available at https://github.com/SandraThieme/BiPredict.

Supplementary information

Supplementary data are available at Bioinformatics Advances online.

Machine Learning in Drug Metabolism Study

Metabolic reactions in the body transform the administered drug into metabolites. These metabolites exhibit diverse biological activities. Drug metabolism is the major underlying cause of drug overdose-related toxicity, adversative drug effects and the drug's reduced efficacy. Though metabolic reactions deactivate a drug, drug metabolites are often considered pivotal agents for off-target effects or toxicity. On the other side, in combination drug therapy, one drug may influence another drug's metabolism and clearance and is thus considered one of the primary causes of drug-drug interactions. Today with the advancement of machine learning, the metabolic fate of a drug candidate can be comprehensively studied throughout the drug development procedure. Naïve Bayes, Logistic Regression, k-Nearest Neighbours, Decision Trees, different Boosting and Ensemble methods, Support Vector Machines and Artificial Neural Network boosted Deep Learning are some machine learning algorithms which are being extensively used in such studies. Such tools are covering several attributes of drug metabolism, with an emphasis on the prediction of drug-drug interactions, drug-target-interactions, clinical drug responses, metabolite predictions, sites of metabolism, etc. These reports are crucial for evaluating metabolic stability and predicting prospective drug-drug interactions, and can help pharmaceutical companies accelerate the drug development process in a less resourcedemanding manner than what in vitro studies offer. It could also help medical practitioners to use combinatorial drug therapy in a more resourceful manner. Also, with the help of the enormous growth of deep learning, traditional fields of computational drug development like molecular interaction fields, molecular docking, quantitative structure-toactivity relationship (QSAR) studies and quantum mechanical simulations are producing results which were unimaginable couple of years back. This review provides a glimpse of a few contextually relevant machine learning algorithms and then focuses on their outcomes in different studies.

A unified drug-target interaction prediction framework based on knowledge graph and recommendation system

Prediction of drug-target interactions (DTI) plays a vital role in drug development in various areas, such as virtual screening, drug repurposing and identification of potential drug side effects. Despite extensive efforts have been invested in perfecting DTI prediction, existing methods still suffer from the high sparsity of DTI datasets and the cold start problem. Here, we develop KGE_NFM, a unified framework for DTI prediction by combining knowledge graph (KG) and recommendation system. This framework firstly learns a low-dimensional representation for various entities in the KG, and then integrates the multimodal information via neural factorization machine (NFM). KGE_NFM is evaluated under three realistic scenarios, and achieves accurate and robust predictions on four benchmark datasets, especially in the scenario of the cold start for proteins. Our results indicate that KGE_NFM provides valuable insight to integrate KG and recommendation system-based techniques into a unified framework for novel DTI discovery.

Progresses and challenges in link prediction

Link prediction is a paradigmatic problem in network science, which aims at estimating the existence likelihoods of nonobserved links, based on known topology. After a brief introduction of the standard problem and evaluation metrics of link prediction, this review will summarize representative progresses about local similarity indices, link predictability, network embedding, matrix completion, ensemble learning, and some others, mainly extracted from related publications in the last decade. Finally, this review will outline some long-standing challenges for future studies.

A drug-biomarker interaction model to predict the key targets of Scutellaria barbata D. Don in adverse-risk acute myeloid leukaemia

A poor prognosis, relapse and resistance are burning issues during adverse-risk acute myeloid leukaemia (AML) treatment. As a natural medicine, Scutellaria barbata D. Don (SBD) has shown impressive antitumour activity in various cancers. Thus, SBD may become a potential drug in adverse-risk AML treatment. This study aimed to screen the key targets of SBD in adverse-risk AML using the drug-biomarker interaction model through bioinformatics and network pharmacology methods. First, the adverse-risk AML-related critical biomarkers and targets of SBD active ingredient were obtained from The Cancer Genome Atlas database and several pharmacophore matching databases. Next, the protein-protein interaction network was constructed, and topological analysis and pathway enrichment were used to screen key targets and main pathways of intervention of SBD in adverse-risk AML. Finally, molecular docking was implemented for key target verification. The results suggest that luteolin and quercetin are the main active components of SBD against adverse-risk AML, and affected drug resistance, apoptosis, immune regulation and angiogenesis through the core targets AKT1, MAPK1, IL6, EGFR, SRC, VEGFA and TP53. We hope the proposed drug-biomarker interaction model provides an effective strategy for the research and development of antitumour drugs.

Network Pharmacology-Based Study of the Underlying Mechanisms of Huangqi Sijunzi Decoction for Alzheimer's Disease

Background

Huangqi Sijunzi decoction (HQSJZD) is a commonly used conventional Chinese herbal medicine prescription for invigorating Qi, tonifying Yang, and removing dampness. Modern pharmacology and clinical applications of HQSJZD have shown that it has a certain curative effect on Alzheimer's disease (AD).

Methods

The active components and targets of HQSJZD were searched in the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP). The genes corresponding to the targets were retrieved using UniProt and GeneCard database. The herb-compound-target network and protein-protein interaction (PPI) network were constructed by Cytoscape. The core targets of HQSJZD were analysed by Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG). The main active compounds of HQSJZD were docked with acetylcholinesterase (AChE). In vitro experiments were conducted to detect the inhibitory and neuroprotective effects of AChE.

Results

Compound-target network mainly contained 132 compounds and 255 corresponding targets. The main compounds contained quercetin, kaempferol, formononetin, isorhamnetin, hederagenin, and calycosin. Key targets contained AChE, PTGS2, PPARG, IL-1B, GSK3B, etc. There were 1708 GO items in GO enrichment analysis and 310 signalling pathways in KEGG, mainly including the cAMP signalling pathway, the vascular endothelial growth factor (VEGF) signalling pathway, serotonergic synapses, the calcium signalling pathway, type II diabetes mellitus, arginine and proline metabolism, and the longevity regulating pathway. Molecular docking showed that hederagenin and formononetin were the top 2 compounds of HQSJZD, which had a high affinity with AChE. And formononetin has a good neuroprotective effect, which can improve the oxidative damage of nerve cells.

Conclusion

HQSJZD was found to have the potential to treat AD by targeting multiple AD-related targets. Formononetin and hederagenin in HQSJZD may regulate multiple signalling pathways through AChE, which might play a therapeutic role in AD.

DTi2Vec: Drug-target interaction prediction using network embedding and ensemble learning

Drug-target interaction (DTI) prediction is a crucial step in drug discovery and repositioning as it reduces experimental validation costs if done right. Thus, developing in-silico methods to predict potential DTI has become a competitive research niche, with one of its main focuses being improving the prediction accuracy. Using machine learning (ML) models for this task, specifically network-based approaches, is effective and has shown great advantages over the other computational methods. However, ML model development involves upstream hand-crafted feature extraction and other processes that impact prediction accuracy. Thus, network-based representation learning techniques that provide automated feature extraction combined with traditional ML classifiers dealing with downstream link prediction tasks may be better-suited paradigms. Here, we present such a method, DTi2Vec, which identifies DTIs using network representation learning and ensemble learning techniques. DTi2Vec constructs the heterogeneous network, and then it automatically generates features for each drug and target using the nodes embedding technique. DTi2Vec demonstrated its ability in drug-target link prediction compared to several state-of-the-art network-based methods, using four benchmark datasets and large-scale data compiled from DrugBank. DTi2Vec showed a statistically significant increase in the prediction performances in terms of AUPR. We verified the "novel" predicted DTIs using several databases and scientific literature. DTi2Vec is a simple yet effective method that provides high DTI prediction performance while being scalable and efficient in computation, translating into a powerful drug repositioning tool.

Network-based pharmacology to predict the mechanism of Ginger and Forsythia combined treatment of viral pneumonia

BACKGROUND

Viral pneumonia (VP) is a common inflammatory disease caused by a virus in the upper respiratory tract. However, current treatment options for pneumonia are limited because of the strong infectivity and lack of research.

METHOD

Based on various databases, the mechanisms of Ginger and Forsythia were predicted by network pharmacology. The possible active ingredients of Ginger and Forsythia were obtained from the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP) and screened by pharmacokinetic parameters. Their possible targets were predicted by the TCMSP database. The VP-related targets were collected from the GeneCards and OMIM databases. The compound-target-disease network was visualized by Cytoscape 3.7.1. In addition, the protein functional annotation and identification of signalling pathways of possible targets were performed with Gene Ontology (GO) and KEGG enrichment analysis. Molecular docking was finally employed for in silico simulation matching between representative Ginger and Forsythia compounds and their core genes.

RESULTS

Twenty-eight active ingredients of Ginger and Forsythia were found and 30 common targets for the combined treatment of VP were obtained. The enrichment analysis of GO functions and KEGG pathways included 186 GO function entries and 56 KEGG pathways. Molecular docking showed that the main ingredients can closely bind three targets (CASP3, JUN, and ESR1). Thus, Ginger and Forsythia play significant roles in the prevention and treatment of VP, and this study showed their mechanism was "multicomponent, multitarget, and multipathway" for the prevention and treatment of VP.

CONCLUSION

We successfully predicted the active components and targets of Ginger and Forsythia for prevention and treatment of VP. This may systematically clarify its mechanism of action and provide a direction for future research.

The Complex Structure of the Pharmacological Drug-Disease Network

The complexity of drug–disease interactions is a process that has been explained in terms of the need for new drugs and the increasing cost of drug development, among other factors. Over the last years, diverse approaches have been explored to understand drug–disease relationships. Here, we construct a bipartite graph in terms of active ingredients and diseases based on thoroughly classified data from a recognized pharmacological website. We find that the connectivities between drugs (outgoing links) and diseases (incoming links) follow approximately a stretched-exponential function with different fitting parameters; for drugs, it is between exponential and power law functions, while for diseases, the behavior is purely exponential. The network projections, onto either drugs or diseases, reveal that the co-ocurrence of drugs (diseases) in common target diseases (drugs) lead to the appearance of connected components, which varies as the threshold number of common target diseases (drugs) is increased. The corresponding projections built from randomized versions of the original bipartite networks are considered to evaluate the differences. The heterogeneity of association at group level between active ingredients and diseases is evaluated in terms of the Shannon entropy and algorithmic complexity, revealing that higher levels of diversity are present for diseases compared to drugs. Finally, the robustness of the original bipartite network is evaluated in terms of most-connected nodes removal (direct attack) and random removal (random failures).

Effect of dapagliflozin on diabetic patients with cardiovascular disease via MAPK signalling pathway

Clinical studies have shown that dapagliflozin can reduce cardiovascular outcome in patients with type 2 diabetes mellitus (T2DM), but the exact mechanism is unclear. In this study, we used the molecular docking and network pharmacology methods to explore the potential mechanism of dapagliflozin on T2DM complicated with cardiovascular diseases (CVD). Dapagliflozin's potential targets were predicted via the Swiss Target Prediction platform. The pathogenic targets of T2DM and CVD were screened by the Online Mendelian Inheritance in Man (OMIM) and Gene Cards databases. The common targets of dapagliflozin, T2DM and CVD were used to establish a protein-protein interaction (PPI) network; the potential protein functional modules in the PPI network were found out by MCODE. Metascape tool was used for Gene Ontology (GO) and Kyoto Encyclopaedia of Genes and Genomes (KEGG) pathway enrichment analysis. A potential protein functional module with the best score was obtained from the PPI network and 9 targets in the protein functional module all showed good binding properties when docking with dapagliflozin. The results of KEGG pathway enrichment analysis showed that the underlying mechanism mainly involved AGE-RAGE signalling pathway in diabetic complications, TNF signalling pathway and MAPK signalling pathway. Significantly, the MAPK signalling pathway was considered as the key pathway. In conclusion, we speculated that dapagliflozin played a therapeutic role in T2DM complicated with CVD mainly through MAPK signalling pathway. This study preliminarily reveals the possible mechanism of dapagliflozin in the treatment of T2DM complicated with CVD and provides a theoretical basis for future clinical research.

Identifying the active compounds and mechanism of action of Banxia Xiexin decoction for treating ethanol-induced chronic gastritis using network pharmacology combined with UPLC-LTQ-Orbitrap MS

BACKGROUND

Banxia Xiexin decoction (BXD), a traditionally prescribed Chinese medicine, has been used to treat chronic gastritis for many years. However, the underlying mechanism and targets for its effects remain unknown. In the present study, we predicted the targets and active compounds of BXD in the treatment of chronic gastritis through network pharmacology and ultra-performance liquid chromatography coupled with linear trap quadrupole-Orbitrap mass spectrometry (UPLC-LTQ-Orbitrap MS).

METHOD

A chronic gastritis model was established in rats by oral administration of 56 % ethanol. BXD was orally administered for 7 days. Stomach tissues were collected for histopathological analysis, and tumour necrosis factor (TNF)-α, interleukin (IL)-2, IL-8, and lactate dehydrogenase (LDH) levels were measured by enzyme-linked immunosorbent assay. UPLC-LTQ-Orbitrap MS was established to analyse compounds in rat plasma following oral BXD administration. The absorbed ingredients were selected as candidate active compounds. The chronic gastritis-related targets were screened using multiple databases. The potential targets for the treatment of chronic gastritis were used to construct a protein-protein interaction (PPI) network and were also analysed using the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases. Finally, molecular docking was used to uncover the interaction between multi-components and putative targets, and the results were verified by surface plasmon resonance (SPR).

RESULTS

Intragastric administration of BXD ameliorated stomach injury resulting from chronic gastritis in rats and decreased the levels of TNF-α, IL-2, IL-8, and LDH. A comprehensive systematic strategy was used to successfully identify 38 candidate targets and 14 active compounds in BXD. Based on the network of compounds-targets and PPI, three hub genes that were associated with BXD therapy for chronic gastritis were selected and included intercellular adhesion molecule-1, peroxisome proliferator-activated receptor gamma and mitogen-activated protein kinase 14. The results of molecular docking and SPR demonstrated that the active compounds in BXD demonstrate affinity for these targets. Additionally, an enrichment analysis revealed that treatment of chronic gastritis with BXD primarily involves cytokine activation, the inflammatory response and nuclear factor-kappa B, hypoxia-inducible factor-1, phosphatidylinositol-3-kinase-protein-serine-threonine kinase and Janus kinase-signal transducer and activator of transcription signalling pathways, which may mediate the effects of BXD in the treatment of chronic gastritis.

CONCLUSION

BXD exhibits a therapeutic effect in ethanol-induced gastritis through multi-compound, multi-target and multi-pathway mechanisms. A strategy of network pharmacology combined with SPR may provide a feasible approach to explore the targets of herbal medicine and uncover novel bioactive components.

LigTMap: ligand and structure-based target identification and activity prediction for small molecular compounds

Target prediction is a crucial step in modern drug discovery. However, existing experimental approaches to target prediction are time-consuming and costly. Here, we introduce LigTMap, an online server with a fully automated workflow that can identify protein targets of chemical compounds among 17 classes of therapeutic proteins extracted from the PDBbind database. It combines ligand similarity search with docking and binding similarity analysis to predict putative targets. In the validation experiment of 1251 compounds, targets were successfully predicted for more than 70% of the compounds within the top-10 list. The performance of LigTMap is comparable to the current best servers SwissTargetPrediction and SEA. When testing with our newly compiled compounds from recent literature, we get improved top 10 success rate (66% ours vs. 60% SwissTargetPrediction and 64% SEA) and similar top 1 success rate (45% ours vs. 51% SwissTargetPrediction and 41% SEA). LigTMap directly provides ligand docking structures in PDB format, so that the results are ready for further structural studies in computer-aided drug design and drug repurposing projects. The LigTMap web server is freely accessible at https://cbbio.online/LigTMap. The source code is released on GitHub (https://github.com/ShirleyWISiu/LigTMap) under the BSD 3-Clause License to encourage re-use and further developments.

Discovery of Natural Products Targeting NQO1 via an Approach Combining Network-Based Inference and Identification of Privileged Substructures

NAD(P)H:quinone oxidoreductase 1 (NQO1) has been shown to be a potential therapeutic target for various human diseases, such as cancer and neurodegenerative disorders. Recent advances in computational methods, especially network-based methods, have made it possible to identify novel compounds for a target with high efficiency and low cost. In this study, we designed a workflow combining network-based methods and identification of privileged substructures to discover new compounds targeting NQO1 from a natural product library. According to the prediction results, we purchased 56 compounds for experimental validation. Without the assistance of privileged substructures, 31 compounds (31/56 = 55.4%) showed IC₅₀ < 100 μM, and 11 compounds (11/56 = 19.6%) showed IC₅₀ < 10 μM. With the assistance of privileged substructures, the two success rates were increased to 61.8 and 26.5%, respectively. Seven natural products further showed inhibitory activity on NQO1 at the cellular level, which may serve as lead compounds for further development. Moreover, network analysis revealed that osthole may exert anticancer effects against multiple cancer types by inhibiting not only carbonic anhydrases IX and XII but also NQO1. Our workflow and computational methods can be easily applied in other targets and become useful tools in drug discovery and development.