Drug-protein Network Research Articles

Complex graph neural networks for medication interaction verification

This paper presents the development and application of graph neural networks to verify drug interactions, consisting of drug-protein networks. For this, the DrugBank databases were used, creating four complex networks of interactions: target proteins, transport proteins, carrier proteins, and enzymes. The Louvain and Girvan-Newman community detection algorithms were used to establish communities and validate the interactions between them. Positive results were obtained when checking the interactions of two sets of drugs for disease treatments: diabetes and anxiety; diabetes and antibiotics. There were found 371 interactions by the Girvan-Newman algorithm and 58 interactions via Louvain.

An Integrative Heterogeneous Graph Neural Network-Based Method for Multi-Labeled Drug Repurposing.

Drug repurposing is the process of discovering new indications (i.e., diseases or conditions) for already approved drugs. Many computational methods have been proposed for predicting new associations between drugs and diseases. In this article, we proposed a new method, called DR-HGNN, an integrative heterogeneous graph neural network-based method for multi-labeled drug repurposing, to discover new indications for existing drugs. For this purpose, we first used the DTINet dataset to construct a heterogeneous drug–protein–disease (DPD) network, which is a graph composed of four types of nodes (drugs, proteins, diseases, and drug side effects) and eight types of edges. Second, we labeled each drug–protein edge, dp i,j = (d i , p j ), of the DPD network with a set of diseases, {δ i,j,1, … , δ i,j,k } associated with both d i and p j and then devised multi-label ranking approaches which incorporate neural network architecture that operates on the heterogeneous graph-structured data and which leverages both the interaction patterns and the features of drug and protein nodes. We used a derivative of the GraphSAGE algorithm, HinSAGE, on the heterogeneous DPD network to learn low-dimensional vector representation of features of drugs and proteins. Finally, we used the drug–protein network to learn the embeddings of the drug–protein edges and then predict the disease labels that act as bridges between drugs and proteins. The proposed method shows better results than existing methods applied to the DTINet dataset, with an AUC of 0.964.

GVDTI: graph convolutional and variational autoencoders with attribute-level attention for drug-protein interaction prediction.

Identifying proteins that interact with drugs plays an important role in the initial period of developing drugs, which helps to reduce the development cost and time. Recent methods for predicting drug-protein interactions mainly focus on exploiting various data about drugs and proteins. These methods failed to completely learn and integrate the attribute information of a pair of drug and protein nodes and their attribute distribution. We present a new prediction method, GVDTI, to encode multiple pairwise representations, including attention-enhanced topological representation, attribute representation and attribute distribution. First, a framework based on graph convolutional autoencoder is constructed to learn attention-enhanced topological embedding that integrates the topology structure of a drug-protein network for each drug and protein nodes. The topological embeddings of each drug and each protein are then combined and fused by multi-layer convolution neural networks to obtain the pairwise topological representation, which reveals the hidden topological relationships between drug and protein nodes. The proposed attribute-wise attention mechanism learns and adjusts the importance of individual attribute in each topological embedding of drug and protein nodes. Secondly, a tri-layer heterogeneous network composed of drug, protein and disease nodes is created to associate the similarities, interactions and associations across the heterogeneous nodes. The attribute distribution of the drug-protein node pair is encoded by a variational autoencoder. The pairwise attribute representation is learned via a multi-layer convolutional neural network to deeply integrate the attributes of drug and protein nodes. Finally, the three pairwise representations are fused by convolutional and fully connected neural networks for drug-protein interaction prediction. The experimental results show that GVDTI outperformed other seven state-of-the-art methods in comparison. The improved recall rates indicate that GVDTI retrieved more actual drug-protein interactions in the top ranked candidates than conventional methods. Case studies on five drugs further confirm GVDTI's ability in discovering the potential candidate drug-related proteins. zhang@hlju.edu.cn Supplementary information: Supplementary data are available at Briefings in Bioinformatics online.

Growth Factor Receptor Signaling Inhibition Prevents SARS-CoV-2 Replication

SARS-CoV-2 infections are rapidly spreading around the globe. The rapid development of therapies is of major importance. However, our lack of understanding of the molecular processes and host cell signaling events underlying SARS-CoV-2 infection hinders therapy development. We use a SARS-CoV-2 infection system in permissible human cells to study signaling changes by phosphoproteomics. We identify viral protein phosphorylation and define phosphorylation-driven host cell signaling changes upon infection. Growth factor receptor (GFR) signaling and downstream pathways are activated. Drug-protein network analyses revealed GFR signaling as key pathways targetable by approved drugs. The inhibition of GFR downstream signaling by five compounds prevents SARS-CoV-2 replication in cells, assessed by cytopathic effect, viral dsRNA production, and viral RNA release into the supernatant. This study describes host cell signaling events upon SARS-CoV-2 infection and reveals GFR signaling as a central pathway essential for SARS-CoV-2 replication. It provides novel strategies for COVID-19 treatment.

Drug-target interaction prediction using semi-bipartite graph model and deep learning

BackgroundIdentifying drug-target interaction is a key element in drug discovery. In silico prediction of drug-target interaction can speed up the process of identifying unknown interactions between drugs and target proteins. In recent studies, handcrafted features, similarity metrics and machine learning methods have been proposed for predicting drug-target interactions. However, these methods cannot fully learn the underlying relations between drugs and targets. In this paper, we propose anew framework for drug-target interaction prediction that learns latent features from drug-target interaction network.ResultsWe present a framework to utilize the network topology and identify interacting and non-interacting drug-target pairs. We model the problem as a semi-bipartite graph in which we are able to use drug-drug and protein-protein similarity in a drug-protein network. We have then used a graph labeling method for vertex ordering in our graph embedding process. Finally, we employed deep neural network to learn the complex pattern of interacting pairs from embedded graphs. We show our approach is able to learn sophisticated drug-target topological features and outperforms other state-of-the-art approaches.ConclusionsThe proposed learning model on semi-bipartite graph model, can integrate drug-drug and protein-protein similarities which are semantically different than drug-protein information in a drug-target interaction network. We show our model can determine interaction likelihood for each drug-target pair and outperform other heuristics.

Drug-target interaction prediction with tree-ensemble learning and output space reconstruction

BackgroundComputational prediction of drug-target interactions (DTI) is vital for drug discovery. The experimental identification of interactions between drugs and target proteins is very onerous. Modern technologies have mitigated the problem, leveraging the development of new drugs. However, drug development remains extremely expensive and time consuming. Therefore, in silico DTI predictions based on machine learning can alleviate the burdensome task of drug development. Many machine learning approaches have been proposed over the years for DTI prediction. Nevertheless, prediction accuracy and efficiency are persisting problems that still need to be tackled. Here, we propose a new learning method which addresses DTI prediction as a multi-output prediction task by learning ensembles of multi-output bi-clustering trees (eBICT) on reconstructed networks. In our setting, the nodes of a DTI network (drugs and proteins) are represented by features (background information). The interactions between the nodes of a DTI network are modeled as an interaction matrix and compose the output space in our problem. The proposed approach integrates background information from both drug and target protein spaces into the same global network framework.ResultsWe performed an empirical evaluation, comparing the proposed approach to state of the art DTI prediction methods and demonstrated the effectiveness of the proposed approach in different prediction settings. For evaluation purposes, we used several benchmark datasets that represent drug-protein networks. We show that output space reconstruction can boost the predictive performance of tree-ensemble learning methods, yielding more accurate DTI predictions.ConclusionsWe proposed a new DTI prediction method where bi-clustering trees are built on reconstructed networks. Building tree-ensemble learning models with output space reconstruction leads to superior prediction results, while preserving the advantages of tree-ensembles, such as scalability, interpretability and inductive setting.

Abstract 3395: TargetTranslator: Big data identifies non-canonical targets for high risk neuroblastoma

Abstract Despite the many advances in the molecular characterization of neuroblastoma, effective treatments for high-risk patients are currently lacking. Using publically available data, integrated computational analysis offers new opportunities to uncover druggable subgroups. In the TargetTranslator project, we have combined state-of-the-art Big Data techniques to identify new targets in high-risk subgroups of neuroblastoma. Starting with clinical or genomic risk factors, TargetTranslator builds a consensus molecular signature across cohorts. This signature is scored against a massive amount of data from (i) childhood tumor biobanks (TARGET, R2), (ii) drug profiling data from cancer cell models (NIH-LINCS), and (iii) drug targets and pathways (STITCH, STRING, MSIGDB). As output, the system provides ranked lists of compounds, molecular targets and pathways for each risk factor, as well as an aggregated probability score. In a proof-of-concept study, we used TargetTranslator to recommend treatments for high-risk neuroblastoma subgroups, including COG high-risk, MYCN amplification, 11q deletion, and signatures of differentiation and ALK activation. Depending on risk group stratification, we detected between 21 and 680 substances (p<0.001), most of which were strongly associated to approximately 10 key targets. In addition to confirming a key role for the PI3K/mTOR and MAPK pathways, we detected high scoring compounds in non-canonical pathways, including neurotransmission, GLI/SMO, ROCK and PKA. Experimental validation of 11 predicted drugs in patient-derived neuroblastoma lines confirmed our signatures and reduced viability. We also identified four new compounds that significantly (p<0.05) suppressed MYCN protein levels (ROCK inhibitor fasudil, CNR2 agonist GW405833, CDK inhibitor AZD5438, lovastatin) at concentrations non-toxic in zebrafish embryos. By integrating data from patients, drugs, and drug-protein networks, we establish a new computational pipeline that predicts protein targets in specific patient subgroups. The TargetTranslator introduces a way to bridge drug effects with patient data and will be available as a user-friendly web tool on www.targettranslator.org in 2018. Citation Format: Elin Almstedt, Caroline Wärn, Ramy Elgendy, Neda Hekmati, Emil Rosén, Ida Larsson, Rebecka Jörnsten, Cecilia Crona, Sven Nelander. TargetTranslator: Big data identifies non-canonical targets for high risk neuroblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3395.

Network Pharmacology and Reverse Molecular Docking-Based Prediction of the Molecular Targets and Pathways for Avicularin Against Cancer.

Avicularin has been found to inhibit the proliferation of HepG-2 cells in vitro in the screening of our laboratory. We intended to explain the molecular mechanism of this effect. Therefore, the combined methods of reverse molecular docking and network pharmacology were used in order to illuminate the molecular mechanisms for Avicularin against cancer. Potential targets associated with anti-tumor effects of Avicularin were screened by reverse molecular docking, then a protein database was established through constructing the drugprotein network from literature mining data, and the protein-protein network was built through an in-depth exploration of the relationships between the proteins, and then the network topology analysis was performed. Additionally, gene function and signaling pathways were analyzed by Go bio-enrichment and KEGG Pathway. The result showed that Avicularin was closely related to 16 targets associated with cancer, and it may significantly influence the pro-survival signals in MAPK signaling pathway that can activate and regulate a series of cellular activities and participate in the regulation of cell proliferation, differentiation, transformation and apoptosis. The network pharmacology strategy used herein provided a powerful means for the mechanisms of action for bioactive ingredients.

The Combination of Three Components Derived from Sheng MaiSan Protects Myocardial Ischemic Diseases and Inhibits Oxidative Stress via Modulating MAPKs and JAK2-STAT3 Signaling Pathways Based on Bioinformatics Approach.

GRS is a drug combination of three components including ginsenoside Rb1, ruscogenin and schisandrin. It derived from the well-known TCM formula Sheng MaiSan, a widely used traditional Chinese medicine for the treatment of cardiovascular diseases in clinic. The present study illuminates its underlying mechanisms against myocardial ischemic diseases based on the combined methods of bioinformatic prediction and experimental verification. A protein database was established through constructing the drug-protein network. And the target-pathway interaction network clustered the potential signaling pathways and targets of GRS in treatment of myocardial ischemic diseases. Several target proteins, such as NFKB1, STAT3 and MAPK14, were identified as the candidate key proteins, and MAPKs and JAK-STAT signaling pathway were suggested as the most related pathways, which were in accordance with the gene ontology analysis. Then, the predictive results were further validated and we found that GRS treatment alleviated hypoxia/reoxygenation (H/R)-induced cardiomyocytes injury via suppression of MDA levels and ROS generation, and potential mechanisms might related to the suppression of activation of MAPKs and JAK2-STAT3 signaling pathways. Conclusively, our results offer the evidence that GRS attenuates myocardial ischemia injury via regulating oxidative stress and MAPKs and JAK2-STAT3 signaling pathways, which supplied some new insights for its prevention and treatment of myocardial ischemia diseases.

Synergistic drug combinations from electronic health records and gene expression.

ABSTRACT Objective: Using electronic health records (EHRs) and biomolecular data, we sought to discover drug pairs with synergistic repurposing potential. EHRs provide real-world treatment and outcome patterns, while complementary biomolecular data, including disease-specific gene expression and drug-protein interactions, provide mechanistic understanding. Method: We applied Group Lasso INTERaction NETwork (glinternet), an overlap group lasso penalty on a logistic regression model, with pairwise interactions to identify variables and interacting drug pairs associated with reduced 5-year mortality using EHRs of 9945 breast cancer patients. We identified differentially expressed genes from 14 case-control human breast cancer gene expression datasets and integrated them with drug-protein networks. Drugs in the network were scored according to their association with breast cancer individually or in pairs. Lastly, we determined whether synergistic drug pairs found in the EHRs were enriched among synergistic drug pairs from gene-expression data using a method similar to gene set enrichment analysis. Results: From EHRs, we discovered 3 drug-class pairs associated with lower mortality: anti-inflammatories and hormone antagonists, anti-inflammatories and lipid modifiers, and lipid modifiers and obstructive airway drugs. The first 2 pairs were also enriched among pairs discovered using gene expression data and are supported by molecular interactions in drug-protein networks and preclinical and epidemiologic evidence. Conclusions: This is a proof-of-concept study demonstrating that a combination of complementary data sources, such as EHRs and gene expression, can corroborate discoveries and provide mechanistic insight into drug synergism for repurposing.

Screening lifespan-extending drugs in Caenorhabditis elegans via label propagation on drug-protein networks.

BackgroundOne of the most challenging tasks in the exploration of anti-aging is to discover drugs that can promote longevity and delay the incidence of age-associated diseases of human. Up to date, a number of drugs, including some antioxidants, metabolites and synthetic compounds, have been found to effectively delay the aging of nematodes and insects.ResultsWe proposed a label propagation algorithm on drug-protein network to infer drugs that can extend the lifespan of C. elegans. We collected a set of drugs of which functions on lifespan extension of C. elegans have been reliably determined, and then built a large-scale drug-protein network by collecting a set of high-confidence drugprotein interactions. A label propagation algorithm was run on the drug-protein bipartite network to predict new drugs with lifespan-extending effect on C. elegans. We calibrated the performance of the proposed method by conducting performance comparison with two classical models, kNN and SVM. We also showed that the screened drugs significantly mediate in the aging-related pathways, and have higher chemical similarities to the effective drugs than ineffective drugs in promoting longevity of C. elegans. Moreover, we carried out wet-lab experiments to verify a screened drugs, 2- Bromo-4’-nitroacetophenone, and found that it can effectively extend the lifespan of C. elegans. These results showed that our method is effective in screening lifespanextending drugs in C. elegans.ConclusionsIn this paper, we proposed a semi-supervised algorithm to predict drugs with lifespan-extending effects on C. elegans. In silico empirical evaluations and in vivo experiments in C. elegans have demonstrated that our method can effectively narrow down the scope of candidate drugs needed to be verified by wet lab experiments.Electronic supplementary materialThe online version of this article (doi:10.1186/s12918-016-0362-4) contains supplementary material, which is available to authorized users.

A Computational Drug Repositioning Approach for Targeting Oncogenic Transcription Factors.

Mutations in transcription factor (TF) genes are frequently observed in tumors, often leading to aberrant transcriptional activity. Unfortunately, TFs are often considered undruggable due to the absence of targetable enzymatic activity. To address this problem, we developed CRAFTT, a computational drug-repositioning approach for targeting TF activity. CRAFTT combines ChIP-seq with drug-induced expression profiling to identify small molecules that can specifically perturb TF activity. Application to ENCODE ChIP-seq datasets revealed known drug-TF interactions, and a global drug-protein network analysis supported these predictions. Application of CRAFTT to ERG, a pro-invasive, frequently overexpressed oncogenic TF, predicted that dexamethasone would inhibit ERG activity. Dexamethasone significantly decreased cell invasion and migration in an ERG-dependent manner. Furthermore, analysis of electronic medical record data indicates a protective role for dexamethasone against prostate cancer. Altogether, our method provides a broadly applicable strategy for identifying drugs that specifically modulate TF activity.

Genome and network visualization facilitates the analyses of the effects of drugs and mutations on protein-protein and drug-protein networks.

BackgroundBiologists generally interrogate genomics data using web-based genome browsers that have limited analytical potential. New generation genome browsers such as the Integrated Genome Browser (IGB) have largely overcome this limitation and permit customized analyses to be implemented using plugins. We illustrate the use of a plugin for IGB that exploits advanced visualization techniques to integrate the analysis of genomics data with network and structural approaches.ResultsWe show how visualization technologies that combine both genomics and network biology can facilitate the selection of the key amino acid contacts from protein-protein and protein-drug interactions. Starting from the MDM2-P53 interaction, which is a high-value target for cancer therapy, and Nutlin, the parent small molecule of an MDM2 antagonist that is currently in clinical trials, we show that this method can be generalized to analyze how drugs and mutations can interfere with both protein-protein and drug-protein networks. We illustrate this point by two additional use-cases exploring the molecular basis of tamoxifen side effects and of drug resistance in chronic myeloid leukemia patients.ConclusionsCombined network and structure biology approaches provide key insights into both the genetic and the edgetic roles of variants in diseases. 3D interactomes facilitate the identification of disease-relevant interactions that can then be specifically targeted by drugs. Recent advances in molecular interaction and structure visualization tools have greatly simplified the mapping of mutated residues to molecular interaction interfaces. Such approaches can now also be integrated with genome visualization tools to enable comparative analyses of interaction contacts.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-016-0908-x) contains supplementary material, which is available to authorized users.

Network pharmacology-based prediction and verification of the molecular targets and pathways for schisandrin against cerebrovascular disease

AimTo illuminate the molecular targets for schisandrin against cerebrovascular disease based on the combined methods of network pharmacology prediction and experimental verification. MethodA protein database was established through constructing the drug-protein network from literature mining data. The protein-protein network was built through an in-depth exploration of the relationships between the proteins. The computational platform was implemented to predict and extract the sensitive sub-network with significant P-values from the protein-protein network. Then the key targets and pathways were identified from the sensitive sub-network. The most related targets and pathways were also confirmed in hydrogen peroxide (H2O2)-induced PC12 cells by Western blotting. ResultsTwelve differentially expressed proteins (gene names: NFKB1, RELA, TNFSF10, MAPK1, CHUK, CASP8, PIGS2, MAPK14, CREB1, IFNG, APP, and BCL2) were confirmed as the central nodes of the interaction network (45 nodes, 93 edges). The NF-κB signaling pathway was suggested as the most related pathway of schisandrin for cerebrovascular disease. Furthermore, schisandrin was found to suppress the expression and phosphorylation of IKKα, as well as p50 and p65 induced by H2O2 in PC12 cells by Western blotting. ConclusionThe computational platform that integrates literature mining data, protein-protein interactions, sensitive sub-network, and pathway results in identification of the NF-κB signaling pathway as the key targets and pathways for schisandrin.

Pathway-pathway network-based study of the therapeutic mechanisms by which salvianolic acid B regulates cardiovascular diseases

Investigation of the potential therapeutic mechanisms of drug candidates is an essential step in the process of new drug discovery. With the rapid development of systems biology, recent network analyses of proteins, drugs, and diseases have enabled great progress in delineating the molecule mechanisms of drug candidates. However, most analyses perform a direct association between gene/protein and disease levels without considering the intermediate biological pathways regulated by the drugs. Given that a protein performs its biological roles through pathways, we propose using a novel pathway-pathway network analysis to investigate the potential therapeutic functions of the drug candidates. Many studies have demonstrated that salvianolic acid B (Sal B) of Salvia miltiorrhiza is an effective therapy for cardiovascular diseases (CVD). Using molecular docking methods to identify direct interacting targets of Sal B, we collected all Sal B-regulated proteins with supporting experimental evidence in PubMed abstracts. FDA-approved CVD drugs and their corresponding targets were also collected. From a traditional drug-protein network analysis, we found that Sal B could affect ACE and REN of the renin-angiotensin-aldosterone system to relax vessels and alleviate hypertension. Subsequent pathway-pathway network analysis was attempted to study the mechanisms of Sal B in treating CVD, and demonstrated that Sal B regulates immunity/inflammation, apoptosis, ion transport and basic metabolism processes in the treatment of CVD. Regulating the immune/inflammation process may be the major mechanism of Sal B. We believe that pathway-pathway network analysis is a novel method for studying the therapeutic mechanisms of herbal ingredients.