Computational drug repositioning using meta-path-based semantic network analysis

Background

Drug repositioning is a promising and efficient way to discover new indications for existing drugs, which holds the great potential for precision medicine in the post-genomic era. Many network-based approaches have been proposed for drug repositioning based on similarity networks, which integrate multiple sources of drugs and diseases. However, these methods may simply view nodes as the same-typed and neglect the semantic meanings of different meta-paths in the heterogeneous network. Therefore, it is urgent to develop a rational method to infer new indications for approved drugs.

Results

In this study, we proposed a novel methodology named HeteSim_DrugDisease (HSDD) for the prediction of drug repositioning. Firstly, we build the drug-drug similarity network and disease-disease similarity network by integrating the information of drugs and diseases. Secondly, a drug-disease heterogeneous network is constructed, which combines the drug similarity network, disease similarity network as well as the known drug-disease association network. Finally, HSDD predicts novel drug-disease associations based on the HeteSim scores of different meta-paths. The experimental results show that HSDD performs significantly better than the existing state-of-the-art approaches. HSDD achieves an AUC score of 0.8994 in the leave-one-out cross validation experiment. Moreover, case studies for selected drugs further illustrate the practical usefulness of HSDD.

Conclusions

HSDD can be an effective and feasible way to infer the associations between drugs and diseases using on meta-path-based semantic network analysis.

     

A Network-Based Classification Model for Deriving Novel Drug-Disease Associations and Assessing Their Molecular Actions

The growing number and variety of genetic network datasets increases the feasibility of understanding how drugs and diseases are associated at the molecular level. Properly selected features of the network representations of existing drug-disease associations can be used to infer novel indications of existing drugs. To find new drug-disease associations, we generated an integrative genetic network using combinations of interactions, including protein-protein interactions and gene regulatory network datasets. Within this network, network adjacencies of drug-drug and disease-disease were quantified using a scored path between target sets of them. Furthermore, the common topological module of drugs or diseases was extracted, and thereby the distance between topological drug-module and disease (or disease-module and drug) was quantified. These quantified scores were used as features for the prediction of novel drug-disease associations. Our classifiers using Random Forest, Multilayer Perceptron and C4.5 showed a high specificity and sensitivity (AUC score of 0.855, 0.828 and 0.797 respectively) in predicting novel drug indications, and displayed a better performance than other methods with limited drug and disease properties. Our predictions and current clinical trials overlap significantly across the different phases of drug development. We also identified and visualized the topological modules of predicted drug indications for certain types of cancers, and for Alzheimer’s disease. Within the network, those modules show potential pathways that illustrate the mechanisms of new drug indications, including propranolol as a potential anticancer agent and telmisartan as treatment for Alzheimer’s disease.     

Predicting drug-disease interactions by semi-supervised graph cut algorithm and three-layer data integration

Background

Prediction of drug-disease interactions is promising for either drug repositioning or disease treatment fields. The discovery of novel drug-disease interactions, on one hand can help to find novel indictions for the approved drugs; on the other hand can provide new therapeutic approaches for the diseases. Recently, computational methods for finding drug-disease interactions have attracted lots of attention because of their far more higher efficiency and lower cost than the traditional wet experiment methods. However, they still face several challenges, such as the organization of the heterogeneous data, the performance of the model, and so on.

Methods

In this work, we present to hierarchically integrate the heterogeneous data into three layers. The drug-drug and disease-disease similarities are first calculated separately in each layer, and then the similarities from three layers are linearly fused into comprehensive drug similarities and disease similarities, which can then be used to measure the similarities between two drug-disease pairs. We construct a novel weighted drug-disease pair network, where a node is a drug-disease pair with known or unknown treatment relation, an edge represents the node-node relation which is weighted with the similarity score between two pairs. Now that similar drug-disease pairs are supposed to show similar treatment patterns, we can find the optimal graph cut of the network. The drug-disease pair with unknown relation can then be considered to have similar treatment relation with that within the same cut. Therefore, we develop a semi-supervised graph cut algorithm, SSGC, to find the optimal graph cut, based on which we can identify the potential drug-disease treatment interactions.

Results

By comparing with three representative network-based methods, SSGC achieves the highest performances, in terms of both AUC score and the identification rates of true drug-disease pairs. The experiments with different integration strategies also demonstrate that considering several sources of data can improve the performances of the predictors. Further case studies on four diseases, the top-ranked drug-disease associations have been confirmed by KEGG, CTD database and the literature, illustrating the usefulness of SSGC.

Conclusions

The proposed comprehensive similarity scores from multi-views and multiple layers and the graph-cut based algorithm can greatly improve the prediction performances of drug-disease associations.

     

Annotating Diseases Using Human Phenotype Ontology Improves Prediction of Disease-Associated Long Non-coding RNAs

Recently, many long non-coding RNAs (lncRNAs) have been identified and their biological function has been characterized; however, our understanding of their underlying molecular mechanisms related to disease is still limited. To overcome the limitation in experimentally identifying disease–lncRNA associations, computational methods have been proposed as a powerful tool to predict such associations. These methods are usually based on the similarities between diseases or lncRNAs since it was reported that similar diseases are associated with functionally similar lncRNAs. Therefore, prediction performance is highly dependent on how well the similarities can be captured. Previous studies have calculated the similarity between two diseases by mapping exactly each disease to a single Disease Ontology (DO) term, and then use a semantic similarity measure to calculate the similarity between them. However, the problem of this approach is that a disease can be described by more than one DO terms. Until now, there is no annotation database of DO terms for diseases except for genes. In contrast, Human Phenotype Ontology (HPO) is designed to fully annotate human disease phenotypes. Therefore, in this study, we constructed disease similarity networks/matrices using HPO instead of DO. Then, we used these networks/matrices as inputs of two representative machine learning-based and network-based ranking algorithms, that is, regularized least square and heterogeneous graph-based inference, respectively. The results showed that the prediction performance of the two algorithms on HPO-based is better than that on DO-based networks/matrices. In addition, our method can predict 11 novel cancer-associated lncRNAs, which are supported by literature evidence.     

Inferring Novel Indications of Approved Drugs via a Learning Method with Local and Global Consistency

Inferring new indication of approved drugs is critical not only for the elucidation of the interaction mechanisms between these drugs and their associated diseases, but also for the development of drug therapy for various human diseases. This paper proposes a network-based approach to reveal the association between 52 human diseases and potential therapeutic drugs based on multiple types of data. The advantage of the approach is that it can obtain the global relevance features for each drug-disease pair in the network by the learning local and global consistency method (LLGC). Cross-validation tests results demonstrate the proposed approach can achieve better performance comparing with previous methods. More importantly, it provides a promising strategy to maximize the value of therapeutic drugs and offer safe and effective treatments for different diseases.     

Drug Repositioning by Kernel-Based Integration of Molecular Structure,Molecular Activity,and Phenotype Data

Computational inference of novel therapeutic values for existing drugs, i.e., drug repositioning, offers the great prospect for faster and low-risk drug development. Previous researches have indicated that chemical structures, target proteins, and side-effects could provide rich information in drug similarity assessment and further disease similarity. However, each single data source is important in its own way and data integration holds the great promise to reposition drug more accurately. Here, we propose a new method for drug repositioning, PreDR (Predict Drug Repositioning), to integrate molecular structure, molecular activity, and phenotype data. Specifically, we characterize drug by profiling in chemical structure, target protein, and side-effects space, and define a kernel function to correlate drugs with diseases. Then we train a support vector machine (SVM) to computationally predict novel drug-disease interactions. PreDR is validated on a well-established drug-disease network with 1,933 interactions among 593 drugs and 313 diseases. By cross-validation, we find that chemical structure, drug target, and side-effects information are all predictive for drug-disease relationships. More experimentally observed drug-disease interactions can be revealed by integrating these three data sources. Comparison with existing methods demonstrates that PreDR is competitive both in accuracy and coverage. Follow-up database search and pathway analysis indicate that our new predictions are worthy of further experimental validation. Particularly several novel predictions are supported by clinical trials databases and this shows the significant prospects of PreDR in future drug treatment. In conclusion, our new method, PreDR, can serve as a useful tool in drug discovery to efficiently identify novel drug-disease interactions. In addition, our heterogeneous data integration framework can be applied to other problems.     

SAveRUNNER: A network-based algorithm for drug repurposing and its application to COVID-19

The novelty of new human coronavirus COVID-19/SARS-CoV-2 and the lack of effective drugs and vaccines gave rise to a wide variety of strategies employed to fight this worldwide pandemic. Many of these strategies rely on the repositioning of existing drugs that could shorten the time and reduce the cost compared to de novo drug discovery. In this study, we presented a new network-based algorithm for drug repositioning, called SAveRUNNER (Searching off-lAbel dRUg aNd NEtwoRk), which predicts drug–disease associations by quantifying the interplay between the drug targets and the disease-specific proteins in the human interactome via a novel network-based similarity measure that prioritizes associations between drugs and diseases locating in the same network neighborhoods. Specifically, we applied SAveRUNNER on a panel of 14 selected diseases with a consolidated knowledge about their disease-causing genes and that have been found to be related to COVID-19 for genetic similarity (i.e., SARS), comorbidity (e.g., cardiovascular diseases), or for their association to drugs tentatively repurposed to treat COVID-19 (e.g., malaria, HIV, rheumatoid arthritis). Focusing specifically on SARS subnetwork, we identified 282 repurposable drugs, including some the most rumored off-label drugs for COVID-19 treatments (e.g., chloroquine, hydroxychloroquine, tocilizumab, heparin), as well as a new combination therapy of 5 drugs (hydroxychloroquine, chloroquine, lopinavir, ritonavir, remdesivir), actually used in clinical practice. Furthermore, to maximize the efficiency of putative downstream validation experiments, we prioritized 24 potential anti-SARS-CoV repurposable drugs based on their network-based similarity values. These top-ranked drugs include ACE-inhibitors, monoclonal antibodies (e.g., anti-IFNγ, anti-TNFα, anti-IL12, anti-IL1β, anti-IL6), and thrombin inhibitors. Finally, our findings were in-silico validated by performing a gene set enrichment analysis, which confirmed that most of the network-predicted repurposable drugs may have a potential treatment effect against human coronavirus infections.     

In silico prediction of potential miRNA‐disease association using an integrative bioinformatics approach based on kernel fusion

Accumulating experimental evidence has demonstrated that microRNAs (miRNAs) have a huge impact on numerous critical biological processes and they are associated with different complex human diseases. Nevertheless, the task to predict potential miRNAs related to diseases remains difficult. In this paper, we developed a Kernel Fusion‐based Regularized Least Squares for MiRNA‐Disease Association prediction model (KFRLSMDA), which applied kernel fusion technique to fuse similarity matrices and then utilized regularized least squares to predict potential miRNA‐disease associations. To prove the effectiveness of KFRLSMDA, we adopted leave‐one‐out cross‐validation (LOOCV) and 5‐fold cross‐validation and then compared KFRLSMDA with 10 previous computational models (MaxFlow, MiRAI, MIDP, RKNNMDA, MCMDA, HGIMDA, RLSMDA, HDMP, WBSMDA and RWRMDA). Outperforming other models, KFRLSMDA achieved AUCs of 0.9246 in global LOOCV, 0.8243 in local LOOCV and average AUC of 0.9175 ± 0.0008 in 5‐fold cross‐validation. In addition, respectively, 96%, 100% and 90% of the top 50 potential miRNAs for breast neoplasms, colon neoplasms and oesophageal neoplasms were confirmed by experimental discoveries. We also predicted potential miRNAs related to hepatocellular cancer by removing all known related miRNAs of this cancer and 98% of the top 50 potential miRNAs were verified. Furthermore, we predicted potential miRNAs related to lymphoma using the data set in the old version of the HMDD database and 80% of the top 50 potential miRNAs were confirmed. Therefore, it can be concluded that KFRLSMDA has reliable prediction performance.     

Analysis of the Robustness of Network-Based Disease-Gene Prioritization Methods Reveals Redundancy in the Human Interactome and Functional Diversity of Disease-Genes

Complex biological systems usually pose a trade-off between robustness and fragility where a small number of perturbations can substantially disrupt the system. Although biological systems are robust against changes in many external and internal conditions, even a single mutation can perturb the system substantially, giving rise to a pathophenotype. Recent advances in identifying and analyzing the sequential variations beneath human disorders help to comprehend a systemic view of the mechanisms underlying various disease phenotypes. Network-based disease-gene prioritization methods rank the relevance of genes in a disease under the hypothesis that genes whose proteins interact with each other tend to exhibit similar phenotypes. In this study, we have tested the robustness of several network-based disease-gene prioritization methods with respect to the perturbations of the system using various disease phenotypes from the Online Mendelian Inheritance in Man database. These perturbations have been introduced either in the protein-protein interaction network or in the set of known disease-gene associations. As the network-based disease-gene prioritization methods are based on the connectivity between known disease-gene associations, we have further used these methods to categorize the pathophenotypes with respect to the recoverability of hidden disease-genes. Our results have suggested that, in general, disease-genes are connected through multiple paths in the human interactome. Moreover, even when these paths are disturbed, network-based prioritization can reveal hidden disease-gene associations in some pathophenotypes such as breast cancer, cardiomyopathy, diabetes, leukemia, parkinson disease and obesity to a greater extend compared to the rest of the pathophenotypes tested in this study. Gene Ontology (GO) analysis highlighted the role of functional diversity for such diseases.     

Prediction of drugs having opposite effects on disease genes in a directed network

Background

Developing novel uses of approved drugs, called drug repositioning, can reduce costs and times in traditional drug development. Network-based approaches have presented promising results in this field. However, even though various types of interactions such as activation or inhibition exist in drug-target interactions and molecular pathways, most of previous network-based studies disregarded this information.

Methods

We developed a novel computational method, Prediction of Drugs having Opposite effects on Disease genes (PDOD), for identifying drugs having opposite effects on altered states of disease genes. PDOD utilized drug-drug target interactions with ‘effect type’, an integrated directed molecular network with ‘effect type’ and ‘effect direction’, and disease genes with regulated states in disease patients. With this information, we proposed a scoring function to discover drugs likely to restore altered states of disease genes using the path from a drug to a disease through the drug-drug target interactions, shortest paths from drug targets to disease genes in molecular pathways, and disease gene-disease associations.

Results

We collected drug-drug target interactions, molecular pathways, and disease genes with their regulated states in the diseases. PDOD is applied to 898 drugs with known drug-drug target interactions and nine diseases. We compared performance of PDOD for predicting known therapeutic drug-disease associations with the previous methods. PDOD outperformed other previous approaches which do not exploit directional information in molecular network. In addition, we provide a simple web service that researchers can submit genes of interest with their altered states and will obtain drugs seeming to have opposite effects on altered states of input genes at http://gto.kaist.ac.kr/pdod/index.php/main.

Conclusions

Our results showed that ‘effect type’ and ‘effect direction’ information in the network based approaches can be utilized to identify drugs having opposite effects on diseases. Our study can offer a novel insight into the field of network-based drug repositioning.

     

生物信息技术加速开发旧药新用途

传统的技术路线研发新药,不仅周期很长而且耗资巨大,开发已获批准药物新的治疗用途,又称为药物重定位,比传统的新药研发具有明显的优势．基于芯片的基因表达谱分析,已常规地广泛用于各种人类疾病的临床研究,提供了在全基因组水平描述疾病状态的特征信号．同时,基因芯片也广泛地用于对比药物处理前后细胞基因表达模式的变化,这也提供了反映药物效应的高质量信号．最近出版的Science Translational Medicine杂志同时发表了一个研究组的两篇论文,为我们展示了如何利用生物信息学手段重新解析和比较全基因组基因表达谱数据,以高效地预测药物的新用途．这两篇论文使用了公共数据库中的100种疾病基因表达谱数据,以及164种药物处理前后细胞基因表达谱数据,通过比较和配对疾病与药物基因表达谱,得到了一些可以逆转疾病异常表达基因的药物,其中证实了一些已知的药物-疾病组合,也预测了一些新的药物-疾病组合．最后通过实验验证了抗溃疡药可用于治疗肺癌,而抗癫痫药可治疗炎症性肠道疾病,进一步证实了他们所采用研究策略的正确性．于是,肺癌和炎性肠道疾病这两种临床上难治的疾病有了新的候选治疗药物,我们也有了一种挖掘已有数据快速发现药物新用途的思路和方法．     

药物重定位的计算分析方法

全新结构药物的研发存在周期长、耗资大、风险高的问题.通过各种技术预测已有药物的新适应症,即药物重定位,可以缩短药物研发时间、降低研发成本和风险.由于疾病种类和已知药物的数量繁多,完全通过实验筛选已知药物的新用途仍然具有很高的成本.随着组学和药物信息学数据的积累,药物重定位进入到了理性设计和实验筛选相结合的阶段,药物重定位的计算预测已经成为计算生物学和系统生物学的重要研究方向.本文将目前药物重定位计算分析的策略归纳为药物-靶标关系分析、药物-药物关系分析和药物-疾病关系分析,对已报道的技术方法及其成功应用实例进行了综述.     

Inferring drug-disease associations from integration of chemical,genomic and phenotype data using network propagation

Background

During the last few years, the knowledge of drug, disease phenotype and protein has been rapidly accumulated and more and more scientists have been drawn the attention to inferring drug-disease associations by computational method. Development of an integrated approach for systematic discovering drug-disease associations by those informational data is an important issue.

Methods

We combine three different networks of drug, genomic and disease phenotype and assign the weights to the edges from available experimental data and knowledge. Given a specific disease, we use our network propagation approach to infer the drug-disease associations.

Results

We apply prostate cancer and colorectal cancer as our test data. We use the manually curated drug-disease associations from comparative toxicogenomics database to be our benchmark. The ranked results show that our proposed method obtains higher specificity and sensitivity and clearly outperforms previous methods. Our result also show that our method with off-targets information gets higher performance than that with only primary drug targets in both test data.

Conclusions

We clearly demonstrate the feasibility and benefits of using network-based analyses of chemical, genomic and phenotype data to reveal drug-disease associations. The potential associations inferred by our method provide new perspectives for toxicogenomics and drug reposition evaluation.

     

Predicting human disease-associated circRNAs based on locality-constrained linear coding

Circular RNAs (circRNAs) are a new kind of endogenous non-coding RNAs, which have been discovered continuously. More and more studies have shown that circRNAs are related to the occurrence and development of human diseases. Identification of circRNAs associated with diseases can contribute to understand the pathogenesis, diagnosis and treatment of diseases. However, experimental methods of circRNA prediction remain expensive and time-consuming. Therefore, it is urgent to propose novel computational methods for the prediction of circRNA-disease associations. In this study, we develop a computational method called LLCDC that integrates the known circRNA-disease associations, circRNA semantic similarity network, disease semantic similarity network, reconstructed circRNA similarity network, and reconstructed disease similarity network to predict circRNAs related to human diseases. Specifically, the reconstructed similarity networks are obtained by using Locality-Constrained Linear Coding (LLC) on the known association matrix, cosine similarities of circRNAs and diseases. Then, the label propagation method is applied to the similarity networks, and four relevant score matrices are respectively obtained. Finally, we use 5-fold cross validation (5-fold CV) to evaluate the performance of LLCDC, and the AUC value of the method is 0.9177, indicating that our method performs better than the other three methods. In addition, case studies on gastric cancer, breast cancer and papillary thyroid carcinoma further verify the reliability of our method in predicting disease-associated circRNAs.     

Systematic drug repositioning based on clinical side-effects

Drug repositioning helps fully explore indications for marketed drugs and clinical candidates. Here we show that the clinical side-effects (SEs) provide a human phenotypic profile for the drug, and this profile can suggest additional disease indications. We extracted 3,175 SE-disease relationships by combining the SE-drug relationships from drug labels and the drug-disease relationships from PharmGKB. Many relationships provide explicit repositioning hypotheses, such as drugs causing hypoglycemia are potential candidates for diabetes. We built Naïve Bayes models to predict indications for 145 diseases using the SEs as features. The AUC was above 0.8 in 92% of these models. The method was extended to predict indications for clinical compounds, 36% of the models achieved AUC above 0.7. This suggests that closer attention should be paid to the SEs observed in trials not just to evaluate the harmful effects, but also to rationally explore the repositioning potential based on this “clinical phenotypic assay”.     

Kinase Identification with Supervised Laplacian Regularized Least Squares

Phosphorylation is catalyzed by protein kinases and is irreplaceable in regulating biological processes. Identification of phosphorylation sites with their corresponding kinases contributes to the understanding of molecular mechanisms. Mass spectrometry analysis of phosphor-proteomes generates a large number of phosphorylated sites. However, experimental methods are costly and time-consuming, and most phosphorylation sites determined by experimental methods lack kinase information. Therefore, computational methods are urgently needed to address the kinase identification problem. To this end, we propose a new kernel-based machine learning method called Supervised Laplacian Regularized Least Squares (SLapRLS), which adopts a new method to construct kernels based on the similarity matrix and minimizes both structure risk and overall inconsistency between labels and similarities. The results predicted using both Phospho.ELM and an additional independent test dataset indicate that SLapRLS can more effectively identify kinases compared to other existing algorithms.     

Large-scale elucidation of drug response pathways in humans

Elucidating signaling pathways is a fundamental step in understanding cellular processes and developing new therapeutic strategies. Here we introduce a method for the large-scale elucidation of signaling pathways involved in cellular response to drugs. Combining drug targets, drug response expression profiles, and the human physical interaction network, we infer 99 human drug response pathways and study their properties. Based on the newly inferred pathways, we develop a pathway-based drug-drug similarity measure and compare it to two common, gold standard drug-drug similarity measures. Remarkably, our measure provides better correspondence to these gold standards than similarity measures that are based on associations between drugs and known pathways, or on drug-specific gene expression profiles. It further improves the prediction of drug side effects and indications, elucidating specific response pathways that may be associated with these drug properties. Supplementary Material for this article is available at www.liebertonline.com/cmb.     

DLGCN:基于图卷积网络的药物-lncRNA[]关联预测

为实现高通量识别新的药物-长链非编码RNA(Long non-coding RNA, lncRNA)关联,本文提出了一种基于图卷积网络模型来识别潜在药物-lncRNA关联的方法DLGCN(Drug-LncRNA graph convolution network)。首先,基于药物的结构信息和lncRNA的序列信息分别构建了药物-药物和lncRNA-lncRNA相似性网络,并整合实验证实的药物-lncRNA关联构建了药物-lncRNA异质性网络。然后,将注意力机制和图卷积运算应用于该网络中,学习药物和lncRNA的低维特征,基于整合的低维特征预测新的药物-lncRNA关联。通过效能评估,DLGCN的受试者工作特性曲线下面积(Area under receiver operating characteristic, AUROC)达到0.843 1,优于经典的机器学习方法和常见的深度学习方法。此外,DLGCN预测到姜黄素能够调控lncRNA MALAT1的表达,已被最近的研究证实。DLGCN能够有效预测药物-lncRNA关联,为肿瘤治疗新靶点的识别和抗癌药物的筛选提供了重要参考。     

In silico screening of drug databases for TSE inhibitors

Today, thousands of different chemical compounds are used as drugs for a wealth of different indications. Here, we demonstrate the use of a novel conformational drug database for the search of compounds with a positive influence on Transmissible Spongiform Encephalopathies (TSEs) by using two- and three-dimensional structural similarity to compounds with known effect. Both methods are combined to deduce a list of 16 candidate drugs. The proposal of a small number of putative inhibitors out of about 2300 approved essential drugs allows testing by expensive or time-consuming methods with the advantage that all agents are well known and suitable for use in humans.