MADGAN:A microbe-disease association prediction model based on generative adversarial networks

Researches have demonstrated that microorganisms are indispensable for the nutrition transportation, growth and development of human bodies, and disorder and imbalance of microbiota may lead to the occurrence of diseases. Therefore, it is crucial to study relationships between microbes and diseases. In this manuscript, we proposed a novel prediction model named MADGAN to infer potential microbe-disease associations by combining biological information of microbes and diseases with the generative adversarial networks. To our knowledge, it is the first attempt to use the generative adversarial network to complete this important task. In MADGAN, we firstly constructed different features for microbes and diseases based on multiple similarity metrics. And then, we further adopted graph convolution neural network (GCN) to derive different features for microbes and diseases automatically. Finally, we trained MADGAN to identify latent microbe-disease associations by games between the generation network and the decision network. Especially, in order to prevent over-smoothing during the model training process, we introduced the cross-level weight distribution structure to enhance the depth of the network based on the idea of residual network. Moreover, in order to validate the performance of MADGAN, we conducted comprehensive experiments and case studies based on databases of HMDAD and Disbiome respectively, and experimental results demonstrated that MADGAN not only achieved satisfactory prediction performances, but also outperformed existing state-of-the-art prediction models.

PharmKG: a dedicated knowledge graph benchmark for bomedical data mining

Biomedical knowledge graphs (KGs), which can help with the understanding of complex biological systems and pathologies, have begun to play a critical role in medical practice and research. However, challenges remain in their embedding and use due to their complex nature and the specific demands of their construction. Existing studies often suffer from problems such as sparse and noisy datasets, insufficient modeling methods and non-uniform evaluation metrics. In this work, we established a comprehensive KG system for the biomedical field in an attempt to bridge the gap. Here, we introduced PharmKG, a multi-relational, attributed biomedical KG, composed of more than 500 000 individual interconnections between genes, drugs and diseases, with 29 relation types over a vocabulary of ~8000 disambiguated entities. Each entity in PharmKG is attached with heterogeneous, domain-specific information obtained from multi-omics data, i.e. gene expression, chemical structure and disease word embedding, while preserving the semantic and biomedical features. For baselines, we offered nine state-of-the-art KG embedding (KGE) approaches and a new biological, intuitive, graph neural network-based KGE method that uses a combination of both global network structure and heterogeneous domain features. Based on the proposed benchmark, we conducted extensive experiments to assess these KGE models using multiple evaluation metrics. Finally, we discussed our observations across various downstream biological tasks and provide insights and guidelines for how to use a KG in biomedicine. We hope that the unprecedented quality and diversity of PharmKG will lead to advances in biomedical KG construction, embedding and application.

Prediction of Potential Associations Between miRNAs and Diseases Based on Matrix Decomposition

It is known that miRNA plays an increasingly important role in many physiological processes. Disease-related miRNAs could be potential biomarkers for clinical diagnosis, prognosis, and treatment. Therefore, accurately inferring potential miRNAs related to diseases has become a hot topic in the bioinformatics community recently. In this study, we proposed a mathematical model based on matrix decomposition, named MFMDA, to identify potential miRNA-disease associations by integrating known miRNA and disease-related data, similarities between miRNAs and between diseases. We also compared MFMDA with some of the latest algorithms in several established miRNA disease databases. MFMDA reached an AUC of 0.9061 in the fivefold cross-validation. The experimental results show that MFMDA effectively infers novel miRNA-disease associations. In addition, we conducted case studies by applying MFMDA to three types of high-risk human cancers. While most predicted miRNAs are confirmed by external databases of experimental literature, we also identified a few novel disease-related miRNAs for further experimental validation.

MSFSP: A Novel miRNA-Disease Association Prediction Model by Federating Multiple-Similarities Fusion and Space Projection

Growing evidences have indicated that microRNAs (miRNAs) play a significant role relating to many important bioprocesses; their mutations and disorders will cause the occurrence of various complex diseases. The prediction of miRNAs associated with underlying diseases via computational approaches is beneficial to identify biomarkers and discover specific medicine, which can greatly reduce the cost of diagnosis, cure, prognosis, and prevention of human diseases. However, how to further achieve a more reliable prediction of potential miRNA-disease associations with effective integration of different biological data is a challenge for researchers. In this study, we proposed a computational model by using a federated method of combined multiple-similarities fusion and space projection (MSFSP). MSFSP firstly fused the integrated disease similarity (composed of disease semantic similarity, disease functional similarity, and disease Hamming similarity) with the integrated miRNA similarity (composed of miRNA functional similarity, miRNA sequence similarity, and miRNA Hamming similarity). Secondly, it constructed the weighted network of miRNA-disease associations from the experimentally verified Boolean network of miRNA-disease associations by using similarity networks. Finally, it calculated the prediction results by weighting miRNA space projection scores and the disease space projection scores. Leave-one-out cross-validation demonstrated that MSFSP has the distinguished predictive accuracy with area under the receiver operating characteristics curve (AUC) of 0.9613 better than that of five other existing models. In case studies, the predictive ability of MSFSP was further confirmed as 96 and 98% of the top 50 predictions for prostatic neoplasms and lung neoplasms were successfully validated by experimental evidences and supporting experimental evidences were also found for 100% of the top 50 predictions for isolated diseases.

MeSHHeading2vec: a new method for representing MeSH headings as vectors based on graph embedding algorithm

Effectively representing Medical Subject Headings (MeSH) headings (terms) such as disease and drug as discriminative vectors could greatly improve the performance of downstream computational prediction models. However, these terms are often abstract and difficult to quantify. In this paper, we converted the MeSH tree structure into a relationship network and applied several graph embedding algorithms on it to represent these terms. Specifically, the relationship network consisting of nodes (MeSH headings) and edges (relationships), which can be constructed by the tree num. Then, five graph embedding algorithms including DeepWalk, LINE, SDNE, LAP and HOPE were implemented on the relationship network to represent MeSH headings as vectors. In order to evaluate the performance of the proposed methods, we carried out the node classification and relationship prediction tasks. The results show that the MeSH headings characterized by graph embedding algorithms can not only be treated as an independent carrier for representation, but also can be utilized as additional information to enhance the representation ability of vectors. Thus, it can serve as an input and continue to play a significant role in any computational models related to disease, drug, microbe, etc. Besides, our method holds great hope to inspire relevant researchers to study the representation of terms in this network perspective.

Corrigendum: IDLDA: An Improved Diffusion Model for Predicting LncRNA-Disease Associations

[This corrects the article DOI: 10.3389/fgene.2019.01259.].

LDAI-ISPS: LncRNA-Disease Associations Inference Based on Integrated Space Projection Scores

Long non-coding RNAs (long ncRNAs, lncRNAs) of all kinds have been implicated in a range of cell developmental processes and diseases, while they are not translated into proteins. Inferring diseases associated lncRNAs by computational methods can be helpful to understand the pathogenesis of diseases, but those current computational methods still have not achieved remarkable predictive performance: such as the inaccurate construction of similarity networks and inadequate numbers of known lncRNA-disease associations. In this research, we proposed a lncRNA-disease associations inference based on integrated space projection scores (LDAI-ISPS) composed of the following key steps: changing the Boolean network of known lncRNA-disease associations into the weighted networks via combining all the global information (e.g., disease semantic similarities, lncRNA functional similarities, and known lncRNA-disease associations); obtaining the space projection scores via vector projections of the weighted networks to form the final prediction scores without biases. The leave-one-out cross validation (LOOCV) results showed that, compared with other methods, LDAI-ISPS had a higher accuracy with area-under-the-curve (AUC) value of 0.9154 for inferring diseases, with AUC value of 0.8865 for inferring new lncRNAs (whose associations related to diseases are unknown), with AUC value of 0.7518 for inferring isolated diseases (whose associations related to lncRNAs are unknown). A case study also confirmed the predictive performance of LDAI-ISPS as a helper for traditional biological experiments in inferring the potential LncRNA-disease associations and isolated diseases.

IDLDA: An Improved Diffusion Model for Predicting LncRNA-Disease Associations

It has been demonstrated that long non-coding RNAs (lncRNAs) play important roles in a variety of biological processes associated with human diseases. However, the identification of lncRNA–disease associations by experimental methods is time-consuming and labor-intensive. Computational methods provide an effective strategy to predict more potential lncRNA–disease associations to some degree. Based on the hypothesis that phenotypically similar diseases are often associated with functionally similar lncRNAs and vice versa, we developed an improved diffusion model to predict potential lncRNA–disease associations (IDLDA). As a result, our model performed well in the global and local cross-validations, which indicated that IDLDA had a great performance in predicting novel associations. Case studies of colon cancer, breast cancer, and gastric cancer were also implemented, all lncRNAs which ranked top 10 in both databases were verified by databases and related literature. The results showed that IDLDA might play a key role in biomedical research.

Predicting lncRNA-miRNA Interaction via Graph Convolution Auto-Encoder

The interaction of miRNA and lncRNA is known to be important for gene regulations. However, the number of known lncRNA-miRNA interactions is still very limited and there are limited computational tools available for predicting new ones. Considering that lncRNAs and miRNAs share internal patterns in the partnership between each other, the underlying lncRNA-miRNA interactions could be predicted by utilizing the known ones, which could be considered as a semi-supervised learning problem. It is shown that the attributes of lncRNA and miRNA have a close relationship with the interaction between each other. Effective use of side information could be helpful for improving the performance especially when the training samples are limited. In view of this, we proposed an end-to-end prediction model called GCLMI (Graph Convolution for novel lncRNA-miRNA Interactions) by combining the techniques of graph convolution and auto-encoder. Without any preprocessing process on the feature information, our method can incorporate raw data of node attributes with the topology of the interaction network. Based on a real dataset collected from a public database, the results of experiments conducted on k-fold cross validations illustrate the robustness and effectiveness of the prediction performance of the proposed prediction model. We prove the graph convolution layer as designed in the proposed model able to effectively integrate the input data by filtering the graph with node features. The proposed model is anticipated to yield highly potential lncRNA-miRNA interactions in the scenario that different types of numerical features describing lncRNA or miRNA are provided by users, serving as a useful computational tool.

Bipartite Heterogeneous Network Method Based on Co-neighbor for MiRNA-Disease Association Prediction

In recent years, miRNA variation and dysregulation have been found to be closely related to human tumors, and identifying miRNA-disease associations is helpful for understanding the mechanisms of disease or tumor development and is greatly significant for the prognosis, diagnosis, and treatment of human diseases. This article proposes a Bipartite Heterogeneous network link prediction method based on co-neighbor to predict miRNA-disease association (BHCN). According to the structural characteristics of the bipartite network, the concept of bipartite network co-neighbors is proposed, and the co-neighbors were used to represent the probability of association between disease and miRNA. To predict the isolated diseases and the new miRNA based on the association probability expressed by co-neighbors, we utilized the similarity between disease nodes and the similarity between miRNA nodes in heterogeneous networks to represent the association probability between disease and miRNA. The model's predictive performance was evaluated by the leave-one-out cross validation (LOOCV) on different datasets. The AUC value of BHCN on the gold benchmark dataset was 0.7973, and the AUC obtained on the prediction dataset was 0.9349, which was better than that of the classic global algorithm. In this case study, we conducted predictive studies on breast neoplasms and colon neoplasms. Most of the top 50 predicted results were confirmed by three databases, namely, HMDD, miR2disease, and dbDEMC, with accuracy rates of 96 and 82%. In addition, BHCN can be used for predicting isolated diseases (without any known associated diseases) and new miRNAs (without any known associated miRNAs). In the isolated disease case study, the top 50 of breast neoplasm and colon neoplasm potentials associated with miRNAs predicted an accuracy of 100 and 96%, respectively, thereby demonstrating the favorable predictive power of BHCN for potentially relevant miRNAs.

A Novel Approach Based on a Weighted Interactive Network to Predict Associations of MiRNAs and Diseases

Accumulating evidence progressively indicated that microRNAs (miRNAs) play a significant role in the pathogenesis of diseases through many experimental studies; therefore, developing powerful computational models to identify potential human miRNA–disease associations is vital for an understanding of the disease etiology and pathogenesis. In this paper, a weighted interactive network was firstly constructed by combining known miRNA–disease associations, as well as the integrated similarity between diseases and the integrated similarity between miRNAs. Then, a new computational method implementing the newly weighted interactive network was developed for discovering potential miRNA–disease associations (WINMDA) by integrating the T most similar neighbors and the shortest path algorithm. Simulation results show that WINMDA can achieve reliable area under the receiver operating characteristics (ROC) curve (AUC) results of 0.9183 ± 0.0007 in 5-fold cross-validation, 0.9200 ± 0.0004 in 10-fold cross-validation, 0.9243 in global leave-one-out cross-validation (LOOCV), and 0.8856 in local LOOCV. Furthermore, case studies of colon neoplasms, gastric neoplasms, and prostate neoplasms based on the Human microRNA Disease Database (HMDD) database were implemented, for which 94% (colon neoplasms), 96% (gastric neoplasms), and 96% (prostate neoplasms) of the top 50 predicting miRNAs were confirmed by recent experimental reports, which also demonstrates that WINMDA can effectively uncover potential miRNA–disease associations.

TLHNMDA: Triple Layer Heterogeneous Network Based Inference for MiRNA-Disease Association Prediction

In recent years, microRNAs (miRNAs) have been confirmed to be involved in many important biological processes and associated with various kinds of human complex diseases. Therefore, predicting potential associations between miRNAs and diseases with the huge number of verified heterogeneous biological datasets will provide a new perspective for disease therapy. In this article, we developed a novel computational model of Triple Layer Heterogeneous Network based inference for MiRNA-Disease Association prediction (TLHNMDA) by using the experimentally verified miRNA-disease associations, miRNA-long noncoding RNA (lncRNA) interactions, miRNA function similarity information, disease semantic similarity information and Gaussian interaction profile kernel similarity for lncRNAs into an triple layer heterogeneous network to predict new miRNA-disease associations. As a result, the AUCs of TLHNMDA are 0.8795 and 0.8795 ± 0.0010 based on leave-one-out cross validation (LOOCV) and 5-fold cross validation, respectively. Furthermore, TLHNMDA was implemented on three complex human diseases to evaluate predictive ability. As a result, 84% (kidney neoplasms), 78% (lymphoma) and 76% (prostate neoplasms) of top 50 predicted miRNAs for the three complex diseases can be verified by biological experiments. In addition, based on the HMDD v1.0 database, 98% of top 50 potential esophageal neoplasms-associated miRNAs were confirmed by experimental reports. It is expected that TLHNMDA could be a useful model to predict potential miRNA-disease associations with high prediction accuracy and stability.

PBHMDA: Path-Based Human Microbe-Disease Association Prediction

With the advance of sequencing technology and microbiology, the microorganisms have been found to be closely related to various important human diseases. The increasing identification of human microbe-disease associations offers important insights into the underlying disease mechanism understanding from the perspective of human microbes, which are greatly helpful for investigating pathogenesis, promoting early diagnosis and improving precision medicine. However, the current knowledge in this domain is still limited and far from complete. Here, we present the computational model of Path-Based Human Microbe-Disease Association prediction (PBHMDA) based on the integration of known microbe-disease associations and the Gaussian interaction profile kernel similarity for microbes and diseases. A special depth-first search algorithm was implemented to traverse all possible paths between microbes and diseases for inferring the most possible disease-related microbes. As a result, PBHMDA obtained a reliable prediction performance with AUCs (The area under ROC curve) of 0.9169 and 0.8767 in the frameworks of both global and local leave-one-out cross validations, respectively. Based on 5-fold cross validation, average AUCs of 0.9082 ± 0.0061 further demonstrated the efficiency of the proposed model. For the case studies of liver cirrhosis, type 1 diabetes, and asthma, 9, 7, and 9 out of predicted microbes in the top 10 have been confirmed by previously published experimental literatures, respectively. We have publicly released the prioritized microbe-disease associations, which may help to select the most potential pairs for further guiding the experimental confirmation. In conclusion, PBHMDA may have potential to boost the discovery of novel microbe-disease associations and aid future research efforts toward microbe involvement in human disease mechanism. The code and data of PBHMDA is freely available at http://www.escience.cn/system/file?fileId=85214.