Prediction of the Drug–Drug Interaction Types with the Unified Embedding Features from Drug Similarity Networks

Drug combination therapies are a promising strategy to overcome drug resistance and improve the efficacy of monotherapy in cancer, and it has been shown to lead to a decrease in dose-related toxicities. Except the synergistic reaction between drugs, some antagonistic drug–drug interactions (DDIs) exist, which is the main cause of adverse drug events. Precisely predicting the type of DDI is important for both drug development and more effective drug combination therapy applications. Recently, numerous text mining– and machine learning–based methods have been developed for predicting DDIs. All these methods implicitly utilize the feature of drugs from diverse drug-related properties. However, how to integrate these features more efficiently and improve the accuracy of classification is still a challenge. In this paper, we proposed a novel method (called NMDADNN) to predict the DDI types by integrating five drug-related heterogeneous information sources to extract the unified drug mapping features. NMDADNN first constructs the similarity networks by using the Jaccard coefficient and then implements random walk with restart algorithm and positive pointwise mutual information for extracting the topological similarities. After that, five network-based similarities are unified by using a multimodel deep autoencoder. Finally, NMDADNN implements the deep neural network (DNN) on the unified drug feature to infer the types of DDIs. In comparison with other recent state-of-the-art DNN-based methods, NMDADNN achieves the best results in terms of accuracy, area under the precision-recall curve, area under the ROC curve, F1 score, precision and recall. In addition, many of the promising types of drug–drug pairs predicted by NMDADNN are also confirmed by using the interactions checker tool. These results demonstrate the effectiveness of our NMDADNN method, indicating that NMDADNN has the great potential for predicting DDI types.

Semantic Interoperability for Logistics and Beyond

A Semantic Mediator was conceived in the CRC 637 (The Collaborative Research Centre 637 “Autonomous Cooperating Logistic Processes” focused on adaptive logistic processes including autonomous capabilities for the decentralised coordination of autonomous logistic objects in a heterarchical structure.) to tackle problems of interoperability of heterogeneous information sources in autonomous cooperating logistics processes. Since the conclusion of the CRC 637, the Semantic Mediator has been developed further and successfully transferred to interoperability problems in different domains, including Industry 4.0, the Internet of Things, and Product Lifecycle Management. This paper will introduce the Semantic Mediator and present examples of its successful application.

An Efficient and Adaptive Framework to Access Heterogeneous Health Information Sources

In health informatics, the need for a consistent and integrated view of distributed and heterogeneous information sources is inevitable. Healthcare, medical education and research could benefit of integrating medical information of patients or about a disease, a treatment or side effects of a drug. This article proposes a flexible incremental update method for the materialized part of the integration system. It permits us to manage the integration system according to the characteristics of the data sources, which can be changed. We present a hybrid data integration approach.In this approach the materialized part of the system in mediator is the object indexation structure based on an instance classification of the sources objects which correspond to the global schema. The object identifier of each object in the indexation structure is materialized together with the attributes which are needed for the incremental updating of this indexation (classifying attributes).

Automated modelling assistance by integrating heterogeneous information sources

Model-Driven Engineering (MDE) uses models as its main assets in the software development process. The structure of a model is described through a meta-model. Even though modelling and meta-modelling are recurrent activities in MDE and a vast amount of MDE tools exist nowadays, they are tasks typically performed in an unassisted way. Usually, these tools cannot extract useful knowledge available in heterogeneous information sources like XML, RDF, CSV or other models and meta-models.We propose an approach to provide modelling and meta-modelling assistance. The approach gathers heterogeneous information sources in various technological spaces, and represents them uniformly in a common data model. This enables their uniform querying, by means of an extensible mechanism, which can make use of services, e.g., for synonym search and word sense analysis. The query results can then be easily incorporated into the (meta-)model being built. The approach has been realized in the Extremo tool, developed as an Eclipse plugin. Extremo has been validated in the context of two domains – production systems and process modelling – taking into account a large and complex industrial standard for classification and product description. Further validation results indicate that the integration of Extremo in various modelling environments can be achieved with low effort, and that the tool is able to handle information from most existing technological spaces.

Extremo: An Eclipse plugin for modelling and meta-modelling assistance

Modelling is a core activity in software development paradigms like Model-driven Engineering (MDE). Therefore, the quality of (meta-)models is crucial for the success of software projects. However, many times, modelling becomes a purely manual activity, which does not take advantage of information embedded in heterogeneous information sources, such as XML documents, ontologies, or other models and meta-models. In order to improve this situation, we present Extremo, an Eclipse plugin aimed at gathering the information stored in heterogeneous sources in a common data model, to facilitate the reuse of information chunks in the model being built. The tool covers the steps needed to incorporate this knowledge within an external modelling tool, supporting the uniform query of the heterogeneous sources and the evaluation of constraints. Flexibility of the main features (e.g., supported data formats, queries) is achieved by means of extensible mechanisms. To illustrate the usefulness of Extremo, we describe a practical case study in the financial domain and evaluate its performance and scalability.