RESTful web services are at the cutting edge of contemporary web technology and have proven to boost flexibility and modularity, promote loosely coupled development and implementation independence thus are the perfect tool for large-scale projects. Moreover, REST-based services are suitable for distributed computing applications. The involvement of our team in the developers group of OpenTox (An FP7-funded EU research project) has offered us great experience as far as REST services are concerned. Popular machine learning and artificial intelligence algorithms such as MLR and SVM were developed in our Laboratory and are deployed as REST web services. These services are used by the applications ToxCreate and ToxPredict for the purposes of predictive toxicology.
The JAQPOT3 web services are OpenTox API-1.2 compliant web services. JAQPOT3 is a web application that supports model training and data preprocessing algorithms such as multiple linear regression, support vector machines, neural networks (an in-house implementation based on an efficient algorithm), an implementation of the leverage algorithm for domain of applicability estimation and various data preprocessing algorithms like PLS and scaling.
ToxOtis is a Java interface to the predictive toxicology services of OpenTox. ToxOtis is being developed to help both those who need a painless way to consume OpenTox web services and for ambitious service providers that don't want to spend half of their time in RDF parsing and creation, database management and security measures.
In order to facilitate considerations on the adequacy of the prediction (model result) in relation to a defined regulatory purpose, JRC has compiled a standard for reporting (Q)SAR predictions for chemical compounds. The QSAR Prediction Reporting Format (QPRF) is a harmonized template for summarizing and reporting substance-specific predictions generated by (Q)SAR models.