Lead: Antreas Afantitis / Deputy: Georgia Melagraki

Participating Partners: NovaMechanics

NM TA will support the application of cheminformatics methods such as quantitative structure activity relationship (QSAR) modelling and clustering to Users datasets in order to establish statistically significant relationships between measured biological activity profiles of engineered nanomaterials and their physical, chemical, and other properties, either measured experimentally or computed from the structure of the nanomaterials.

NM will offer support to Users in the design of their experiments to be directly compatible with the data needs for QSAR development, or work with Users that already have datasets, in order to computationally enrich the datasets, and then to build QSARs. Full computational time and power, technical expertise, bioinformatics support and analysis and paper writing support are included in the package of TA to be offered by NM in order to fully support Users nanoinformatics needs.

For the development of predictive models for Users based on their needs and datasets, several clustering and machine-learning techniques for data analysis will be employed in order to support the customised needs of Users, by building KNIME workflows. NM will use existing and develop new custom-made KNIME nodes, drawing on the user-friendly and comprehensive open-source data integration of KNIME to develop a customised processing, analysis, and exploration platform, in line with the current Enalos KNIME nodes (developed by NovaMechanics’ group).

The computation tools offered for TA fall under four categories:

• nanoWorkflow support to Users experimental design to ensure data suitability for QSAR development
• Data Processing and Analysis services including quantum mechanical descriptors and image analysis
• Predictive nanotoxicology tools, focussing on QSARs
• Data visualisation tools, including the in-silico toxicity platform that will be developed under JRA4