OpenMolGRIND: Molecular Science and Engineering in a Grid Context

Modern approaches to chemistry and pharmacology deal with large-scale molecular design problems. The molecular design is essentially based on data warehousing and data mining. Data warehousing techniques are needed to collect relevant data from distributed and heterogeneous databases. Data mining techniques are used to build predictive quantitative structure-property and activity relationship models. Increasingly, the computational resources (databases and analysis and modeling programs) needed for molecular engineering are geographically distributed. Grid technology offers a framework to build distributed molecular engineering systems. The OpenMolGRID project is focused on integrating existing software modules into a Grid infrastructure, and on making a solid foundation for next step molecular engineering tools. This involves the design of a seamless and unified user interface and the provision of adapters to make software Grid-aware. The system will be used to develop prototype applications for the generation of molecular structures with given chemical properties or biological activities. It will be intensively tested by academic and industrial users on real applications (multi-drug resistance, Gprotein-coupled receptors activity, and toxicity).