Towards Data Integration for Computational Chemistry

Increasingly, understanding complex problems in computational chemistry requires the use of several applications as components of computational workflows. Unfortunately, this process is severely hindered by a lack of data interchange standards. The eCCP and eMinerals UK eScience projects are developing a framework to facilitate the interoperability of such applications. Data sets are serialised in a way that conforms to existing XML languages, including the Chemical Markup Language, and collaboratively designed extensions. Various methods are used to express relationships between data sets, based on the XLink standard. Explicit semantics are associated to data sets by mappings to concepts specified in OWL ontologies. These mappings make use of RDF and XPointer standards and enable the users of framework data to work with its logical structure. This provides a number of advantages, including the ability to work with data that conforms to differing data models. Framework tools are under development that will enable the manipulation of such data, being designed with an emphasis on easy integration into the current working practices of computational communities, such as the UK’s Collaborative Computational Projects.