Modelling Highly Symmetrical Molecules: Linking Ontologies and Graphs

Oliver Kutz, Janna Hastings, Till Mossakowski

In: 15th International Conference on Artificial Intelligence: Methodology, Systems, and Applications. International Conference on Artificial Intelligence: Methodology, Systems, and Applications (AIMSA-2012) 15th September 13-15 Varna Bulgaria Seiten 103-111 Lecture Notes in Artificial Intelligence (LNAI) 7557 Springer 2012.


With dramatically increasing volumes of chemical data becoming available in recent years, finding automated means to manage the deluge becomes ever more urgent. Traditional automated classification of chemical entities depends on identifying interesting parts and properties of the molecules. However, classes of chemical entities which are highly symmetrical and which contain large numbers of homogeneous parts (such as carbon atoms) are not straightforwardly classified in this fashion. One such class of molecules is the recently developed fullerene family, discovery of which led to the award of the Nobel prize for chemistry in 1996. Fullerene molecules show potential for many novel applications including in biomedicine. Standard cheminformatics approaches to classification do not easily allow the automatic classification of chemical entities as members of the fullerene class based on their chemical structure, and neither does OWL, due to the tree model requirement. Furthermore, while individual members of the fullerene class can be modelled in standard FOL, expressing the properties of the class as a whole (independent of the count of atoms of the members) requires second-order quantification over the set of atoms in a molecule. Yet, given the size of chemical ontologies such as ChEBI, using second-order expressivity in the general case is prohibitively expensive to practical applications. To address these conflicting requirements, we introduce a novel framework in which we heterogeneously integrate standard ontological modelling with monadic second-order reasoning over chemical graphs, enabling various kinds of information flow between these distinct representational layers.

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence