The task of automatically extracting insights or building computational models from knowledge on complex systems greatly relies on the choice of appropriate representation. This work makes an effort towards building a framework suitable for representation of fragmented knowledge on complex systems and its semi-automated curation---continuous collation, integration, annotation and revision.
We propose a knowledge representation system based on hierarchies of graphs related with graph homomorphisms. Individual graphs situated in such hierarchies represent distinct fragments of knowledge and the homomorphisms allow relating these fragments. Their graphical structure can be used efficiently to express entities and their relations. We focus on the design of mathematical mechanisms, based on algebraic approaches to graph rewriting, for transformation of individual graphs in hierarchies that maintain consistent relations between them. Such mechanisms provide a transparent audit trail, as well as an infrastructure for maintaining multiple versions of knowledge.
We describe how the developed theory can be used for building schema-aware graph databases that provide schema-data co-evolution capabilities. The proposed knowledge representation framework is used to build the KAMI (Knowledge Aggregation and Model Instantiation) framework for curation of cellular signalling knowledge. The framework allows for semi-automated aggregation of individual facts on protein-protein interactions into knowledge corpora, reuse of this knowledge for instantiation of signalling models in different cellular contexts and generation of executable rule-based models.