Biomedical knowledge engineering tools based on experimental design: a case study based on neuroanatomical tract-tracing experiments

Curating information from the literature for storage in databases is a crucial task in biological research, and many groups assign a particular person or team to that process. We are developing a general-purpose approach to representing the design of a biomedical experiment and provide a manageable template for that experiment's data: "Knowledge Engineering from Experimental Design" (KEfED). The KEfED model allows us to impose a formal and well-grounded structure on the data contained in scientific articles, based on relationships between the dependent and independent variables that make up a scientific experiment. We use this structure to add value to the data contained in the articles by performing directed information retrieval, adding basic forms of reasoning using additional information such as anatomical atlases and taxonomies from external ontologies. We use a graphical interface for constructing KEfED models and a first-order logic reasoning system that performs inference over such models.