Supplementary MaterialsAdditional file 1 Appendix A. produced from triple components of various other annotations. Outcomes We present a job- and domain-independent ontological model for capturing annotations and their linkage with their Imatinib Mesylate cell signaling denoted understanding representations, which may be singular principles or even more complex pieces of assertions. We’ve applied this model as an expansion of the info Artifact Ontology in OWL and managed to get freely offered, and we present how it could be integrated with many prominent annotation and provenance versions. We present many app areas for the model, which range from linguistic annotation of textual content to the annotation of disease-associations in genome sequences. Conclusions With this model, progressively more technical annotations could be composed from various other annotations, and the provenance Imatinib Mesylate cell signaling of compositional annotations could be represented at the annotation level or at the amount of individual components of the RDF triples composing the annotations. Therefore permits progressively richer annotations to end up being made of previous annotation initiatives, the complete provenance recording of which facilitates evidence-based inference and error tracking. formats suited to one particular type of annotation or task but are not broadly applicable or interoperable. Several prominent annotation models not limited to specific types of tasks or information have been produced, and components that enable annotations to denote knowledge structures more complex than atomic concepts have been added very recently to these models [3,4]. Yet there have been no mechanisms put forth by which these more complex annotations can refer to other annotations and by which their provenance can be unambiguously recorded. There have also been prominent efforts in scientific workflow provenance [5,6]. That work, however, primarily focuses on annotating experimental data, typically annotating lists of identifiers or numeric data with their origins, not on annotating with dynamically composed and compositional knowledge structures. An effective annotation model, in addition to being applicable to many annotation use cases and supporting the specification of complex knowledge structures, needs to be able to unambiguously represent annotation provenance. While ontologies strive to be total, it is likely that specific applications will require dynamic construction of concepts, either through data-driven methods [7] or compositional concept formation [8]. To support and document the provenance of these more complex annotations, annotators (both human and computational) need the ability to refer to existing annotations as the basis of more complex annotations. For example, in the linguistic domain, an annotation representing part of a syntactic parse tree may wish to build upon existing token or part-of-speech annotations. Similarly, in the biomedical domain, a protein interaction event annotation may wish to leverage existing annotations identifying specific proteins. As annotation efforts become more ambitious, they will naturally build upon previous annotation efforts, and tracking the provenance of constructed knowledge representations being used for annotations at a fine-grained level will be important to facilitate inference and error analysis. This paper proposes a task- and domain-independent formal ontological model for the creation of annotations and their linkage to their denoted knowledge representations, which can be singular concepts or more complex knowledge in the form of units of RDF assertions. With this model, progressively more complex annotations can be composed from other annotations, and this provenance can be unambiguously represented at either a coarse- or fine-grained level. We have designed our annotation model to be generic in order to facilitate the concurrent usage of multiple types of annotations (the term apple denotes the particular apple or the even more general idea of an apple). We hold an annotation is certainly a kind of information content material entity, since it is for some reason about the entity it really is annotating. We Imatinib Mesylate cell signaling are involved in the ongoing procedure for submitting our model to the IAO for inclusion. An OWL representation of our model as an expansion of the IAO is certainly provided in Extra document 3. Namespace and notation Our in-house knowledge bottom of biomedicine (KaBOB) may be the aggregator of our function. KaBOB extensions of an ontology are Rabbit Polyclonal to Histone H3 called by prefixing the ontologys namespace with the letter k; the namespace will be utilized to recognize concepts. Class brands start out with a capital letter, while example and property brands begin.