- WP8 - Demo of systems/physiology model metadata standards

Activity type: RTD

Principal Partner(s):

Objectives

This WP aims to establish the methodology to annotate the physics- and multiscale anatomy-related concepts in model  variables/parameters, using CORDO dictionaries. This use-case effort will focus on the annotation, and metadata representation in at least one SBML and one CellML model.

It is also a goal of this WP to interact closely with Work packages 3, 4, and 5 to use the experience of this WP in developing a composite annotation architecture and VPH annotation repository that capable of annotating, storing and querying overlaps between biosimulation models.

Description

The systematic and consistent metadata representation of model variables/parameters with CORDO IDs is the central focus of this WP, and is also fundamentalto VPH ToolKit development. The ability to support such a coherent level of curation across the whole spectrum of the VPH is a key requisite to community-wide model interoperability. This pilot WP will make use of mutiscale structural terms, as well as concepts from physics, to generate standard patterns of descriptors. This work will be carried out in the context of the design plans emerging from WP2-5. This WP will focus on metadata representation of two models (one in SBML and one in CellML) to research variable/parameter decomposition and annotation, as well as the communal strategy to encode such information into the model's metadata.

Model interoperability may stem from more subtle structural relationships, other than outright identity, that occur between process participants. The VPH's aim to achieve 'vertical integration' of physiological models that interoperate across dimensional scales, for instance, requires methods that detect relationships between biological entities of significantly different structural granularity. Mapping model variables/parameters to structural terms that belong to a well-developed ontology would allow the detection of model overlap based on reasoning operations over the relationship of structures (e.g. if a parameter in model X maps to the "distal convoluted tubule", and model Y refers to the "glomerulus", then both models are compatible with model Z that maps to the "nephron" concept ? the relationship between these three anatomical concepts would be available through the FMA partonomy).

Other indicators of functional overlap may be derived from physical properties described in model parameters such as 'pressure', 'resistance' and 'compliance'. Such parameters can be mapped to physical concepts using terms from ontologies such as PATO, the OPB or SBO.

This WP therefore seeks work with WPs 3-5 to establish community-wide annotation methods and standards through the metadata representation over two systems-level models one encoded in SBML and one in CellML