WP 4: Information system and e-infrastructures
Workpackage 4 ‘Information system and e-infrastructures’ prepares the organization and analysis of phenomic datasets in such a way that they follow the FAIR requirements (Findable, Accessible, Interoperable, Reusable). This will allow the Plant Science community to share, reuse and visualize plant phenomic datasets.
The first objective was to map the current situation, based on interviews with operators of local infrastructures. This revealed a diverse landscape for the naming of objects and measured variables and for the handling of datasets. We also mapped the situation for image analysis that revealed the same diversity, in particular for the packages that in many case cannot be reused by other scientists. However, approaches consisting of imposing standardized terms or solutions to a community have all failed in the past. EMPHASIS develops a strategy for understanding user’s needs, then formalize them for vocabulary and dataset management, then build solutions e.g. ontologies, formats, imaging methods and e-infrastructure, and finally test their usefulness in a broad community of phenomics.
The second objective is to ensure that data can be analyzed across experiments and phenotyping installations (‘Reusable’). We have developed methods that allow environmental data to be consistent across installations, with a common minimum list of variables characterized each by a definition, a method and a unit. Phenomic variables are tracked with the same triplet, and imaging methods are organized in workflows, in such a way that any future user can exactly know what was measured or how image analysis was performed. Data formats that can be read in 20 years time are encouraged.
The third objective is to design information systems and imaging pipelines able to be deployed in a diversity of phenomic installations, including controlled conditions and field (‘Accessible’). This requires a precise identification of objects, events and traits with uniform resource identifiers (URI, attached figure). We also design a suite of methods allowing information systems at different sites to be ‘Interoperable’, so a user can access datasets in different sites with a common query.
The information system is/will be interfaced with systems developed by the genomic and modelling communities for accessing genomic information and genotype-specific model parameters together with phenomic information (e.g. https://emphasis.plant-phenotyping.eu/Collaboration_ELIXIR)
Fig. WP4 Example of use of Uniform Resource Identifiers (URIs) for identifying unambiguously all objects present in single images taken in (a) greenhouse and (b) field experiment. For example, in the greenhouse, an image <m3p:arch/2017/ic17002295855> of a given plant <m3p:arch/2017/c17000118> that is placed in a pot
Work Package participants: INRA (lead), JUELICH, VIB, CNR, UNOTT, WR, UCL.
Contacts: Francois Tardieu