These notes are relevant to the GitHub repository for the profile: https://github.com/eScienceLab/bge-ro-crate-profile.
To run the scripts you must install the requirements listed in requirements.txt.
The scripts use roc-validator to automatically run validation against the specification when generating RO-Crates. This should tell you if you did anything wrong according to the base spec (but does not mean that everything is definitely correct - e.g. it is possible to link things incorrectly and still pass validation).
The make_crate_genome.py and make_crate_barcode.py scripts create example crates that represent the profile.
Currently the scripts are not in sync with the profile text - as of v0.2 the scripts have been advanced further than the profile text.
The scripts generate RO-Crates in the bge-crate-genome and bge-crate-barcode folders respectively. To generate a HTML preview of the crate (useful for checking things are linked as intended):
npm install ro-crate-html
rochtml bge-crate-genome/ro-crate-metadata.json
The make_profile_crate.py script builds a Profile Crate (an RO-Crate that represents the profile itself) using the ROCratePCC package. The resulting crate should be suitable for submission to the RO-Crate Profile Portal.
If there are changes to the context.json or the URIs for the profile/examples, the script should be updated accordingly. When releasing a new profile version, ensure all the metadata in the script (authors etc) is up to date.
- People: https://orcid.org/
- Organizations: https://ror.org/
- Places: https://www.geonames.org/v3/
- Taxonomy: https://www.ncbi.nlm.nih.gov/taxonomy
- BioProjects: https://www.ncbi.nlm.nih.gov/search/ (though at time of writing, search results may give 500 errors)
- Protocols: https://protocols.io/
- Samples (example accession SAMEA114402090):
- Sequencing experiments (example accession ERX12519568):
- Assembled genome (example accession GCA_964187845.1):