NFDI4ING Betty

Minutes and task from the meetings

2025-11-06

Divyansh ** adding software tool with version number (input files), ** discussion on howto reference the software tool with a URI ** add reference to the MathmodDB ** Documentation of howto add a benchmark
Philipp ** Julia implementation of plate with a hole (either nextflow, or snakemake, or both)
Bernd ** Integration of jupyter notebook with Markdown to couple documentation and execution of notebooks ** Benchmark implementation of rotating cylinders ** Documentation of howto add a benchmark (together with Divynash)
Mahdi ** Comparison of concepts in snakemake and nextflow provenance ** ROhub uploading/downloading ** provenance information in named graph only?
All ** Test concept (what should we test, on what level of detail, what is the requirement on external APIs (such as ROhub, jupyterhub) ** what is the interface to users (result users, tool developer, benchmark developer) - python package ** one tool/benchmark per ROCrate

datacite has an api from jupyter to download research object
versioning of data in a doi might allow to only use the latest results and invalidate old ones, but there is the alternative to use a handle rather than a permanent DOI (then datacite does not work, since it requires a DOI, a handle registration is then related to the actual repository used such as radar and darus)
additional metadata field could be introduced to define the maturity level (e.g. benchmarks creation/development updating-version final
options for the repository
- zenodo, option to register a doi on a git repo
- radar
- darus, python interfaces e.g. here
https://github.com/BioDT/pydidit
ProvStor For querying many RO-Crates
Loads RO-Crate JSON-LD into Fuseki Triple store (for SPARQL), and the ZIP of RO-Crate into MinIO/S3
https://elixir-europe.org/services/openebench may also be of interest for benchmarking
dashboard over RO-Crates using Sampo-UI

OpenFOAM Benchmark Case
SmartSim Orchestrator Software - similar to NextFlow?
nextflow workflow implemented
goal is to use the workflow ROCrate discussed here
meeting with MARDI, their plan is to describe the model (plate with a hole) in the MathModDB or MathAlgoDB such that we can reference that
another collection of workflow also based on the RO-Crate discussed here
„analyzer” for converting the log of different wf system (i.e. nextflow) to the standardized workflow format
implementation of snakemake plugin based on the metadata4ing ontology

next steps BAM