FAIR4RS & Robustness Assessment

A review of how well biobb_md_workflows meets FAIR4RS and data-pipeline-engineering principles, plus a roadmap to close the gaps.

Where the project stands today

Legend: 🟢 settled · 🟡 partial · 🔴 missing

FAIR4RS

Principle

Reality in the repo

Status

Findable

Public GitHub repo, git tags (0.0.1–0.1.0), GitHub Pages docs, Colab tutorial. __version__ now exposed (via importlib.metadata) and logged at the start of every run. CITATION.cff added. Zenodo DOI planned.

🟡

Accessible

Repo + tags + LICENSE present. License is CC BY-NC-SA — academic-open / commercially-restricted choice.

🟢

Interoperable

Standard formats in/out (PDB/GRO/TOP/XTC), YAML config, thin argparse CLIs, importable workflow functions.

🟢

Reusable

--help + docs exist; reproducibility in place (pinned deps + tagged forks).

🟢

Robustness

Principle

Reality in the repo

Status

Reproducibility

All forks are pinned to fork-namespaced tags; other dependencies are version-pinned.

🟢

Provenance

configuration and absolute input paths are written per run; log.out and per-step logs exist. Package version is now logged in the log header. Full CLI command, and input-file checksums are still not recorded (full run_manifest.json remains future work).

🟡

Environment portability

Conda env files exist; deps + forks now pinned. No container yet.

🟡

Modularity

Genuinely modular at the BioBB-step level. create_config_file/config scaffolding is copy-pasted across all four workflows. Creation of common modules pending.

🟡

Validation

Step outputs are validated internally by the biobbs — each building block checks its own outputs. Remaining gaps are at the CLI boundary: no argparse choices for enums, no --ph bounds, and several bad-arg paths log an error then silently return. Only md_gromacs has a real check_inputs().

🟡

Roadmap

1. Improve clarity of errors

  • Surface a clearer workflow-level message for the different failure modes.

2. Provenance

  • One shared helper writing output/run_manifest.json — git commit (if resolvable), full sys.argv, SHA-256 of each input file, UTC timestamp, resolved config.yml path, and a conda env export / pip freeze snapshot.

3. Validation

  • argparse choices= for the format/output_format/protonation-tool enums; --ph bounds [0, 14]; convert “log-error-then-return” silent exits into raise.

  • Generalize the md_gromacs.check_inputs() pattern to the other three workflows.

4. Testing & automation

  • Minimal GitHub Actions CI: ruff lint + pip install .

  • One tiny end-to-end smoke test on a small system (reuse tests/*/input.yml)

  • Dependency automation (Dependabot/Renovate) watching the two forks.

5. FAIR4RS metadata & portability

  • Mint a Zenodo DOI for easy academic citation: (1) org admin authorizes Zenodo for the NBDsoftware org, (2) enable the repo in Zenodo, (3) cut a GitHub Release → DOI auto-minted, (4) paste the concept-DOI into CITATION.cff and a README badge.

  • CHANGELOG.md.

  • License: settled — Explore other options more appropiate for code.

  • Build an Apptainer/Singularity image per release for one-command portable execution.