MetaPathways 3.5.2 release readiness

Release instructions · Tester checklist

Status updated 2026-10-02. This is a maintainer record of pre-release testing. Version remains 3.5.2. No release tag or registry publication is authorized by these checks.

Registry and container status

Public metadata checked on 2026-10-02: Anaconda lists 3.5.1 as latest; the latest GitHub release is v3.5.1-build1; Quay tags list 3.5.1 images. The 3.5.2 installation examples are release instructions, not evidence of publication. Until publication, testers install the reviewed checkout.

The Conda recipe includes pinned MAGSplitter and Camelot sources at build time; the Quay image uses that same Conda artifact. prepare_test copies bundled inputs and reference seeds into writable storage. Final Conda and Docker/Apptainer artifacts must pass release integration before publication; unit tests and documentation checks alone do not establish that validation.

Completed checks

  • 125 runtime/regression tests and 17 release-control tests pass. Tests cover independent per-database taxonomy, SwissProt taxon parsing, database order, report joins, abundance exports, workflow scheduling, checkpoint behavior, sequence staging and release safeguards.

  • The user completed the three-sample CAMI II test workflow (Meyer et al., 2022) locally with SwissProt and --skip_ptools: 51/51 successful tasks. Detailed result audit.

  • All CDS records, mapped gene/RNA abundance rows, contig measurements and nine genome bins were reconciled. Report SQLite integrity and foreign keys pass. Expected RNA warnings and unavailable PGDBs are explained in the audit.

  • CLI documentation and local documentation links pass validation.

  • The earlier 3.5.2 candidate was built and installed successfully as a Conda package. Artifacts are specific to their source commit; rebuild from the final feature commit to include subsequent taxonomy and explorer changes.

  • Source archives, installed native binaries and bundled three-sample fixture assets have been checked. Public packaging excludes licensed Pathway Tools installers, SIFs and MetaCyc databases.

  • Feature pushes run smoke checks. Publication helpers require reviewed dev/master; no feature-branch publication is permitted.

Before merge and public release

  1. Build/install the final committed candidate and record artifact checksums alongside its source commit.

  2. Complete the full local benchmark with the licensed Pathway Tools image and MPDB, retaining task traces and logs.

  3. Complete the HPC Slurm test and benchmark with shared-storage paths, recording requested resources separately from measured consumption.

  4. Have the tester review the feature PR before merging to dev; promote to production through a separate dev → master PR and validate the production commit.

  5. Run the release integration and optional container checks, verify registry credentials and metadata, and publish only after explicit release approval. Anaconda, Quay and Zenodo uploads are not part of the local test or feature push.

CAMI references

  • CAMI: Sczyrba, A., Hofmann, P., Belmann, P., et al. (2017). Critical Assessment of Metagenome Interpretation—a benchmark of metagenomics software. Nature Methods 14(11), 1063–1071. DOI: 10.1038/nmeth.4458. CAMI project website.

  • CAMI II: Meyer, F., Fritz, A., Deng, Z.-L., et al. (2022). Critical Assessment of Metagenome Interpretation: the second round of challenges. Nature Methods 19(4), 429–440. DOI: 10.1038/s41592-022-01431-4. CAMI project website.

  • Source dataset for the MP test subset: CAMI II multi-sample human microbiome dataset. Dataset DOI: 10.4126/FRL01-006425518. The bundled inputs are selected and cropped subsets of this collection; their exact transformations and file hashes are recorded in the bundle provenance.