Roadmap and Known Issues

Roadmap & Known Issues

What is this?

Where the pipeline came from, what is planned, and known problems waiting for a fix.

Why does it matter?

So nobody is surprised by a known bug, and new contributors can pick something useful to work on.

What do I do?

Before starting new work, check here; fix items as separate, reviewed changes (Best Practices) and update this page.

timeline
    2024-04 : Custom B/Victoria Nextclade datasets
    2024-07 : Influenza A Nextclade datasets
    2024-11 : Builds move to one Snakemake pipeline
    2025-02 : Automated ingest (flusort + fludb), genome builds
    2026-10 : Surveillance report, conda env, run reports, data guard, source_prep

Loading

Milestones from the repository history.

History

• 2024-11-26: all builds constructed with a simplified Snakemake pipeline.
• Commit 9478cb9: ingest automated (flusort typing, fludb upload, per-build downloads); manual curation of sequences.fasta/metadata.tsv no longer needed; concatenated genome builds for H1N1 and H3N2.
• October 2026: single-script surveillance report (HTML + PowerPoint) replacing build-reports.py/render-reports.qmd; standalone conda environment replacing the Nextstrain Docker runtime (local nextstrain view was hard to reach on localhost, especially on WSL, and the Docker image may not include the reports' extra dependencies; see Getting Started); run provenance and run reports; GISAID/JHH data guard; prepare_source.py replacing manual source preparation; subclade colors matching the official Nextstrain builds (config/{subtype}/colors.tsv, scripts/sync_subclade_colors.py), fixing a colors setting in segments.smk that was never passed to augur export.
• October 2026 (known-issues fixes): GISAID acknowledgement in config/description.md; one upload script, scripts/nextstrain_upload.py (private by default, --public checks the GISAID acknowledgement, typed confirmation), replacing three broken helpers; Nextclade HA/NA datasets git-ignored, with every dataset's tag and the input files' checksums written to the run provenance file (scripts/record_provenance.py); tutorial samples confirmed released on GISAID; lab SharePoint link and PI contact filled in.

Roadmap

• Automated surveillance report (HTML + PowerPoint) for all builds.
• Automated concatenated genome builds for H1N1 and H3N2 (9478cb9).
• t-SNE for all builds with pathogen-embed (manuscript: https://bedford.io/papers/nanduri-cartography/).
• Convert legacy Jupyter notebooks in notebooks/ to marimo.

Known issues

Issue	Impact	Suggested fix
Maintainer email on Home is a placeholder	New users can only reach the maintainer through GitHub	Add the maintainer's JHU email
The tanglegram URL formula in config/description.md has a stray }, hard-codes PekoszLab in its second half, omits the seasonal-flu/ folder and lists segment m instead of mp	The formula shown on every Nextstrain page gives broken links	Correct the formula text (examples below it are right)

Troubleshooting / FAQ

• I found a new bug. Open a GitHub issue (no data in it!) and add a row here.
• I want to work on a roadmap item. Check its tier and talk to the maintainer first.

Next step → Home