Skip to content

feat(github): add a form for the Studies using pathotypr category - #20

Merged
Paururo merged 1 commit into
mainfrom
feat/studies-discussion-form
Aug 17, 2026
Merged

feat(github): add a form for the Studies using pathotypr category#20
Paururo merged 1 commit into
mainfrom
feat/studies-discussion-form

Conversation

@Paururo

@Paururo Paururo commented Aug 17, 2026

Copy link
Copy Markdown
Member

The category was renamed from Show and tell and had no form, so it collects whatever people feel like writing.

What it asks

Field Required Why
The work yes DOI, preprint, repository, or just a title
Version yes see below
Which commands no which of the five did the work
Marker panel no published Zenodo panel, or one they curated
Reference genome yes the most consequential detail on the form
Organism yes where it is being pushed past what has been validated
What you used it for yes one sentence
What was awkward no the most useful field

Two choices worth explaining

The version field carries a real warning, not a tidiness request. Before 1.0.1, classify scanned only the forward strand and missed markers on reverse-oriented contigs, and genomes with no hits were dropped from the summary rather than reported as unclassified. An analysis run on draft assemblies with 1.0.0 has to be read with that in mind, and the only way anyone ever finds that out is if the version was written down.

The reference genome is required and the panel is not. A panel read against a different assembly produces confident, wrong calls and says nothing about it. Knowing which panel was used is interesting; knowing which assembly it was read against is what makes the numbers interpretable.

Verified

The filename was checked against the live category slugs from the API, not assumed:

marker-proposals          -> Marker proposals
q-a                       -> Q&A
studies-using-pathotypr   -> Studies using pathotypr

That check exists because getting it wrong is exactly how the original marker proposal form sat invisible: it was named marker_proposal.yml against categories that never included one.

The category was renamed from Show and tell and had no form, so it collected
whatever people felt like writing. This asks for the handful of things that
make a report useful to whoever reads it later.

The version field carries a real warning rather than a tidiness request.
Before 1.0.1, classify scanned only the forward strand and missed markers on
reverse-oriented contigs, and genomes with no hits were dropped from the
summary instead of being reported as unclassified. An analysis run on draft
assemblies with 1.0.0 has to be read with that in mind, and the only way anyone
finds that out is if the version is written down.

The reference genome is required and the marker panel is not. That is
deliberate: a panel read against a different assembly produces confident, wrong
calls and says nothing about it, so the reference is the single most
consequential detail on the form.

The last field asks what was awkward. A real analysis knows things about a tool
that no feature request written from the outside does.

Filename checked against the live category slug, which is how the original
marker proposal form came to be invisible for months.
@github-actions github-actions Bot added area:markers Marker panels, their format and their validation area:ci Workflows, checks and repository automation labels Aug 17, 2026
@Paururo
Paururo merged commit c14cc35 into main Aug 17, 2026
4 checks passed
@Paururo
Paururo deleted the feat/studies-discussion-form branch August 17, 2026 07:07
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

area:ci Workflows, checks and repository automation area:markers Marker panels, their format and their validation

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant