about summary refs log tree commit diff
path: root/test/python
diff options
context:
space:
mode:
authorPjotr <wrk@bosa.lan>2026-11-09 12:23:53 +0000
committerPjotr <wrk@bosa.lan>2026-11-09 12:23:53 +0000
commitc560ce726bc0c45ac87b8f0f9bf83870e25ac195 (patch)
treef5514258eb4e79b8503f9323ab3c0e13b59720dd /test/python
parent29eb1363caea8c0eb13eb91c5a44d4f2967b69da (diff)
downloadgenenetwork4-c560ce726bc0c45ac87b8f0f9bf83870e25ac195.tar.gz
compute: let GEMMA name the trait, so its plot can be titled
Two GEMMA runs over one BXD panel drew genuinely different pictures --
traces differ on all 20 chromosomes, peaks -log10 p 5.48 at chr8 97.0 Mb
against 5.43 at chr1 175.6 Mb -- and both came out headed
**"trait gemma: GEMMA association along the genome"**, because GEMMA's
request had no trait label at all: `/compute/gemma` fell back to
`<group>::gemma`, and `/plot/gemma` fell back to "gemma". The plot of each
trait was told apart by its annotation and nothing else.

`trait_name` is now an optional field on `GemmaRequest`, used for the
`Trait: ...` note (first in `notes`, as on `/compute/rqtl`), for the PheWAS
and R/qtl suggestions, and for the `/plot/gemma?trait=` suggestion -- so a
posted response titles its own plot, and `?trait=` still overrides.
Without it nothing changes: the label stays `<group>::gemma`.

* src/routes/gemma.rs: the field, the label resolution, the note, the
  plot href.
* tests/gemma.rs: the covariate run names its trait and asserts the note is
  first; tests/plot.rs: a GEMMA plot titles itself from the note, and an
  explicit `?trait=` wins over it.
* README.md (`/compute/gemma`, and the plot section), docs/www/SKILL.md
  (the GEMMA options list, and what the plot title comes from).

281 tests pass; the same four as before do not.
Diffstat (limited to 'test/python')
0 files changed, 0 insertions, 0 deletions