pirl-unc / hitlist / 27359346616
80%

DEFAULT BRANCH: main
Ran
Jobs 1
Files 29
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst
coverage: 79.925% (+0.1%) from 79.788%
27359346616

push

github

web-flow 
v1.41.0: add `export ms`/`peptide-summary` + serotype broadening (reimplements #142) (#338)

Reimplements the high-value parts of the stale draft #142 cleanly against
current main (that branch was ~150 commits behind and unmergeable). #142 is
closed in favor of this.

New CLI surface:
- `hitlist export ms` is now the canonical name for the MS observations
  export; `export observations` stays as a backward-compatible alias.
- `hitlist export peptide-summary --gene PRAME --serotype A24` collapses MS
  observations to one row per peptide, scoring support on a single target
  allele/serotype across ranked tiers (mono/multi exact-allele, same-serotype,
  class-only sample-genotype, unknown) plus a cancer/healthy/adjacent/other
  source breakdown. Aimed at "which PRAME peptides might present on A24 in
  cancers?".
- `--peptide` filter added to `export ms` (generate_observations_table already
  supported the kwarg; only the CLI flag was missing).

Serotype broadening (curation.py): a split serotype like `A2403` now also maps
to its broad parent `A24` when that parent exists in the mhcgnomes table, so a
broad `--serotype A24` query matches A*24:03. The broad name becomes canonical
(`allele_to_serotype("HLA-A*24:03") == "HLA-A24"`). This is what makes the
peptide-summary serotype query behave correctly for split serotypes.

export.py: new generate_ms_peptide_summary_table() + helpers. cli.py: _export_ms
/_export_peptide_summary dispatch helpers (kept pandas out of cli import path).

Deliberately dropped from #142: the speculative export-context refactor and the
1.15.x version churn. Tests added for the peptide-summary buckets, the CLI
helpers, the scope/single-target validation, and the serotype broadening.

5721 of 7158 relevant lines covered (79.92%)

0.8 hits per line

Coverage Regressions

Lines Coverage ∆ File
148
51.85
 0.77% cli.py
114
84.44
 0.23% export.py
24
91.93
 0.06% curation.py
Jobs
ID Job ID Ran Files Coverage
1 27359346616.1 29
79.92
 GitHub Action Run
Source Files on build 27359346616