pirl-unc / hitlist / 25049901808
80%

DEFAULT BRANCH: main
Ran
Jobs 1
Files 25
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: 65.395% (-0.6%) from 65.98%
25049901808

push

github

web-flow 
v1.29.0: hitlist pmhc query + top-level reshuffle (#171)

Adds the protein x allele evidence query that downstream training and
modelling work has been hand-rolling.  ``hitlist pmhc query`` takes
proteins and MHC alleles, returns one flat row per (gene, allele,
peptide) with per-row PMID source list and an optional NetMHCpan /
MHCflurry binding-affinity score.

Default text output is a "table grouped under section headers" — gene
and allele are headers; peptide rows align in a column-formatted block
beneath each allele.  ``--format csv`` and ``--format json`` are also
supported.

Top-level reshuffle promotes commands that aren't pMHC-MS evidence out
of the ``export`` group:

  hitlist samples         (was: hitlist export samples)
  hitlist proteomics      (was: hitlist export bulk-proteomics)
  hitlist expression      (was: hitlist export line-expression)

``export`` now keeps only the pMHC-evidence flavors (observations,
binding, training, peptide-counts).  Old paths still work for one
release with a stderr deprecation notice.

Argparse setup factored into ``_add_samples_args`` and
``_add_line_expression_args`` helpers so the new top-level and the
deprecated ``export`` aliases share one source of truth for flags.

Tests:
- tests/test_pmhc_query.py — 5 tests covering grouping, multi-mapping,
  protein filter, empty result schema, and grouped-table rendering.

3033 of 4638 relevant lines covered (65.39%)

0.65 hits per line

Coverage Regressions

Lines Coverage ∆ File
230
29.35
 -2.51% cli.py
Jobs
ID Job ID Ran Files Coverage
1 25049901808.1 25
65.39
 GitHub Action Run
Source Files on build 25049901808