• Home
  • Features
  • Pricing
  • Docs
  • Announcements
  • Sign In

pirl-unc / hitlist / 33006655814
81%

Build:
DEFAULT BRANCH: main
Ran 26 Aug 2026 07:48PM UTC
Jobs 1
Files 30
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

26 Aug 2026 07:43PM UTC coverage: 80.696% (+0.2%) from 80.498%
33006655814

push

github

web-flow
v1.46.0: fix APM flag contamination + silent arm collapse (#353, #354, #355) (#356)

Three defects that made the KO-vs-WT contrast silently wrong.

#353 — apm_columns_for_sample() unioned the parent study's
perturbations list into every sample's per-gene flags, so a CRISPR
panel marked all its arms (control included) as perturbed for every
gene. HAP1 wildtype reported apm_erap1_perturbed=True; 184/688 samples
were flagged apm_perturbed while their own condition_category was
unperturbed. Per-gene flags now come from the sample's own condition;
the panel context moves to study_apm_perturbed / study_apm_genes.

#355 — condition_category and the apm_* block were derived from two
different normalizations of `condition`, so they disagreed on 51
samples. Both now go through simplify_condition(), which also stops
discarding the "unperturbed + X" form (7 IFN-gamma-treated tumor
samples were labelled controls).

#354 — the observation->sample join first-picked an arm whenever no
discriminator separated the candidates, routing 19 of 41 contrast
studies entirely onto one arm with no column marking the guess. Now:
IEDB's per-peptide elution-condition enumeration is parsed and
outranks token scoring; narrative fields never decide an arm; ties
between disagreeing arms are refused; and undetermined rows keep only
what every candidate agrees on, marked sample_attribution=
pmid_ambiguous. Adds sample_attribution and is_control_arm.

Also fixes a pre-existing ruff RUF036 failure in scanner.py that broke
./lint.sh on main.

Claude-Session: https://claude.ai/code/session_014cBpazLoFvftLSLavT7T9X

5982 of 7413 relevant lines covered (80.7%)

0.81 hits per line

Coverage Regressions

Lines Coverage ∆ File
152
85.32
0.88% export.py
Jobs
ID Job ID Ran Files Coverage
1 33006655814.1 26 Aug 2026 07:48PM UTC 30
80.7
GitHub Action Run
Source Files on build 33006655814
  • Tree
  • List 30
  • Changed 3
  • Source Changed 0
  • Coverage Changed 3
Coverage ∆ File Lines Relevant Covered Missed Hits/Line
  • Back to Repo
  • Github Actions Build #33006655814
  • 2ed2517c on github
  • Prev Build on main (#27842275440)
  • Next Build on main (#33007348727)
  • Delete
STATUS · Troubleshooting · Open an Issue · Sales · Support · CAREERS · ENTERPRISE · START FREE TRIAL · SCHEDULE DEMO
ANNOUNCEMENTS · TWITTER · TOS & SLA · Supported CI Services · What's a CI service? · Automated Testing

© 2026 Coveralls, Inc