25187188068
80%

Ran 30 Apr 2026 08:18PM UTC

Jobs 1

Files 27

Run time 1min

Badge

Embed ▾

Committed 30 Apr 2026 08:16PM UTC coverage: 70.772% (+0.08%) from 70.692%

Build # 25187188068

Build Type

push

github

Committed by

web-flow

Commit Message

v1.30.17: Refine class-label flag into severity tiers (#201) (#203)

Splits the previously-binary mhc_class_label_suspect into a four-tier
mhc_class_label_severity column (ok/borderline/suspect/implausible).

The binary v1.30.0 cutoffs (>=18aa class I, <=10aa class II) conflated
two distinct populations: bulged class-I peptides (13-17aa, biologically
real) and clear curation drift (e.g. 18+aa labeled class I). Same on
the class-II side: 8-10aa class-II ligands exist in the literature but
were lumped together with 5-7aa rows that almost always indicate label
mistakes.

Tiers per row (using peptide length and curated mhc_class):
- ok           canonical lengths
- borderline   bulged class-I (13-14aa) / short class-II (8-10aa)
- suspect      class-I 15-17aa / class-II 5-7aa
- implausible  class-I >=18aa or <=7aa / class-II <=4 or >=31aa

Backwards-compat: mhc_class_label_suspect = severity in {suspect, implausible}.
Adds exclude_class_label_implausible loader filter for the relaxed
cleaning path that retains bulged class-I peptides.

Coverage Stats

3632 of 5132 relevant lines covered (70.77%)

0.71 hits per line

Coverage Regressions

Lines	Coverage	∆	File
6	94.52	0.58%	observations.py

Jobs

ID	Job ID	Ran	Files	Coverage
1	25187188068.1	30 Apr 2026 08:17PM UTC	27	70.77	GitHub Action Run

pirl-unc / hitlist / 25187188068
80%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Coverage Regressions

Jobs

Source Files on build 25187188068

pirl-unc / hitlist / 25187188068 80%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Coverage Regressions

Jobs

Source Files on build 25187188068

pirl-unc / hitlist / 25187188068
80%

README BADGES
x