pirl-unc / tsarina / 27316174951
77%

DEFAULT BRANCH: main
Ran
Jobs 1
Files 37
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: 74.119%. Remained the same
27316174951

push

github

web-flow 
Add placental onco-germline antigens + fix mislabeled paralog IDs (#111)

* Add placental onco-germline antigens + fix mislabeled paralog IDs

Broadens the CTA candidate universe to the placenta-restricted arm of the
cancer-germline antigen family and corrects three paralog entries that were
silently no-ops.

Placental antigens (tsarina#110): add the CGB (hCG-beta), PSG, placental
syncytin/ERV envelope, placental galectin (PP13/PP14) and placental
lactogen/GH families as candidates tagged `placental_antigen`, then let the
reproductive-restriction filter judge them rather than excluding by hand.
hCG-beta (CGB8) is dropped from NON_CTA_EXCLUDED for the same reason. The
somatically-broad members (CSH1, GH2, most CGB/PSG) self-exclude; CSH1,
ERVH48-1, ERVV-2, LGALS13/14, PSG2/6 survive into the expressed set.

Paralog ID fixes: MAGEA2B, SSX4B and CT45A5 previously carried their
siblings' Ensembl gene IDs, so they collided with existing rows and were
never actually added. Corrected to their own primary-assembly chrX loci
(verified against Ensembl 111). MAGEA2B/SSX4B encode proteins identical to
MAGEA2/SSX4 (the NY-ESO-1 CTAG1A/B and XAGE1A/B situation) so they must be
in the universe to keep their peptides out of the non-CTA negative set.

Relax the add_cta_gene.py guard: filter-failing candidates are now recorded
with passes_filters=False (universe-but-excluded) instead of refused, so a
curated set filters down by the data, not by hand.

Counts: expressed CTAs 271 -> 272 (SSX4B), filter-passers 286 -> 297.

Bumps version to 1.12.1.

* Review fixes: correct issue refs (#110 is the build-leak issue) + CSH1 note

- The placental/paralog work cited tsarina#110, but #110 is the unrelated
  *.regen.csv build-leak issue (fixed here by the deploy.sh/pyproject change).
  Repoint those comment refs to this PR (tsarina#111).
- The script claimed 'CSH1, GH2 self-exclude', but CSH1 is placenta-dominant
  (~9508 nTPM placenta) and passes the reproductive filter into th... (continued)

2692 of 3632 relevant lines covered (74.12%)

0.74 hits per line

Jobs
ID Job ID Ran Files Coverage
1 27316174951.1 37
74.12
 GitHub Action Run
Source Files on build 27316174951