pirl-unc / tsarina / 27098176197
77%

DEFAULT BRANCH: main
Ran
Jobs 1
Files 36
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: 71.568%. Remained the same
27098176197

push

github

web-flow 
Regenerate bundled CTA table from pinned HPA v23 (RNA + IHC) (#101)

The bundled table had mixed HPA provenance (tissue-derivation audit): the
filter-driving rna_deflated_reproductive_frac was ~current, but per-tissue
columns (rna_max_ntpm, rna_max_somatic_*, rna_*_max_ntpm) were from an
older release -- 27 rows referenced brain subregions absent from the
current consensus, and rna_max_ntpm was below a per-tissue value in 253
rows (impossible within one snapshot).

Add scripts/regenerate_table.py: re-derives ALL RNA and protein/IHC
columns for every row from a single pinned HPA release (v23 -- the newest
release serving both rna_tissue_consensus and normal_tissue), reusing the
same generators as add_cta_gene.py. Protein columns are reverse-engineered
from normal_tissue.tsv (validated to reproduce the shipped columns at
97.5%); a curated _CROSS_REACTIVE_IHC override forces IHC to 'no data' for
sequence-near-identical paralog antigens (MAGEB6, XAGE2, CT45A8, CT45A9)
whose shared antibody cross-reacts (Low protein where RNA is ~0 nTPM).

Membership delta (expressed CTA_gene_names 261 -> 263):
  + CT47A8/CT47A9/CT47A10  -- v23 gives them clean testis-only IHC
  - OOEP                   -- v23 deflated frac 0.9496 < 0.95 (Approved)
The four cross-reactive genes are preserved (override). Per-tissue columns
are now internally consistent and reproducible from one release.

Run: python scripts/regenerate_table.py [--apply]. Docs + count tests
updated; 407 tests pass.

2575 of 3598 relevant lines covered (71.57%)

0.72 hits per line

Jobs
ID Job ID Ran Files Coverage
1 27098176197.1 36
71.57
 GitHub Action Run
Source Files on build 27098176197