pirl-unc / tsarina / 27318830096
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: 74.147% (+0.03%) from 74.119%
27318830096

push

github

web-flow 
Group identical-protein paralogs SSX4/SSX4B and MAGEA2/MAGEA2B at pMHC level (#112)

SSX4 (ENSG00000268009) and SSX4B (ENSG00000269791) are distinct chrX loci that
encode the identical 188-aa (and 153-aa) SSX4 protein; MAGEA2 (ENSG00000268606)
and MAGEA2B (ENSG00000183305) likewise encode the identical 314-aa MAGE-A2
protein. Verified against Ensembl 111. This is the same situation as the
existing CTAG1A/CTAG1B (NY-ESO-1) and XAGE1A/XAGE1B groups: one antigen at the
pMHC level whose members must collapse so identical peptides are not
double-counted in a panel.

Register both as explicit _CTA_GROUPS with member aliases (incl. the
non-standard but plausible 'SSX4A'/'MAGEA2A' forms — HGNC names the primary
loci SSX4 and MAGEA2, with no 'A' suffix). Mirrors the CTAG1/XAGE1 pattern
exactly; selecting via the bare 'SSX4' alias now resolves to SSX4/SSX4B and
reports both member genes.

Note: this only fixes pMHC-level grouping. Proteoform-level TPM summation
(summing each sample's TPM across identical-protein members before computing
prevalence/percentile, across all datasets, in a public API) is a separate,
larger change landing in cancerdata — the current tsarina expression
aggregation across group members is max, TCGA-only, and private.

Stacked on the placental-antigen branch (needs SSX4B/MAGEA2B in the CTA table).
Bumps version to 1.13.0.

2696 of 3636 relevant lines covered (74.15%)

0.74 hits per line

Coverage Regressions

Lines Coverage ∆ File
84
91.05
 0.04% spanning.py
Jobs
ID Job ID Ran Files Coverage
1 27318830096.1 37
74.15
 GitHub Action Run
Source Files on build 27318830096