pirl-unc / mhcgnomes / 33474521325
90%

DEFAULT BRANCH: main
Ran
Jobs 1
Files 41
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: 89.393%. Remained the same
33474521325

push

github

web-flow 
Curate SLA (swine) haplotypes (#143) (#163)

haplotypes.yaml did have swine entries -- keyed by MHC prefix as SLA:, not by
latin name, which is why the issue reported no Sus scrofa entry. But both
wrote their member alleles with the species prefix ("SLA-1*01:01") while the
loader hands the parser the string after the species, so each resolved
carrying zero alleles and printed a warning to stdout on every parse. Those
six warnings were the only ones in the whole ontology.

Hp-2.0 was also wrong on the data. Table 2 of Baekbo et al. 2017 (PMC5472656)
gives it as SLA-1*0201/*0701, SLA-3 null, SLA-2*0201; the entry read
SLA-3*02:01, the SLA-2 column filed one locus over. SLA-3*02:01 is a real
allele, so nothing complained.

Curates all nine class I haplotypes the SLA Nomenclature Committee has
designated, from the "Known haplotypes" half of that table. The A.0-Z.0 rows
below it are that study's own proposals, several resting on unnamed sequences,
and are left out.

Answers the open question about the Hp prefix from the literature rather than
adding a field. Reiner et al. 2024 (PMC10925748) states the ISAG/IUIS-VIC
scheme: swine haplotypes carry their own prefix, Hp- high resolution and Lr-
low resolution, numbered <class I>.<class II>. The prefix is part of the name,
so "Hp-17.0" and "SLA-Hp-17.0" both parse and no `haplotype prefix` key is
needed.

Adds the general guard: every one of the 71 curated haplotypes must keep every
allele it lists, across all species. The failure mode here was a warning on
stdout and a haplotype that claims a name and carries nothing.

Filed #162 for what could not be sourced -- Lr- haplotypes are defined by
allele groups (SLA-1*15XX, Blank) which the format cannot express, and the
class II Hp-0.y compositions are in a paywalled table.

0 of 36,752 corpus names change.

Claude-Session: https://claude.ai/code/session_012s4siLj2Vm33eNMGjNSazH

4239 of 4742 relevant lines covered (89.39%)

0.89 hits per line

Jobs
ID Job ID Ran Files Coverage
1 33474521325.1 41
89.39
 GitHub Action Run
Source Files on build 33474521325