RAS2 curation notes

2026 IBA re-review

Re-checked the four current RAS2 IBA annotations against GOA and the cached PANTHER
PTHR24070 PAINT table:

• GO:0005886 plasma membrane and GO:0003924 GTPase activity both trace to
  PANTHER:PTN000631348, a Ras-family node that has current IBDs for membrane localization
  and the core Ras catalytic activity.
• GO:0007163 establishment or maintenance of cell polarity traces to
  PANTHER:PTN000631919, an opisthokont node whose polarity assertion is compatible with
  the yeast Ras2 role in Cdc42/MAPK-dependent filamentous growth and Lte1 bud-cortex
  recruitment.
• GO:0007265 Ras protein signal transduction traces to PANTHER:PTN008393310, matching
  the conserved Ras GTPase switch role that Ras2 performs upstream of Cyr1/cAMP/PKA in
  budding yeast.

The SGD:S000005042 donor that appears in the GO:0005886 and GO:0003924
WITH/FROM sets is RAS2 itself. That is expected for target-seeded PAINT nodes: the
direct yeast SGD evidence helped place the IBD at PTN000631348, and the resulting
IBA records that plasma-membrane localization and GTPase activity are inherited
Ras-family properties rather than lineage-specific RAS2 observations.

PTN000631348 also carries GO:0007264 small GTPase mediated signal transduction,
the parent of GO:0007265 Ras protein signal transduction; RAS2 currently inherits
the more specific child from PTN008393310 without a redundant GO:0007264 IBA row in
the fetched GOA snapshot.

All four IBA rows are biologically sound for RAS2, so the YAML now keeps them as ACCEPT
and records propagation_review.root_cause: NO_FAILURE_CORE with the PTN ancestral node,
not the extant WITH/FROM genes, as the source entity. The full GOA WITH/FROM sets were
copied into supporting_entities for traceability.

The newer-paper search did not turn up a post-2023 primary paper that changes these RAS2
core function calls. Recent hits were pathway-level or engineering papers about yeast
Ras/cAMP/PKA signaling rather than new direct evidence for Ras2 itself.

2026-10-16: GO:0000411 row re-examined, still UNDECIDED (issue #4027)

GO:0000411 positive regulation of transcription by galactose (IMP,
PMID:16292676). The full text (Curr Genet 2006) is paywalled with no PMC copy.
SGD's phenotype curation for the paper
(https://www.yeastgenome.org/backend/reference/16292676/phenotype_details,
retrieved 2026-10-16) records only the RAS2-val19 activation allele
("utilization of carbon source: decreased", galactose, severe growth defect),
so the IMP most likely rests on that gain-of-function phenotype. The abstract
says the hyperactive allele causes "a defect in the expression of genes, like
GAL1" PMID:16292676 and explains suppression by PGM1/PCM1 as better use
of limiting glucose-1-phosphate. Read alone, that points to Ras/PKA hyperactivity
antagonising galactose induction, which does not support a positive term. But
the full text could contain a ras2 loss-of-function result, and the SGD curator
read it, so the row stays UNDECIDED (not REMOVE or MODIFY). The question is
recorded in suggested_questions.