sta-2 (C. elegans) notes

Literature review integration (2026-10-04)

Added two reviews (full text cached) to the review YAML:

• Taffoni & Pujol 2015 (PMID:26716073), review of epidermal innate immunity:
• STA-2 and ELT-3 are needed for nlp AMP transcription PMID:26716073.
  Added to GO:0006357, GO:0010628 and the core function.
• SNF-12 binds STA-2 in endocytic vesicles PMID:26716073. Added to GO:0030139.
• STA-2 is nuclear and hemidesmosome-associated PMID:26716073
  and binds MUP-4 PMID:26716073.
  Added to GO:0030056, GO:0098733 and the nucleus IDA.
• The severe-injury response is p38-independent but needs STA-2 PMID:26716073. Added to GO:0009611.
• Activation is non-canonical and not fully understood PMID:26716073. Added to the GO:0007259 REMOVE.
• Wang & Levy 2012 (PMID:24058748). Its F58E6.1 is sta-2 (UniProt ORFNames F58E6.1).
• STA-2 is a divergent paralog from a nematode-specific duplication PMID:24058748. Added to the GO:0007259 REMOVE.
• Caveat for the DNA-binding terms: only a weak DNA-binding domain match PMID:24058748,
  and no STAT-motif binding in an unpublished (data not shown) EMSA PMID:24058748.
  Added as a caveat to the GO:0000978 reason. The action stays ACCEPT, but GO:0000978 may be worth revisiting (see suggested_questions on STA-2 binding sites).
• Liongue & Ward 2013 (PMID:24058787) was not added. It makes no STA-2-specific point beyond what the sta-1 review already cites.
• Also added the missing propagation_review (PTN000927860, LINEAGE_OR_TAXON_MISMATCH) to the GO:0007259 IBA REMOVE.
• No actions changed.

Cross-gene briefing for GO editors and PAINT curators: https://claude.ai/artifact/4MLfbbbWWnLchDvspwdPuM

OpenScientist: does STA-2 keep a functional STAT DNA-binding domain? (2026-10-04)

Report: sta-2-hypotheses/sta2-dbd-dna-binding/openscientist.md (hypothesis framed neutrally;
the review's ACCEPT on GO:0000978 was withheld from the run).

• Verdict: the STAT DNA-binding-domain fold is retained (AlphaFold model confident over the
  region), but sequence-specific DNA binding is not supported.
• Checked against repo data before wiring: STA-2's UniProt entry has only the fold-level
  superfamilies IPR008967 and IPR012345 and no STAT DNA-binding-domain signature, whereas
  STA-1's entry has Pfam PF02864 (STAT_bind), InterPro IPR013801 and CDD cd14801. STA-2 is
  567 aa against STA-1's 706 aa.
• Also from the run (not independently re-derived): none of the three base-contacting
  residues of DNA-bound STAT1 (PDB 1BF5) is conserved in STA-2, and STA-2 lacks the
  activating tyrosine.
• Consistent with Wang & Levy 2012 PMID:24058748.
• Changes: GO:0000978 ACCEPT -> MARK_AS_OVER_ANNOTATED; GO:0000981 and GO:0003700 ACCEPT ->
  MODIFY to GO:0140110 transcription regulator activity; GO:0003677 ACCEPT -> UNDECIDED;
  core function MF GO:0000981 -> GO:0140110. The worm STAT module's STA-2 annoton follows.
• Decisive experiment: STA-2 ChIP-seq in the epidermis (with and without SNF-12), or EMSA of
  recombinant STA-2 DNA-binding domain on AMP promoter elements.