## CNPY3
- **UniProt:** Q9BT09 · **batch:** proteostasis-batch-2026-06-07 · **review status:** COMPLETE
- **PN placement:** `ER proteostasis|Chaperone|HSP90 system|HSP90 cochaperone|Folding of TLRs` ; **PN-node mapping:** leaf `[subtype] Folding of TLRs` `no_mapping`; `[type] HSP90 cochaperone` → `mapped` GO:0051879 Hsp90 protein binding (`more_specific_than_existing_goa`); group/class/branch unmapped.
- **Consistency:** Strongly consistent. Notes, YAML, and PN all describe CNPY3 (PRAT4A) as an ER-luminal TLR-specific co-chaperone for HSP90B1/gp96 (ATP-sensitive interaction), required for TLR1/2/4/5/7/9 folding/ER exit (not TLR3), causal in DEE60. GO:0051879 verified real. The review independently proposes GO:0051879 (action NEW) for the gp96 interaction — directly matching the PN projection.
- **PN story / NEW pressure:** PN asserts Hsp90 binding (GO:0051879) — genuinely absent from GOA (the GOA file has NO Hsp90/Hsp90-binding row, only signaling receptor binding + bare protein binding + ER lumen). So the dossier's `more_specific_than_existing_goa` label is technically too weak: it is effectively `new_to_goa`. The review goes further and proposes a coherent NEW set (GO:0051082 unfolded protein binding, GO:0051879, GO:0034975 protein folding in ER, GO:0045087 innate immune response, GO:0034123 positive regulation of TLR signaling) — all defensible and well beyond the single PN term. ADD GO:0051879 confirmed.
- **Mapping strategy:** PN maps conservatively at `[type]` (cochaperone), leaving the "Folding of TLRs" leaf and HSP90-system containers unmapped — appropriate. Only refinement: the goa_status should be `new_to_goa` (no Hsp90-binding annotation exists in GOA), not `more_specific_than_existing_goa`.
- **Evidence alignment:** PN listed no reference titles for this row. Review evidence is UniProt (gp96/TLR interactions) + Reactome (R-HSA-1678923/1678944 TLR folding by GP96/CNPY3) + mouse PRAT4A mechanism (PMID:20865800, cited in notes); GOA IPI rows (SLITRK4 etc.) correctly dismissed as HT noise. No conflict.
- **Verdict:** Consistent; PN GO:0051879 is correct and already in the review's NEW set, but its goa_status should be `new_to_goa` (not "more_specific"). Review's broader NEW terms are well supported.

**Recommended edits:** [MAP] Correct goa_status for GO:0051879 on the CNPY3 row from `more_specific_than_existing_goa` to `new_to_goa` (no Hsp90-binding annotation exists in CNPY3 GOA). (Review already proposes GO:0051879 + the full chaperone/TLR NEW set; no YAML change needed.)