## CACYBP
- **UniProt:** Q9HB71 · **batch:** proteostasis-batch-2026-06-06 · **review status:** COMPLETE
- **PN placement:** `Cytonuclear proteostasis|Chaperone|HSP90 system|HSP90 cochaperone|CS domain containing` ; **PN-node mapping:** cochaperone type → mapped/ok_for_propagation GO:0051879 Hsp90 protein binding (goa_status=more_specific_than_existing_goa); subtype/group/class/branch → no_mapping.
- **Consistency:** Major divergence. PN classifies CACYBP as an **HSP90 cochaperone** (CS/p23-like domain) and projects GO:0051879 Hsp90 protein binding. The review, the notes, and the Falcon deep research instead establish CACYBP/SIP as a **Ca2+-regulated adaptor in a Siah1-based (non-cullin) E3 ubiquitin ligase** that degrades beta-catenin, and as an **S100A6/calcyclin-binding** protein (PMID:16085652, 18803400, 22295074). The review has NO Hsp90-binding annotation and does not mention HSP90 cochaperone function. The CS/p23-like (HSP20-like, IPR007052/IPR037893) domain is the only basis for the HSP90-cochaperone call; CS-domain presence does not establish Hsp90 binding for this protein.
- **PN story / NEW pressure:** PN asserts an Hsp90-cochaperone role NOT in existing GO annotations and NOT supported by the review/literature. GO:0051879 is a real term (verified) but is unsupported for CACYBP — the deep research found no primary Hsp90-binding data (the only Hsp90 link is review-level mention of radicicol-induced abundance changes, not binding). Conclusion: over-reaches — projecting Hsp90 binding onto CACYBP is a domain-similarity over-transfer. The protein's real shared MF is molecular adaptor activity (GO:0060090, already ACCEPT) and S100/ubiquitin-ligase binding.
- **Mapping strategy:** This gene argues the HSP90-cochaperone node should not project GO:0051879 onto CACYBP. CS-domain membership alone is too broad to assert Hsp90 binding (CS domains occur in non-Hsp90 contexts, as here). Treat as a node where this member is mis-bucketed.
- **Evidence alignment:** PN cites no references. Review/deep-research PMIDs (16085652 Siah1 assembly; 18803400 S100A6 structure; 22295074 gastric-cancer beta-catenin) are entirely about E3-adaptor/S100 biology — zero overlap with any Hsp90-cochaperone literature.
- **Verdict:** Domain-based misclassification — CACYBP is a Siah1 E3-ligase adaptor / S100A6-binding protein, not a demonstrated HSP90 cochaperone; Hsp90-binding projection is unsupported. **Recommended edits:** [MAP] remove/flag GO:0051879 projection onto Q9HB71 and reconsider the HSP90-cochaperone placement; the gene's shared MF is better represented by molecular adaptor activity (GO:0060090) / ubiquitin protein ligase binding, already in the review.