cryaba (Danio rerio) review notes

Zebrafish alpha-crystallin B chain a (alphaB1-crystallin / alphaBa-crystallin), UniProt Q9PUR2 (TrEMBL). The entry has no FUNCTION and no SUBCELLULAR LOCATION line, only the small heat shock protein family assignment, so all functional evidence comes from primary zebrafish literature.

Re-review 2026-09-29

• Reconciled the yaml against the refreshed GOA. Dropped the two obsoleted GO:0051082 unfolded protein binding rows (IBA and IDA), the GO:0046872 metal ion binding IEA row and a duplicate GO:0005212 IEA row under GO_REF:0000120, none of which GOA now carries; resolved the two PENDING rows, GO:0005212 IEA (GO_REF:0000002, InterPro:IPR003090) and GO:0140309 unfolded protein holdase activity IDA (PMID:15692462), both to ACCEPT.
• As for cryaa, GO:0140309 is accepted rather than replaced by a bespoke NTR: the term's carrier/escort definition sits awkwardly on a small heat shock protein, but GO lists "holdase" among its synonyms and GOA migrated the gene onto it, so the mismatch is recorded as a suggested question and proposed_new_terms was dropped.
• Corrected a paralog mix-up in the description. The previous text said cryaba is lens-specific and then attributed to it "lower expression in heart, brain, skeletal muscle, and liver (PMID:16420472)" - but that expression pattern belongs to alphaB2 (cryabb). The paper states the opposite for this gene PMID:16420472 and concludes PMID:16420472.
• That correction drives the grading of the four PMID:25866181 morpholino rows, which come from a screen of human myofibrillar myopathy gene orthologs. GO:0007519 skeletal muscle tissue development and GO:0030239 myofibril assembly remain KEEP_AS_NON_CORE, on the grounds that chaperoning sarcomeric clients is a real participation mechanism for an sHSP (it is the basis of human CRYAB myopathy), but with the morphant and paralog-expression caveats stated. GO:0060047 heart contraction and GO:0007626 locomotory behavior changed KEEP_AS_NON_CORE -> MARK_AS_OVER_ANNOTATED: contraction is performed by the sarcomere and behaviour by the whole animal, so both are readouts of the structural defect already captured, two steps removed from the molecular activity PMID:25866181. No row was removed: the curator read the full text and the morpholino is recorded in the annotation.
• GO:0036438 maintenance of lens transparency IMP ACCEPT, now grounded in the null-mutant result PMID:38705506 with the attribution control PMID:38705506.
• GO:0140309 holdase ACCEPTed as a core function with an explicit quantitative caveat: PMID:15692462. The in vivo cataract result is what justifies treating the weaker activity as functionally central for this paralog.
• GO:0005886 plasma membrane IDA ACCEPT -> KEEP_AS_NON_CORE. The row comes from a paper about lengsin (PMID:18406404) whose cached record is abstract-only; per the do-not-overrule rule the ZFIN curator's reading is accepted and the row kept, but membrane association is not treated as a core location, and plasma membrane was removed from core_functions.
• GO:0009408 response to heat IBA ACCEPT -> KEEP_AS_NON_CORE and GO:0042026 protein refolding IBA MODIFY -> MARK_AS_OVER_ANNOTATED, with the same arguments and propagation_reviews used for cryaa (node PTN000897708 is seeded by inducible invertebrate sHSPs; holdases do not perform the refolding step).
• Added reference_review to all five PMIDs, including a note on PMID:16420472 recording the previous misattribution, and kept one deep-research statement: the proteomics confirmation that an alphaBa-crystallin peptide present in wild-type adult lenses is absent in cryaba nulls. This notes file did not previously exist. Validation: zero errors, zero warnings.

Alignment to the holdase new-term convention (2026-09-29)

The two GO:0140309 IDA rows were first accepted in this re-review, but that diverged from the convention used for the paralogs (cryabb) and for the other small heat-shock proteins in this repository (human CRYAA, HSPB2, HSPB3, fly Hsp26/Hsp27), which treat GO:0140309 as still carrying a carrier-specific definition (escort to an acceptor; go-ontology#30552) and request a new "holdase chaperone activity" term instead. For consistency the rows are now MODIFY with an NTR replacement, a proposed_new_terms entry was added, and the core function uses proposed_molecular_function. The experimental evidence and its interpretation are unchanged; only the target term differs.