GNB4

Summary: Retain the LAMP2 interaction evidence outside generic binding annotation.

Reason: Validated yeast two-hybrid records support the GNB4–LAMP2 contact. The literal partner P13473-2 is preserved; the supplement's gene-level row does not independently identify its clone sequence. Generic protein binding does not specify the signaling function.

Summary: Retain the CASP6 interaction evidence outside generic binding annotation.

Reason: GNB4–CASP6 occurs in the original interaction table and validated yeast two-hybrid records. This positive contact does not by itself establish a caspase-regulatory mechanism. Generic protein binding adds little functional information.

Summary: Retain the SH3GLB1 interaction evidence outside generic binding annotation.

Reason: GNB4–SH3GLB1 occurs in the original interaction table and validated yeast two-hybrid records. The assay does not establish a membrane-remodeling role for GNB4. Remove the uninformative binding annotation without rejecting the contact.

Summary: Retain the ancestral activity-location assertion with its exact qualifier.

Reason: The Q9HAV0 target inherits the cytoplasmic assertion from AN0. The literal is_active_in relation is preserved: cytosolic client locations in folding events do not replace this activity assertion. Human GNB2 in the donor list is a distinct experimentally annotated paralog, not GNB4 self-evidence or circular support. The core section explains why no membrane-sidedness refinement is asserted.

Summary: Retain the placental lysosomal membrane association.

Reason: The original enrichment table explicitly lists G protein beta4, IPI00012451, in human placental lysosomal membranes: 13 peptides and adjusted p=0.023. This supports contextual association. Peptide uniqueness was not independently replayed, and neither integral membrane topology nor a lysosome-specific function follows.

Summary: Retain the cytosolic unfolded GNB4 client location.

Reason: Retain cytosolic localization of the unfolded beta-subunit client during CCT capture as non-core; the chaperonin performs the folding.

Summary: Retain the cytosolic GNB4 client during ATP-driven folding.

Reason: CCT-mediated folding contains unfolded and partially folded beta-subunit clients in the cytosol. Retain this maturation location as non-core, without assigning ATP hydrolysis or chaperone activity to the client.

Summary: Retain the cytosolic GNB4 client in chaperonin–PDCL assembly.

Reason: PDCL binds the partially folded beta-subunit client on CCT. Its explicit cytosolic location is retained as non-core maturation context.

Summary: Retain the cytosolic GNB4 client during ATP binding to CCT/TRiC.

Reason: ATP binds the CCT chaperonin around its unfolded beta-subunit client. Retain the client's cytosolic location as non-core; ATP binding is not assigned to the beta subunit.

Summary: Retain the cytosolic target leaf during G-beta–gamma assembly.

Reason: PDCL-assisted beta-gamma assembly retains a cytosol-labeled beta-subunit member within its plasma-membrane product. Keep this preparatory client location as non-core.

Summary: Retain mature-complex member labeling as contextual.

Reason: The receptor-driven event contains a mature heterotrimer and released beta-gamma dimer. Its cytosol-labeled target is nested in plasma-membrane complexes. Retain this member location as non-core; the label does not establish the active complex's membrane sidedness or freely soluble signaling.

Summary: Retain the cytosolic GNB4 client during chaperonin release.

Reason: PDCL releases the folded beta-subunit client from CCT into a cytosolic PDCL-bound intermediate. Retain this maturation location as non-core.

Summary: Retain the mature heterotrimer member location as contextual.

Reason: The output is a mature heterotrimer containing a cytosol-labeled beta-subunit member. Retain this database member location as non-core, distinct from the plasma-membrane parent complex; it does not directly locate the active signaling interface.

Summary: Retain membership in the inherited heterotrimeric G-protein complex.

Reason: The GNB4 leaf inherits the AN0 complex assertion. Bovine donor structures and human beta4 heterotrimer formation support membership, without assigning alpha-subunit GTPase chemistry to beta4.

Summary: Retain the conserved mouse-to-human heterotrimer membership.

Reason: Mouse P29387 has direct complex annotations. Human GNB4–gamma2 with alpha-oA forms a heterotrimer in rat neurons, corroborating the orthology transfer.

Summary: The broad signaling term is less informative than the existing GPCR process.

Reason: GO:0007186 is an is_a descendant of GO:0007165 and is already retained. The InterPro mapping is biologically compatible, but the existing GPCR annotation provides the informative process specificity.

Summary: Retain the inherited GPCR signaling process.

Reason: Human WT beta4 enhanced bradykinin-dependent IP3 signaling in COS7 cells expressing PLCB2; disease variants impaired it, corroborating the inherited process.

Summary: Expression-based (HEP) annotation from a proteomic study of adult human substantia nigra (PMID:22926577); detection in the tissue does not show a role in its development.

Reason: GNB4 was detected in a quantitative proteomic comparison of adult post-mortem substantia nigra from disease and control brains, with increased abundance reported in Huntington disease; the detection is a real observation. The study does not examine formation or maturation of the substantia nigra, so presence of the protein in the adult tissue is the whole basis of the developmental term. That over-reads an expression result; nothing refutes it, and the row is marked as an over-annotation, as is done for this class of proteomic expression annotations, rather than removed.

Summary: Retain the inherited signaling-complex adaptor activity.

Reason: The exact Q9HAV0 sequence maps to AN218, descending from the annotated AN0/PTN000457487 in PTHR19850. The incoming Git5 donor has experimental grounding: fission-yeast Git5 enhances the Git3–Gpa2 interaction in a budding-yeast assay. This supports assembly, not beta-subunit nucleotide catalysis. Later donor-list additions are retained separately; no target-specific loss was established.

Summary: Retain contextual association with a Radil-containing complex.

Reason: Original Figure 3B reports GNB4 in three of three purifications with both mouse Radil and its PDZ-deleted construct in human MDA-MB-231 cells. This supports complex association, without establishing purified binary binding or a GNB4-specific KIF14 mechanism.

Summary: Retain the contextual mouse-to-human synaptic activity transfer.

Reason: The actual P29387 mouse donor carries experimental SynGO is_active_in annotations from PMID:34932372. Retain the human orthology inference and qualifier as contextual. The original target-specific assay or peptide assignment was not independently recovered; the paper's abstract emphasizing other beta subunits does not refute the curated donor.

Summary: Retain the contextual extracellular-vesicle preparation association.

Reason: Supplemental Table 2 row 629 contains Q9HAV0 in the three-protein group P62879;Q9HAV0;P16520. Its peptide counts belong to that group, not independently to GNB4. Retain the curator's pooled EPS-urine preparation annotation with this ambiguity and co-isolation limits; no exosome-biogenesis role is inferred.