The InterPro2GO mapping assigns YET2 the BAP29/BAP31 signatures (IPR008417, IPR040463; Pfam PF05529) and the GO terms attached to them - endoplasmic reticulum, membrane, and intracellular protein transport. These are family-level consequences of the domain match, so the two localization terms are well founded while the transport term inherits a mammalian BAP31 role never demonstrated for Yet2p.

SGD assigned the root terms GO:0003674 and GO:0008150 to YET2 with the ND evidence code, formally recording that no experimental data establish either a molecular function or a biological process for Yet2p. The ER localization is the only aspect with direct observational support.

PAINT placed Yet2p inside the BAP29/BAP31 clade (PANTHER PTHR12701, PTN000294723), propagating the ER membrane location together with mammalian BAP31 process terms (ER-to-cytosol retrograde transport, positive regulation of ubiquitin-dependent catabolism). The location is corroborated independently by topology, the C-terminal di-lysine motif and direct imaging; the process terms rest entirely on the mammalian descendants, with no yeast experiment supporting them for Yet2p.

The ER membrane annotation from this source is a vocabulary mapping of the UniProt SUBCELLULAR LOCATION line "Endoplasmic reticulum membrane; Multi-pass membrane protein", which UniProt itself tags ECO:0000250 (by similarity) rather than as a direct observation on Yet2p.

Yet2p was localized to the ER as part of a systematic N-terminal GFP-tagged library of ~1,800 endomembrane proteins. This is the source of the HDA C:endoplasmic reticulum annotation (ECO:0007005): a genuine direct observation, but a high-throughput one, and not a functional characterization of the gene.

The stable unit of the yeast BAP31 family is a Yet1p-Yet3p heterocomplex whose assembly is needed for subunit stability and correct ER localization. Yet2p is not reported as a member of this complex, which is the basis for treating its partners as unknown rather than assuming paralog redundancy.

The Yet complex is not packaged into COPII vesicles and so does not act as an ER export receptor; instead a fraction associates with the Sec translocation apparatus, and that association increases under ER stress. This argues against transplanting the mammalian BAP31 "ER export receptor" role onto the yeast family, including Yet2p.

The inositol-starvation growth defect is reported for yet1 and yet3 deletions specifically; no such requirement is attributed to YET2 anywhere in the abstract, so inositol prototrophy is a paralog phenotype and not a YET2 annotation basis.

UniProt's only FUNCTION statement for Yet2p is hedged ("May play a role") and is sourced to the non-PubMed-indexed Toikkanen et al. 2006 reference, so it does not constitute evidence sufficient to annotate a molecular function or a transport process.

Yet2p is a 160-residue multi-pass ER membrane protein with three transmembrane helices (3-23, 46-66, 104-124), a lumenal N-terminus and a cytoplasmic C-terminal tail (125-160) - the canonical BAP31-type architecture.

A di-lysine motif at residues 157-160 in the cytoplasmic tail provides the ER-retrieval signal underpinning steady-state ER residence.

Yet2p belongs to the BCAP29/BCAP31 family (Pfam PF05529 Bap31; InterPro IPR008417, IPR040463; PANTHER PTHR12701), the family shared with Yet1p, Yet3p and mammalian BAP29/BAP31.

Yet2p is the lowest-abundance member of the yeast BAP31 family and, unlike Yet1p and Yet3p, is not constitutively expressed - the observation (from Zung et al. 2024) that most directly argues Yet2p is not simply redundant with the Yet1p-Yet3p branch but is conditionally regulated. The inducing conditions are not established.

The report finds essentially no YET2-specific primary literature: the gene is characterized almost entirely through its better-studied paralogs, so the family-level material it summarizes (Yet3 ergosterol/ERGosome scaffolding, Yet1p-Yet3p Opi1/inositol regulation, mammalian BAP31 sorting) is paralog context rather than YET2 evidence.