vas2: original ProtNLM2 function paragraph

Source: preserved XML, pre-release post-processed-2026_02_28k.xml. The source paragraph is reproduced verbatim. Current sequence and taxonomy are identity checks, not proven prediction-time input.

Component of the adaptor complexes which link clathrin to receptors in coated vesicles. Clathrin-associated protein complexes are believed to interact with the cytoplasmic tails of membrane proteins, leading to their selection and concentration.

Atomic claim assessments

Claim 1: Component of the adaptor complexes

Assessment: CNN (score 2). Target pull-downs and subunit deletions establish Aps1 as the sigma subunit of AP-1. The generic plural describes the adaptor-complex class; it is supported for AP-1 and does not establish membership of Vas2 in AP-2 or AP-3.

Assessment: CNN (score 2). AP-1 is a conserved clathrin-associated cargo adaptor, and its gamma/sigma hemicomplex recognizes cargo sorting motifs. Target Apl4-Aps1 interaction and endosomal traffic defects establish the corresponding adaptor architecture. The linkage is executed by the assembled complex, not by isolated Vas2 simultaneously supplying all cargo and clathrin contacts.

Claim 3: Clathrin-associated protein complexes are believed to interact with the cytoplasmic tails of membrane proteins, leading to their selection and concentration.

Assessment: CNN (score 2). Mammalian AP-1 gamma/sigma pull-downs establish recognition of cytosolic dileucine cargo motifs. The sigma-specific target family assignment, Apl4-Aps1 association, and Syb1 sorting phenotype support transfer of the shared cargo-selection mechanism. Exact motif preferences or receptor identities in fission yeast are not inferred from mammalian isoform preferences. This complex-level cargo sorting mechanism is already represented by curated clathrin-cargo adaptor and transport annotations.

These are prose-claim assessments; no GO mappings were invented for the paragraph. CNN denotes established equivalent biology, including justified conserved-family inference, and makes no assertion about training-data membership.