Cargo receptor ligand activity is going

GO:0140355 is obsoleted with no replacement term

AI Gene Review · projects/CARGO_RECEPTOR_LIGAND_ACTIVITY_OBSOLETION · 2026

Bottom line

  • Being recognised by a receptor is not an activity. The receptor gets GO:0038024 cargo receptor activity with the ligand as has_input.
  • 172 annotations, but 157 are Ensembl projections from 3 seeds. The real job is 6 experimental rows.
  • Here, TCN2 and CBLIF list the term in core_functions and will fail validation when it goes. Scoped, waiting upstream.

The B12 uptake system

Why GO is dropping the term

  • Definition: "interacts with a cargo receptor and initiates endocytosis", which describes what happens to the protein.
  • It is is_a protein binding with no logical link to GO:0038024.
  • The resemblance to "receptor ligand" terms is nominal: those sit in the signaling receptor branch, which is disjoint from cargo receptors.
  • Minted in 2019 for ApoB as the LDLR ligand; no apolipoprotein carries it today.

The footprint

Impact on this repo

Review Row action core_functions Tier
human TCN2 ACCEPT (EXP) GO:0140355 MF 1: breaks validation
human CBLIF ACCEPT (EXP) GO:0140355 MF 1: breaks validation
human TCN1 KEEP_AS_NON_CORE (EXP) none 2: re-term row
human HSPA12A prose only none 3: reword
CD320, CUBN, AMN carry GO:0038024 receptor side 4: check lossless

TCN2 also marks its protein-binding row over-annotated because GO:0140355 is "the informative MF"; that reasoning needs rewriting.

Status and next steps

  1. Wait for go-ontology#32466 before editing core_functions; the ticket has reversed direction once.
  2. Fix TCN2 and CBLIF together: REMOVE the rows, re-point core_functions to GO:0031419 and possibly GO:0140104 molecular carrier activity.
  3. Decide GO:0140104 once and apply it to all three carriers.
  4. Flag upstream: yeast ATG5 IDA and orphaned mouse Hpse ISO.

Read more: projects/CARGO_RECEPTOR_LIGAND_ACTIVITY_OBSOLETION.md