Vesicle Tethering BP Subtree — Obsoletion & MF Refactor (GO:0099022)

Bottom line: Before a vesicle docks and fuses, a tether (a long
coiled-coil golgin or a multisubunit complex such as TRAPP, COG or the
exocyst) makes first contact and bridges it to the target membrane. GO
has obsoleted GO:0099022 vesicle tethering and four children because
tethering is a binding activity, and minted the MF GO:7770062 vesicle
membrane tethering activity in its place; the recommended pattern is
that MF with a part_of link to the transport process it serves. We
recorded the upstream plan, the flagged InterPro and UniRule mappings,
and the reviews in this repo that touch the terms. The
obsoletion has landed: a QuickGO recheck on 2026-10-31 found all five BP
terms obsolete and GO:7770062 live, so the ontology side of this project is
complete. Eight human TRAPP subunit reviews
(TRAPPC1, 3, 4, 5, 8, 11, 12, 13) already MODIFY their NAS row on the
obsolete GO:0099022 to GO:0006888 ER to Golgi vesicle-mediated
transport. TMF1 is fixed in PR #3237 (merged): its NEW MF row and core
MF move from the stand-in GO:0060090 to GO:7770062, the obsolete
GO:0099041 NEW row and core BP are removed, and the "vesicle tethering
activity" proposed_new_terms entry is dropped now that GO has minted
it. The same PR moves USO1's docking row to GO:7770062. The
tether-subunit new reviews (Tier 2 onward) have not started.

This is the tethering step of the vesicle refactor. The docking step is
tracked in VESICLE_DOCKING_OBSOLETION
(#6379) and its regulation sub-issue in
SYNAPTIC_VESICLE_DOCKING_OBSOLETION
(#6415); the targeting terms that fold into transport processes are in
VESICLE_TARGETING_OBSOLETION (#6424).

Overview

A GO obsoletion proposal retired GO:0099022 vesicle tethering and four
children (BPs), redirecting the biology to a single molecular-function term:

• GO:7770062 vesicle membrane tethering activity.

The rationale matches the parallel
vesicle docking,
ER-PM tethering,
mito-ER tethering,
ciliary basal-body docking, and
synaptic vesicle docking trackers:
the tethering step is best represented as a binding activity (MF) rather
than a process (BP). Note that the existing GO:0099023 vesicle tethering
complex is a cellular-component term, not the MF — it must not be reused as
the MF replacement.

Per @ValWood on the upstream issue, the recommended post-obsoletion pattern is
the new MF in a part_of annotation extension to the relevant transport BP,
e.g. GO:7770062 vesicle membrane tethering activity part_of GO:0006888 endoplasmic reticulum to Golgi vesicle-mediated transport. Curators may also
fall back to existing or expanded "fusion" descendant BPs where they cover the
full tethering→docking→fusion pathway.

Upstream tickets

• Annotation tracker: geneontology/go-annotation#6375
  — "Review annotations to GO:0099022 vesicle tethering and children"
• Ontology tickets (from upstream body):
  • geneontology/go-ontology#31868
  • geneontology/go-ontology#31871
  • geneontology/go-ontology#31872
  • geneontology/go-ontology#31881
• Annotation review spreadsheet (not machine-accessible):
  https://docs.google.com/spreadsheets/d/1PN2Z6gl1XhUsNHs3eJXpm8ZkQCJYtfPDj3ogLWxgAa8
• Local follow-up: ai4curation/ai-gene-review#646
  — tracks the remaining EXOC4, EXOC6, C17orf75/Njmu-R1, TRIP11, GOLGA5,
  GORAB and RAB6A review queue

Obsoletion plan (per upstream)

At project creation, all five BP terms were still active in OLS on
2026-05-25 and the replacement MF GO:7770062 was not yet resolvable. By
the 2026-09-26 audit, GO:0099022 and GO:0099041 were obsolete and GO:7770062
had been minted. A QuickGO recheck on 2026-10-31 found all five BP terms
obsolete and GO:7770062 live, so the ontology side of this project is complete.

None of the five carries a term replaced by: GO records GO:7770062 only as
consider, which is correct — a biological process cannot be replaced outright
by a molecular function, so each annotation needs a curator rather than a
mechanical repoint. The exocyst-scoped subset of this plan is now recorded
structurally in modules/dmel_exocyst.yaml under proposed_ontology_changes
(see Ontology Changes), which is what
just export-ontology-changes reports upstream. The four mapping removals below
are still open there; the obsoletion itself is recorded as FIXED.

Affected upstream groups (from issue body, 2026-04-15)

Mappings flagged by upstream for review (InterPro2GO / UniRule):

The pinned local rules/arba/_interpro2go.txt still lists all three obsolete
InterPro2GO mappings above, so recheck InterPro2GO before starting EXOC4,
EXOC6 or C17orf75 to confirm whether GO:7770062 has been added. Recheck
UniRule2GO separately for UR000459766 and UR001419783.

Impact on this repo

The first local impact scan found TMF1 and the vesicle-docking gene USO1. The
later TRAPP reviews added eight more direct GO:0099022 rows:

A 2026-10-05 audit of the live YAML found TMF1 fixed, USO1 carrying one
GO:0048211 docking row that now MODIFYs to GO:7770062, and TRAPPC1/3/4/5/8/11/12/13
each carrying a GO:0099022 NAS row that MODIFYs to GO:0006888.

TMF1 was the highest-priority follow-up here and is fixed in #3237
(merged). The review now uses the minted GO:7770062 MF directly. The remaining
work is new review coverage for Tier 2 and later tether subunits, especially
EXOC4/EXOC6 and the C17orf75/Njmu-R1 family from the upstream InterPro2GO list.

Scope

• Organisms: predominantly human and yeast (ComplexPortal 95 + UniProt 11 +
  SGD 6 are the pending bulk), with smaller cross-kingdom tails (FlyBase, MGI,
  WB already done). Most ComplexPortal rows are macromolecular tethering
  complexes (TRAPPI/II/III, COG, exocyst, GARP, HOPS/CORVET, Dsl1, Golgin
  family) rather than individual genes.
• GO branch: a BP→MF refactor, not a within-branch swap. Annotations
  that meant "this protein is a tether/bridge" map cleanly onto
  GO:7770062 (likely MODIFY). Annotations that meant "this protein
  participates in a tethering-mediated transport event" should retain the
  upstream transport BP (GO:0006888 ER→Golgi, GO:0006891 intra-Golgi,
  GO:0006893 Golgi→PM, GO:0048278 vesicle docking, etc., the last of which is
  itself being obsoleted under #6379 — so
  cross-check with that tracker) and add the MF in part_of extensions.
• Type of fix: structural ontology change; per-gene curator judgment is
  required to choose the correct upstream transport BP and to distinguish
  "performs tethering" from "regulates tethering".

Candidate genes for initial review

Verify each with just fetch-gene <organism> <gene> and confirm UniProt
accessions before starting.

Tier 1 — refresh required (already in repo)

1. TMF1 (human, UniProt P82094) — genes/human/TMF1/. Direct
   GO:0099041 NAS row + proposed_new_terms already references GO:0099022 and
   GO:7770062. The single cleanest test case for transferring an existing
   review onto the new tether MF. Fixed in #3237 (merged): now on GO:7770062.

Tier 2 — InterPro2GO–flagged tether subunits (high-priority new reviews)

These genes are directly implicated in the upstream issue's flagged mapping
table — they are the most defensible new reviews because the obsoletion
spreadsheet will name them explicitly.

1. EXOC4 / Sec8 (human, UniProt Q96A65) — exocyst component. Flagged
   in #6375 via InterPro2GO mapping IPR039682 → GO:0090522. Also listed as a
   Tier 2 candidate in the sibling VESICLE_DOCKING_OBSOLETION;
   pick one tracker to host the review when it is started.
2. EXOC6 / Sec15 (human, UniProt Q8TAG9) — exocyst component. Flagged
   in #6375 via InterPro2GO mapping IPR007225 → GO:0090522. Also listed as a
   Tier 2 candidate in the sibling VESICLE_DOCKING_OBSOLETION.
3. C17orf75 / Njmu-R1 (human, UniProt Q9HAS0) — flagged in #6375 via
   InterPro2GO mapping IPR028280 → GO:0099041. The current human symbol for
   Protein Njmu-R1 is C17orf75, also known as SRI2; fetch this gene as
   C17orf75.

Tier 3 — canonical golgin / tether complex subunits

These are well-known tethers absent from the repo. They sit in the
ComplexPortal (95) and UniProt (11) buckets and would substantially populate
the new MF with high-quality experimental anchors.

1. GMAP-210 / TRIP11 (human, UniProt Q15643) — golgin; tethers
   COPI-derived vesicles to the cis-Golgi via N-terminal amphipathic helix.
2. Golgin-84 / GOLGA5 (human, UniProt Q8TBA6) — golgin tether for
   intra-Golgi transport vesicles; shares the N-terminal motif noted in
   TMF1-ai-review.yaml as conserved across golgin tethers.
3. GORAB (human, UniProt Q5T7V8) — GRIP-domain golgin tether involved
   in retrograde trafficking; disease-relevant (gerodermia osteodysplastica).
4. RAB6A (human, UniProt P20340) — recruits TMF1 and other golgin
   tethers via GTP-bound state; central upstream node for several of the
   genes above.

Proposed approach

1. Tier 1 refreshes are done. #3237 closed the loop for TMF1 and USO1
   after GO:7770062 was minted, and the TRAPPC1/3/4/5/8/11/12/13 reviews
   already carry the correct MODIFY calls for their obsolete GO:0099022 rows.
2. Queue Tier 2 (EXOC4, EXOC6, C17orf75/Njmu-R1) as new reviews —
   these are the genes upstream curators will inevitably touch when working
   through the InterPro2GO mapping fix.
3. Coordinate with sibling trackers:
   • VESICLE_DOCKING_OBSOLETION (#6379) —
     EXOC4 and EXOC6 are shared Tier 2 candidates; pick one tracker to host
     each new review when started.
   • ER_PM_TETHERING_OBSOLETION (#6383),
     MITO_ER_TETHERING_OBSOLETION (#6397),
     CILIARY_BASAL_BODY_DOCKING_OBSOLETION (#6405),
     SYNAPTIC_VESICLE_DOCKING_OBSOLETION (#6415) —
     parallel BP→MF refactors; review-text wording should stay consistent
     across the cluster.
4. Re-validate each affected review with
   just validate <organism> <gene> after editing.

Related obsoletions

This is one item in a coordinated cluster of "BP-tether/docking → MF binding
activity" obsoletions. Cross-reference the parallel trackers:

• VESICLE_DOCKING_OBSOLETION — sibling
  obsoletion (GO:0048278 and BP children, #6379). Notes that #6375 is the
  parallel tether obsoletion; this project file is that parallel tracker.
• SYNAPTIC_VESICLE_DOCKING_OBSOLETION
  — sub-issue of #6379 (#6415, GO:0099148).
• CILIARY_BASAL_BODY_DOCKING_OBSOLETION
  — sibling docking obsoletion (GO:0097711, #6405).
• ER_PM_TETHERING_OBSOLETION — sibling
  tether obsoletion (GO:0061817, #6383).
• MITO_ER_TETHERING_OBSOLETION — sibling
  tether obsoletion (GO:1990456, #6397).

Priority

Medium. The direct in-repo refreshes are done, and the TRAPP rows now
avoid an unsupported mammalian tethering-MF assertion. The project remains a
useful queue for EXOC4, EXOC6 and C17orf75/Njmu-R1 because those families are
named by the upstream InterPro2GO and UniRule cleanup, and later golgin/RAB
reviews would help populate the new MF with well-supported experimental
anchors.

Status

• 2026-05-25 — Project file created. Tracking upstream
  go-annotation#6375
  (last active 2026-04-23) and the four ontology tickets in the cluster.
  Obsoletion not yet applied; all five BP terms still active in OLS;
  replacement MF GO:7770062 not yet minted. Affected existing reviews:
  TMF1 (direct GO:0099041 row; review's proposed_new_terms already
  references GO:0099022 / GO:7770062). No gene reviews started or refreshed
  yet under this tracker.
• 2026-09-26 — TMF1 refresh fixed in #3237 (merged): NEW MF row and core
  MF GO:0060090 → GO:7770062 vesicle membrane tethering activity;
  obsolete GO:0099041 NEW row and core BP removed; proposed_new_terms
  entry removed because GO has minted the MF. USO1's GO:0048211 row
  also moves to GO:7770062 in the same PR. The TRAPPC1/3/4/5/8/11/12/13
  GO:0099022 rows already MODIFY to GO:0006888 and are unchanged. Tier 2
  new reviews not started, so maturity stays IN_PROGRESS.
• 2026-09-27 — #3237 merged, so the TMF1 and USO1 refreshes are on
  main. Tier 2 new reviews not started; maturity stays IN_PROGRESS.