PN placement:ALP|Autophagosome closure maturation and lysosome fusion|Sealing of autophagophore membrane|ESCRT-I complex component ; PN-node mapping: leaf mapped / GO:0000813 ESCRT I complex; "Sealing" group mapped / ok_for_propagation / GO:0000045 autophagosome assembly; class context_onlyGO:0016236. Projected: GO:0000813 (already_in_goa_exact), GO:0000045 (new_to_goa).
Consistency: Two issues. (1) GO:0000813 ESCRT I complex is consistent (ACCEPTed). (2) The PN new_to_goa projection of GO:0000045 autophagosome assembly has no support in the MVB12B review or GOA, and parallels MVB12A where macroautophagy is MARK_AS_OVER_ANNOTATED — so PN over-reaches here. (3) Internal review bug: the GO:0043130 ubiquitin binding row is correctly flagged negated: true (GOA = NOT|enables ... IDA PMID:20654576, an experimental NOT call) yet given action: UNDECIDED with reason "cannot be confirmed... before accepting or removing." The curator experimentally established NON-binding; per project rules this NOT annotation should be ACCEPTed (retained), not left UNDECIDED.
PN story / NEW pressure: PN asserts a brand-new autophagosome-assembly role for MVB12B (new_to_goa) via ESCRT-I sealing. No MVB12B-specific evidence supports it; the shared structural paper assays VPS28, and phagophore-closure ESCRT evidence (PMID:31519728) is for VPS37A. Conclusion: over-reaches; defer to expert question rather than annotate.
Mapping strategy: Same as MVB12A — the "Sealing" node's blanket ok_for_propagation of GO:0000045 to every ESCRT-I-component leaf is too strong for MVB12B (membership ≠ demonstrated closure function). Recommend candidate-only propagation.
Evidence alignment: Shared: PMID:32424346 (ESCRT-I scaffold) underpins both PN row and review CC/fission annotations; PMID:20654576 (MVB12A/B PTMs) is the basis of both the EGFR-context and the negated ubiquitin-binding row. Overlap strong; divergence only on autophagy inference.
Verdict: ESCRT-I membership consistent; PN GO:0000045 (new_to_goa) over-reaches; review has a negation/action inconsistency. Recommended edits: in MVB12B YAML, change the GO:0043130 ubiquitin binding row from action: UNDECIDED to ACCEPT (retain the experimental NOT annotation; GOA NOT|enables IDA PMID:20654576) and fix the summary to reflect NON-binding rather than "cannot be confirmed." Flag the "Sealing" node so GO:0000045 is not auto-propagated to MVB12B.