Consistency: Coherent. Notes, review, PN row ("TRAPP complex, core subunit"), and node mapping agree TRAPPC5 (Trs31 ortholog, small core subunit) acts via complex-level RAB1 GEF + ER-to-Golgi transport. Review carries GO:0030008 (IDA/IEA) and TRAPPII/TRAPPIII memberships. Notable, defensible deviation from raw GOA: the IBA GO:1990070 TRAPPI complex annotation is MARK_AS_OVER_ANNOTATED because mammalian TRAPP I existence is "in question" (PMID:21525244, PMID:27066478) — well argued, not a contradiction.
PN story / NEW pressure: PN Notes mention ATG9/ATG2 trafficking. Review keeps autophagy as TRAPPIII-membership context only; TRAPPC5 has no gene-specific autophagy evidence. Defensible. Conclusion: already captured via complex membership; no NEW term.
Mapping strategy: No change to node. Projected GO:0030008 (OLS-verified) already in GOA. One subtlety worth flagging: the review demotes TRAPPI (GO:1990070) as a mammalian artifact, but the PN node projects only the parent GO:0030008, so no conflict with the projection.
Evidence alignment: PN titles ("Membrane Trafficking in Autophagy"; PMID:27066478) both in review; review adds PMID:21525244 and Reactome. Generic protein-binding rows (PMID:25416956/28514442/31515488/32296183/33961781) over-annotated. No miscitation.
Verdict: Consistent; ACCEPT mapping. Includes a sound TRAPPI-overannotation call beyond the PN scope.