PN placement:ER proteostasis|Protein transport|Removal of misinserted transmembrane proteins (group, no_mapping); PN-node mapping: propagating term comes from the parent classER proteostasis|Protein transport = mapped, ok_for_propagation_to_go, GO:0015031 protein transport (new_to_goa, confirmed absent from goa.tsv).
Consistency: Deep research (falcon), review YAML, and PN are mutually consistent on the modern model: ATP13A1 is a P5A-ATPase transmembrane-helix dislocase (GO:0140567 / GO:0140569), not a cation pump. The review aggressively (and defensibly) REMOVEs the legacy Mn2+/cation-transport annotations (GO:0071421, GO:0098655, GO:0034220, GO:0015410, GO:0006874) and MODIFYs the P-type-transporter terms to dislocase. No internal contradictions.
PN story / NEW pressure: The PN leaf names exactly the gene's core role ("removal of misinserted TM proteins") and is left no_mapping (correct — too heterogeneous). No NEW GO pressure: the specific role is already fully captured by existing GO:0140567 + GO:0140569 (both ACCEPTed, with IDA support). proposed_new_terms is empty. Already captured.
Mapping strategy: Gene does not change the node. The only projected term, GO:0015031 protein transport, derives from the broad class node and is far broader than the gene's dislocase function — it is a generic umbrella, not a claim about ATP13A1's specific activity. Acceptable as a class-level propagation but uninformative for this gene; the gene-specific terms live in the review, not the projection.
Evidence alignment: PN dossier carries no reference titles for this gene; review anchors on PMID:32973005 (P5A dislocase), PMID:36264797 (MTCH2/ER dislocase), PMID:24392018 (ER localization). No divergence to reconcile.
Verdict: CONSISTENT — no NEW term warranted; PN class projection (protein transport) is broader-but-harmless. No edits required.