## FBXW7
- **UniProt:** Q969H0 · **batch:** proteostasis-batch-2026-06-13 · **review status:** COMPLETE (very comprehensive, ~2220 lines)
- **PN placement (3 rows):** Row1 ALP `Autophagy-Lysosome Pathway|Pre-initiation autophagy signaling|mTORC1 pathway, direct|Modulator of mTORC1 activity` (SHOC2-Raptor); Row2 `UPS|...|Cul1 substrate receptor|F-box|WD40`; Row3 `UPS|Ubiquitin and UBL binding|E3 ligase|CUL1 receptor|idiosyncratic Ub binding / WD40` (PMID:21070969). **PN-node mapping:** Row2 group=mapped GO:1990756 (already_in_goa_exact); Row3 group=mapped GO:0061630 (new_to_goa); Row1 ALP no_mapping (GO:0010506 reg. of autophagy held too_broad).
- **Consistency:** Rows 1–2 consistent. FBXW7 is the textbook tumor-suppressor SCF receptor; review has GO:1990756 (IDA x2, ACCEPT) — matching Row2 "already_in_goa_exact" — plus GO:0050816 phosphothreonine binding, GO:0030332 cyclin binding, and a large substrate set (cyclin E, MYC, NOTCH ICD, MCL1, RICTOR, MLST8, NR1D1, EGFR). Row1 mTORC1/SHOC2 role corresponds to KEEP_AS_NON_CORE downstream outputs. **Row3 INCONSISTENT:** PN projects GO:0061630 ubiquitin protein **ligase** activity from PMID:21070969 (Pashkova 2010, [DOI](https://doi.org/10.1016/j.molcel.2010.10.018)), a paper (per PubMed) about WD40 propellers as ubiquitin-**binding** domains regulating F-box turnover — with Cdc4 (the FBXW7 ortholog) shown to use Ub-binding to promote its own **auto-ubiquitination**, not substrate ligation. The review independently flags exactly this: GO:0043130 ubiquitin binding (IBA) is MARK_AS_OVER_ANNOTATED ("FBXW7 recognizes phosphodegrons, not free ubiquitin"). So the YAML directly contradicts the PN Row3 catalytic projection.
- **PN story / NEW pressure:** Row2 GO:1990756 already in GOA (IDA) and accepted — no new pressure. Row3 GO:0061630 over-reaches (wrong activity class for a non-catalytic receptor; and the source paper is about Ub binding/auto-turnover). Both GO:0061630 and GO:0043130 verified real via OLS. No defensible NEW catalytic term.
- **Mapping strategy:** Row2 correct (GO:1990756 matches review core MF; already_in_goa_exact). Row1 ALP no_mapping correct. **Row3 mapping wrong:** GO:0061630 should be GO:0043130 ubiquitin binding (and even that the review judges over-annotated for FBXW7 since recognition is phosphodegron-based) or no_mapping. Not broader/narrower issue — it is a wrong-activity issue.
- **Evidence alignment:** Strong overlap on the receptor story; Row2 PN PMID:15340381 (family review) vs review's gene-specific PMID:17434132 (cyclin E structure), PMID:15103331/15150404 (MYC), PMID:28007894 (STYX), PMID:35395208 (DEDHIL). Row3 PMID:21070969 is NOT cited in the FBXW7 review (only its IBA-derived ubiquitin-binding term is, via GO_REF:0000033) and is mischaracterized by the GO:0061630 mapping.
- **Verdict:** Rows 1–2 CONSISTENT; Row3 PN node OVER-REACHES (contradicts the review's own over-annotation call). **Recommended edits:** [MAP] change Row3 `Ubiquitin and UBL binding|E3 ligase` projection for FBXW7 from GO:0061630 ubiquitin protein ligase activity to GO:0043130 ubiquitin binding or no_mapping — PMID:21070969 = WD40 ubiquitin BINDING controlling F-box auto-turnover, not catalytic ligase; the FBXW7 review already marks ubiquitin binding as over-annotated.