NBN (nibrin, NBS1) review notes
Literature search (no deep-research provider available)
No deep-research provider keys were available, so this review rests on a manual search. Sources checked:
- UniProt O60934 (
NBN-uniprot.txt): FUNCTION, SUBUNIT, DOMAIN and PTM comments. The PTM comment names CDK2 (not CK2) as the Ser-432 kinase.
- All 68 GOA-cited publications, cached with
just fetch-gene-pmids human NBN. Most are abstract-only; full text was available for PMID:28216226, 18411307, 18583988, 19804756, 27918544, 30657944, 38961290, 26215093, 29709199, 22369660 and others.
- 94 Reactome events cached under
reactome/ (92 nucleoplasm, 2 cytosol).
- Context: telomere end-joining pathway choice (module and GO-CAM review proposed in ai4curation/ai-gene-review#5904; not on this branch and not used as evidence). The cached GO-CAM source model
gocams/65a1f4f800002884/65a1f4f800002884-src.yaml was read.
Key findings with provenance
Curation decisions worth flagging
- Protein binding (36 rows): REMOVE, except 6 rows MODIFIED to GO:0140031 phosphorylation-dependent protein binding. In those 6 the cited paper shows that binding depends on partner phosphorylation (gamma-H2AX, MDC1, RAD17).
- Damaged DNA binding: the IBA and IEA rows (enables) are marked over-annotated, because DNA binding in MRN is carried by MRE11 and RAD50. The IC row (contributes_to) is kept as non-core. The validator flags this as an action mismatch on the same term; the mismatch is deliberate, since the qualifiers make different claims.
- PELI1-derived ISS rows: K63-linked ubiquitination is removed (NBN is the PELI1 substrate, not a ligase, and does none of the ubiquitination). Positive regulation of HR (GO:1905168) is MODIFIED to GO:0000724 HR rather than removed: NBN does promote HR, but as a participant (MRE11 nuclease activation, MRN tethering), and a participant is annotated to the process rather than to its regulation. The bad PELI1 transfer is recorded in that row's propagation_review (PR #5944 review).
- NEW: GO:0170053 nuclease activator activity (IDA, PMID:30787182). Three chromatin-protein adaptor rows cite in vitro nuclease papers. They are accepted because the adaptor claim holds for the gene, and the nuclease activation they actually show is added as its own NEW row.
- GO-CAM 65a1f4f800002884 puts chromatin-protein adaptor activity on NBN with has_input TERF2. The TRF2-NBS1 structure paper describes TRF2 as the hub PMID:28216226. Under the GO:0140463 definition ("or the complex to which it belongs"), NBN docking on TRF2 to hold MRN at telomeres still fits, so the NBN row is accepted. The question is recorded in suggested_questions.
- GO:0110025 (resection at replication forks): kept as non-core. The cited IDA papers use DSB-end substrates; the fork claim rests on PMID:22565321 (fork restart).
- GO:1904354 negative regulation of telomere capping (PMID:17694070): MODIFY to GO:0032206 positive regulation of telomere maintenance, following the paper's model in which TRF1 release promotes telomerase access.
PR #5944 review follow-up
- Rule applied across the NBN and LIG4 reviews: a developmental or tissue-level process term whose only basis is the knockout phenotype of a housekeeping DNA repair factor (lethality, apoptosis, progenitor loss) records necessity, not participation, and is MARK_AS_OVER_ANNOTATED; it is kept as non-core only where the gene performs a step inside the process itself (as LIG4 does in V(D)J joining during lymphocyte differentiation). Applied to GO:0001701 and GO:0001832 (both MARK_AS_OVER_ANNOTATED).