Menin binds RPA2 directly, the contact maps to RPA2 aa 43-171 across its low-affinity ssDNA site, and three MEN1 patient missense mutations abolish it - a 2003 finding that this graph had never cited despite menin being its central node

PubMed abstract plus a targeted extraction from the PMC open-access full text for the deletion mapping, the mutation panel, the ssDNA-site statement and the replication-foci negative. The figures were not viewed and NO quantitation - no Kd, no band intensity, no n - was extracted; 'abolished', 'impaired' and 'little or no effect' are the paper's own gradings. Systems: yeast two-hybrid (two formats), GST pulldowns with purified proteins, gel filtration, HeLa co-IP and immunofluorescence. NOT haematopoietic, NO fusion in any arm, no primary tissue. 2003, before menin's crystal structure, so the deletion mapping defines determinants rather than an epitope, and the two separated determinants (1-40 and 286-448) are consistent with either a bipartite contact or with a folding requirement - the paper cannot distinguish these and nor does this entry. The expected_absent_terms field records that the ABSENCE of KMT2A/MLL from this paper is part of the claim.

• The paper contains the negative that most limits it, and the limitation is directional: menin does NOT mobilise into replication foci, so this contact is not a replication-fork event, which removes the mechanism a reader reaches for first. Two further weaknesses: the in-vitro species preference (free RPA2) and the in-cell co-precipitation (RPA1 present) point in opposite directions and the paper does not resolve which species menin binds in a nucleus; and in twenty-three years the binding itself appears never to have been independently retested - it has been restated in reviews and extended functionally, which is inheritance rather than replication. Agarwal SK et al. 2004

A 14-of-14 string match establishes that the numbering is CONSISTENT with this accession; it does not prove the paper used this convention, because an offset that happened to preserve residue identities would pass. It validates nothing about the biology of any mutation. Menin's isoform situation is the specific hazard here - a 610-aa displayed isoform with longer described isoforms - and ana-q36-menin-residue-axis is the corpus entry that holds the numbering question; it should be read before any further menin residue arithmetic. Fetched once, on one day, from one release.

• This is the weakest available form of coordinate verification and is reported as such rather than as confirmation. It cannot detect an identity-preserving offset; it says nothing about whether the paper's expression constructs matched the canonical isoform; and no paper-side statement of the reference sequence exists to check against. The honest status is 'consistent with O00255 canonical numbering', and if a later beat needs these residues for structural work it must resolve the numbering against the paper's own constructs, not against this match. This repository 2026

Abstract level only; the full text was not opened, so the binding assay is unspecified - 'binding assays indicated' is the abstract's own hedge and names no method. MCF7 and MDA-MB-231 breast cancer lines, overexpression and knockdown rather than endogenous perturbation, not haematopoietic, no fusion. This claim ALREADY EXISTS in this corpus as evidence E5 of gene-rela and is carried here because the interface now has an entry of its own; gene-rela remains the entry of record for the NF-kB axis and this item must not be counted as independent corroboration of it.

• Direction of causation is not established by this design. RPA2 is itself reported here as an E2F1 target, so it rises with proliferation - which makes 'RPA2 up, NF-kB up, tumour worse' equally consistent with RPA2 being a proliferation marker that an overexpression experiment then forces into a functional-looking role. And a breast-epithelial NF-kB axis is a long way from infant B-ALL; this corpus's own menin-NF-kB entries are careful about that distance and this item should inherit their caution rather than borrow their confidence. Heppner C et al. 2001

PREDICTION ONLY - homology modelling, docking, GROMACS MD and MM/PBSA, with no experiment of any kind. Graded `speculative` for that reason and not because the work is poor. Abstract level; the predicted menin residues are NOT recorded in this entry because the abstract does not name them and the full text was not read, so importing 'two predicted sites' as coordinates would be fabrication. The authors' own framing is that the experimental approaches are 'expensive, time-consuming, and challenging', i.e. that this substitutes for an experiment that has not been done. Note the mutation studied, S606F, is NOT in the Sukhodolets panel and lies beyond the 448 truncation point that retained binding, so it is not obviously a test of the mapped determinants. MM/PBSA absolute binding free energies are not comparable to measured affinities and should never be quoted as Kd-equivalent.

• This paper predicts an interface that the founding experimental paper had already mapped by deletion to menin aa 1-40 plus 286-448 and to RPA2 aa 43-171, and the abstract does not state whether its predicted sites agree with that mapping. A computational interface prediction that does not reproduce the existing experimental constraint before extending it cannot be scored, and this entry therefore uses it only as evidence that the interface has attracted no further experimental work - which is itself the informative point. Sukhodolets KE et al. 2003

Bounded by titles and abstracts (rule 80). The RPA32 spelling returns ZERO against menin, so a name list built from the kDa nomenclature would have missed this entire literature - the relevant name list actually tried was RPA2, RPA32, 'replication protein A', menin and MEN1, and NOT tried were RPA34, REPA2, RFA2 on the RPA side. The six-record menin list is the same under RPA2[tiab] AND MEN1[tiab], which is a useful consistency check but not an independent one.

• A six-record literature with two genuinely independent experimental papers fourteen years apart is thin, and quoting 'six papers link menin and RPA2' would overstate it. Stripped to independent experimental work on the contact, the count is two - one 2003 paper from the NIH MEN1 group and one 2017 paper from unrelated Taiwanese groups - plus one computational study and three papers that inherit the claim. Nobody has retested the binding itself. Dilley WG et al. 2005

The OB-fold coordinates are HUMAN, from the fetched P15927 record, and the overlap with 43-171 is simple arithmetic on those numbers. The allosteric-coupling mechanism is NOT human - PMID 23070815 crystallised a USTILAGO MAYDIS heterotrimer (MeSH 'Fungal Proteins', 'Ustilago'; no 'Humans' term) - so the mechanism is imported and no residue from that paper is used. Graded `moderate` for that reason: the overlap is a fact, the functional reading of the overlap is a cross-species inference. The DBD-A/DBD-B nomenclature is RPA1's, not RPA2's, so the low-affinity protein-binding surface named in the coupling model is not itself the menin-binding surface - the inference is about the complex's state, not about which residues menin touches.

• The founding paper tested the natural consequence of this overlap and it was negative: menin had no detectable effect on RPA-DNA binding in vitro. So the overlap, on the only evidence that exists, has no functional consequence for DNA binding - and this item exists as much to record that negative as to record the overlap. Any future claim that menin modulates RPA's ssDNA engagement has to get past a 2003 experiment that looked and found nothing. Sukhodolets KE et al. 2003