JMY - one AP-MS row with AFF4, in HCT116 only, from a paper whose whole argument is that interactomes differ between cell lines
also called JMY · Junction-mediating and -regulatory protein · Q8N9B5
JMY does two unrelated jobs: in the nucleus it helps the tumour suppressor p53 switch genes on, and in the cytoplasm it builds actin filaments. It was found once as a co-purifying protein of AFF4, in colon cancer cells and not in the kidney cells the same study also used.
The same thing, for a scientist
JMY (UniProt **Q8N9B5, 988 aa**) is a dual-function protein: a nuclear p53/TP53 co-activator acting through p300/EP300, and a cytoplasmic nucleation-promoting factor that activates the Arp2/3 complex and also nucleates unbranched filaments through its WH2 domains [E3]. It enters this graph on **a single IntAct row**: AFF4 as tagged bait, JMY as prey, anti-tag co-immunoprecipitation, **HCT116**, spoke-expanded, MI-score 0.35, from BioPlex 3.0 alone [E1]. **The cell line is the interesting part and it cuts both ways.** BioPlex 3.0's central claim is that interactomes are extensively rewired between cell lines, and JMY is absent from the AFF4 pulldown in the HEK293T arm of the same study and from BioPlex 2.0's 293T AFF4 pulldown, which did recover PBX2 and the whole SEC [E2]. So JMY is either a genuine HCT116-specific contact - exactly what that paper exists to report - or a single-arm hit with no replication anywhere. **There is no way to tell from the data, and the entry says so rather than choosing.** JMY is not a hub (11 partners in BioPlex 3.0, 18 across all of IntAct) so promiscuity does not explain it [E2]. `JMY[tiab] AND (MLL[tiab] OR KMT2A[tiab])` returns **0**; the three leukaemia records title-check off-beat [E4]. Written `weak`. **One thing this entry does add to the graph's existing structure:** JMY's nuclear function runs through **EP300**, which this graph already carries as `gene-ep300`, and its nuclear localisation in UniProt is **ECO:0000250 - by similarity only, with no experimental support** [E3], which is a live instance of the distinction brief rule 91 asks for.
This is a prediction, not a measurement. No one has seen this protein in this shape. The colours say how much the software trusts itself, and that is worth reading before anything else — the orange stretches are places the model is effectively saying this part has no fixed shape.
I have never read one of these pictures before
A protein is one long chain of amino acids that folds up. The ribbon traces that chain from one end to the other; it is not a wire, it is a simplification of where the backbone runs. Drag to rotate, scroll to zoom.
Domains are the stretches of chain that do a particular job, and they are what the coloured blocks above the picture mark out. Click one and the view flies to it. In this project the important question about any domain is usually whether the fusion still has it — the labels say lost where it does not.
Anything in yellow sticks is a bound small molecule, which is usually the drug.
One row, one cell line, and why that is not automatically a defect
The instinct on a single AP-MS row is to discount it. Here the discount has to be argued more carefully, because the paper it comes from is the one that established cell-specific interactome remodelling as a real phenomenon: Huttlin et al. 2021 built two networks, 293T and HCT116, precisely to show that thousands of interactions are cell-line specific and that shared interactions concentrate in core complexes while cell-specific ones link those complexes.
So "present in HCT116, absent in 293T" is a result that study is designed to produce, not a contradiction. The honest position is that a single-arm hit in a two-arm study is uninterpretable on its own: it is what a real cell-specific interaction looks like and what a single false positive looks like.
What would resolve it: any AFF4 or AFF1 pulldown in a third cell type, or an HCT116 reciprocal with JMY as bait. Neither exists.
And HCT116 is a colon carcinoma line. Nothing about it is informative for a B-lineage leukaemia, and no AFF4 interactome has ever been measured in a haematopoietic cell at all - which is a larger gap than this entry.
The p53/EP300 axis, recorded without a mechanism attached
JMY's nuclear job is to increase p53-dependent transcription by working with p300/EP300, and it has been described in the p53 stress response to DNA damage. This graph carries gene-ep300, gene-crebbp and gene-tp53.
There is no experiment joining any of this to AFF4, to AFF1 or to a KMT2A fusion, and I am not proposing a mechanism. The reason to write it down is the shape of the coincidence: a co-activator that works through EP300 turning up beside an elongation scaffold is the kind of adjacency that generates a confident-sounding paragraph, and the useful thing is to record that the adjacency was noticed and that nothing supports it.
One genuine caution about the annotation itself. UniProt's Nucleus call on JMY is ECO:0000250 - inferred by similarity, while its cytoplasmic vesicle, cytoskeleton and autophagosome-membrane calls are ECO:0000269. Brief rule 91 point 4 asks for exactly this distinction: JMY's nuclear residence, which is the half of its biology that could matter here, is the half with no experimental annotation.
| uniprot | Q8N9B5 |
|---|---|
| gene symbol | JMY |
| length AA | 988 |
| alphafold ID | AF-Q8N9B5-F1 |
| protein class | WASP-family nucleation-promoting factor with WH2 domains; p53 transcriptional cofactor |
| uniprot resolution | gene_exact:JMY AND organism_id:9606 AND reviewed:true returned ONE reviewed row on 2026-09-21; primary gene name read back as JMY; entry JMY_HUMAN; 988 aa; no gene synonyms. Re-checked with gene:JMY (synonym-inclusive) - still one record, so no collision. |
| localization | Nucleus (UniProt Q8N9B5, ECO:0000250 - by similarity, NOT experimental); cytoplasmic vesicle, cytoplasm/cytoskeleton and autophagosome membrane all ECO:0000269; endomembrane system uncoded. Retrieved 2026-09-21. |
| localization evidence code | ECO:0000250 for the nuclear call; ECO:0000269 for every cytoplasmic call. The compartment relevant to an AFF4 interaction is the one with no experimental support. |
| domains note | No domain boundaries asserted; the WH2 and CA region ranges were not read back from UniProt with residue numbers this cycle. |
| shown for fusion | Nobody has looked in any fusion. The bait is wild-type AFF4 ORF in HCT116 colon carcinoma cells; there is no JMY experiment in AFF1, in any KMT2A fusion, or in any haematopoietic cell. |
domains
[]
interaction with the core complex
{
"partner": "AFF4",
"partner_accession": "Q9UHB7",
"our_protein_role": "prey",
"bait": "AFF4 (tagged ORF)",
"method": "psi-mi:\"MI:0007\"(anti tag coimmunoprecipitation)",
"interaction_type": "psi-mi:\"MI:0914\"(association)",
"expansion": "psi-mi:\"MI:1060\"(spoke expansion) - NOT a measured binary interaction",
"host_system": "Homo sapiens HCT116 colon carcinoma cell",
"source_publications": [
"33961781"
],
"distinct_experiments": 1,
"source_class": "proteome-scale AP-MS atlas (BioPlex 3.0), HCT116 arm only",
"intact_acs": [
"intact:EBI-54879646"
],
"intact_miscore": 0.35,
"recovered_in_the_293t_arm": false,
"recovered_in_bioplex_2_0": false,
"reciprocal_pulldown": false,
"region_mapped": false,
"affinity_measured": false,
"structures": []
}promiscuity controls 2026 09 21
{
"distinct_partners_in_bioplex_3_0": 11,
"distinct_partners_all_of_intact": 18,
"distinct_publications_all_of_intact": 6,
"interpretation": "not a hub; promiscuity does not explain this row"
}counted absences
{
"kmt2a": {
"query": "JMY[tiab] AND (MLL[tiab] OR KMT2A[tiab])",
"records": 0
},
"leukaemia": {
"query": "JMY[tiab] AND (leukemia[tiab] OR leukaemia[tiab])",
"records_raw": 3,
"records_on_topic": 0,
"pmids": [
"35041720",
"33683341",
"31776129"
]
},
"aliases_searched": [
"JMY"
],
"aliases_available": "none - UniProt lists no gene synonym for JMY",
"not_searched": "the protein name 'junction-mediating and regulatory protein'",
"date": "2026-09-21",
"bound": "zero in titles and abstracts; supporting figures not searched"
}adjacency recorded without a mechanism
{
"observation": "JMY's nuclear function is as a p53/TP53 co-activator acting through p300/EP300; this graph carries gene-ep300, gene-crebbp and gene-tp53.",
"verdict": "no experiment joins any of this to AFF4, AFF1 or a KMT2A fusion. Recorded so that the adjacency is found already noticed and already unsupported."
}- Nothing here supports an intervention. The risk is that 'p53 cofactor binds the elongation scaffold' is a sentence that writes itself, and the entry exists partly so that sentence has a recorded refutation attached before anyone writes it.
- JMY's cytoplasmic job is Arp2/3-dependent actin nucleation, which is required for autophagosome formation and for cell motility in normal cells. Anything that inhibited JMY would hit those functions everywhere, so even a real interaction would not make it a target.
- Is the AFF4-JMY contact genuinely HCT116-specific, or a single-arm false positive? A third cell line, or a reciprocal pulldown with JMY as bait in HCT116, decides it - and neither exists. This is the cheapest resolvable question in the whole beat.
- No AFF4 or AFF1 interactome has ever been measured in a haematopoietic cell. Every AP-MS row in this beat comes from HEK293T or HCT116, and that is a larger gap than any individual partner in it.
- Does JMY have a measured nuclear pool in a human cell? Its nuclear annotation is by similarity only, and the nucleus is the only compartment in which the reported interaction could occur.
- E1weak confidence 2 of 5: Preliminary, small, indirect, or not yet peer reviewed.database
IntAct records a single AFF4-JMY row, from BioPlex 3.0 (PMID 33961781), anti-tag coimmunoprecipitation with AFF4 as bait and JMY as prey, in HCT116 colon carcinoma cells, spoke-expanded, MI-score 0.35.
IntAct molecular interaction database, PSICQUIC MITAB 2.7, query id:Q8N9B5 AND id:Q9UHB7, 1 row retrieved 2026-09-21 · IntAct (EMBL-EBI) 2026What this evidence cannot show
One row, one publication, one arm of a two-arm study, MI-score 0.35 - the lowest of any pair in this beat. Spoke expansion means this asserts co-purification with the bait, not a binary contact. HCT116 is a colon carcinoma line; the bait is a tagged wild-type AFF4 ORF; no region, affinity, structure or reciprocal pulldown exists.
Evidence that argues against this (1)
- The same study's 293T arm, and BioPlex 2.0's 293T AFF4 pulldown, do not recover JMY while recovering the SEC and PBX2. A hit present in one of two arms of the same pipeline is either the cell-specific rewiring the paper reports or a single false positive, and nothing in the data distinguishes them. Huttlin EL et al. 2017
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
BioPlex 3.0 comprises affinity purifications of 10,128 human proteins in 293T and 5,522 immunoprecipitations in HCT116, and its stated finding is that comparison across cell lines validates thousands of interactions while revealing extensive cell-specific rewiring; JMY has 11 distinct partners in that dataset and 18 across all of IntAct, so it is not a promiscuous prey.
Dual proteome-scale networks reveal cell-specific remodeling of the human interactome. · Huttlin EL, Bruckner RJ, Navarrete-Perea J, et al., Gygi SP · Cell 2021What this evidence cannot show
Read as abstract, not full text. Partner counts computed from IntAct's import on 2026-09-21, deduplicated to distinct accessions per publication, not from the paper's supplementary tables. The paper's cell-specificity argument is made network-wide and licenses no inference about any individual single-arm hit, which is precisely the ambiguity this entry records.
Evidence that argues against this (1)
- A low prey degree removes an artefact explanation and establishes nothing positive. It is equally consistent with JMY being a rare, specific contaminant of chromatin-associated baits in HCT116. Rolland T et al. 2014
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
JMY is UniProt Q8N9B5, 988 aa, primary gene name JMY with no synonyms; UniProt describes it as both a nuclear p53/TP53 cofactor acting via p300/EP300 and a cytoplasmic nucleation-promoting factor for the Arp2/3 complex, and annotates Nucleus as ECO:0000250 (by similarity) while cytoplasmic vesicle, cytoskeleton and autophagosome membrane carry ECO:0000269.
UniProtKB entry Q8N9B5 (JMY_HUMAN, Junction-mediating and -regulatory protein), subcellular location and function comments with evidence codes · UniProtKB/Swiss-Prot 2026What this evidence cannot show
Retrieved 2026-09-21 with evidence codes read, per brief rule 91. The nuclear call is inferred from orthologue data, not from a human experiment UniProt cites, and the cited experimental papers were not opened. gene:JMY AND organism_id:9606 AND reviewed:true returns exactly one reviewed record, so there is no paralogue collision on this symbol - the brief flagged JMY as collision-shaped and it is not.
Evidence that argues against this (1)
- A protein with two spatially separated functions is the hardest case for an interactome row to interpret: an AP-MS hit cannot say which pool of JMY was captured, and the cytoplasmic pool is the one with experimental localisation support while the nuclear pool is the one that could plausibly meet AFF4. Huttlin EL et al. 2021
- E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
No literature connects JMY to KMT2A or MLL: JMY[tiab] AND (MLL[tiab] OR KMT2A[tiab]) returns 0 records, and the three leukaemia records title-check to an HTLV-1 ATLL co-expression study, an AML relapse gene-expression signature and a follicular lymphoma transformation case study.
PubMed counted-absence search for JMY against MLL, KMT2A and leukaemia, run 2026-09-21 · PubMed (NCBI) 2026What this evidence cannot show
Operand controls: JMY[tiab] alone returns 92 and JMY[tiab] AND (leukemia[tiab] OR leukaemia[tiab]) returns 3, so the instrument returns ones. Off-beat hits are PMID 35041720, 33683341 and 31776129. JMY has no UniProt gene synonyms, so the single-token search is not undercounting through aliases; it could still undercount through the protein name, which was not searched. Zero in titles and abstracts; supporting figures not searched.
Evidence that argues against this (1)
- A three-character symbol is exposed to author-initial and abbreviation collisions in the inflationary direction, so 92 is an upper bound on the real JMY literature rather than a clean count. That affects the operand control, not the zero. Huttlin EL et al. 2017
single BioPlex 3.0 AP-MS row, HCT116 arm only, spoke-expanded; absent from the 293T arm of the same study
JMY's nuclear function is as a p53 co-activator acting through p300/EP300; recorded as an adjacency with no experiment joining it to AFF4
nuclear p53/TP53 cofactor increasing p53-dependent transcription
the other nuclear call, and it is by similarity while every cytoplasmic call is experimental