KMT2A-KRR1 is one spoke-expanded BioPlex 3.0 AP-MS row in HEK293T with KRR1 as bait and KMT2A as prey, sitting in a seven-member ribosomal block at the floor score - declined
also called KMT2A-KRR1 · MLL-KRR1 · HRB2 · EBI-54612384
A database lists KRR1, a protein that helps build ribosomes in the nucleolus, as a KMT2A partner. The experiment tagged KRR1 and pulled it out of kidney cells, and KMT2A was one of the things that came along - as it did for six other ribosome proteins in the very same screen.
The same thing, for a scientist
**Verdict: DECLINED. Detection method is anti-tag co-immunoprecipitation (AP-MS) in BioPlex 3.0, spoke-expanded, with KRR1 as bait and KMT2A as prey** [E1]. One IntAct record, EBI-54612384, PMID 33961781, HEK293T, intact-miscore 0.35. KRR1 is the best-annotated of the three nucleolar candidates in this beat: UniProt Q13601 gives Nucleus/nucleolus at **ECO:0000269** and an explicit small-subunit (SSU) processome function [E3], so unlike ABT1 its nucleolar identity is not itself an inference. **The decline rests on a different fact, and it is the one the dispatched hypothesis asked for: KRR1 is not an isolated nucleolar hit, it is one member of a seven-protein ribosome and nucleolus block inside a single screen.** Of the 24 BioPlex baits that recovered KMT2A, seven are ribosomal or ribosome-biogenesis proteins - ABT1, KRR1, RPL36, RPLP0, RPS9, Rpl13a and the RPL3/RPL3L group - plus PIH1D1 of the R2TP snoRNP-assembly chaperone, and **all eight sit at the 0.35 floor score while WDR5, RBBP5 and ASH2L sit at 0.96, 0.95 and 0.91** [E1, E4]. A block of ribosomal proteins arriving together at the bottom of one screen's score distribution is the textbook signature of affinity-purification background, not of a complex.
The seven-member ribosomal block, which is the actual measurement in this entry
The beat's hypothesis was that ABT1, KRR1 and UTP3 - three nucleolar ribosome-biogenesis factors on two different baits - might indicate real nucleolar biology for KMT2A and PSIP1. The pre-registered discriminator required a base rate before any shared-screen observation could count.
Pulling the base rate turned the n from 3 into 7, in one dataset. The 24 BioPlex baits that recovered KMT2A as prey, with live intact-miscores 1:
| group | baits | miscore |
|---|---|---|
| genuine MLL1 complex | WDR5, RBBP5, ASH2L | 0.96, 0.95, 0.91 |
| CK2 holoenzyme + HCF | CSNK2A2, CSNK2B, HCFC2, CSNK2A1 | 0.53-0.79 |
| ribosome / nucleolus | ABT1, KRR1, RPL36, RPLP0, RPS9, Rpl13a, RPL3/RPL3L group (+PIH1D1, R2TP) | 0.35, all of them |
| splicing factors | SRSF4, SRSF5, SRSF6 | 0.35 |
| secreted / extracellular | ANOS1, FGFBP1 | 0.35 |
| other | ASB7, OASL, PRKRA, SET | 0.35 |
Two things follow. First, the nucleolar 'triple' is not three proteins on two baits; it is a seven-member ribosomal block inside one screen plus one BioID hit on the other bait. Second, the same floor score contains ANOS1 and FGFBP1, both experimentally annotated as secreted 4 - pairs that cannot be physiological, sitting at the identical score, in the identical experiment. That internal negative control is what makes the block interpretable rather than merely unproven.
What would rescue KRR1, and why none of it exists
Reciprocal confirmation is structurally unavailable. All 31 BioPlex rows touching Q03164 have KMT2A as prey 1 - the 3,969-residue protein is not among the 10,128 baits - so the strongest of the three checks the BioPlex authors name was never run and cannot be run on the deposited data 2.
Replication in the second cell line did not happen. This record is HEK293T only; unlike FGFBP1, it does not appear in the HCT116 arm 1.
No orthogonal method exists. The pair has one publication and one detection method in all of IntAct 1.
And there is a compartment argument that runs the other way, which is worth stating because it is the honest counter. KRR1 is not implausible in a nucleus: it is nuclear and nucleolar with experimental codes, and KMT2A's rDNA and nucleolar biology is genuinely unmeasured in this project rather than excluded. The decline is not 'KRR1 cannot be near KMT2A'; it is that this record cannot distinguish that from background, and its six ribosomal companions say which way to bet.
| partner a | gene-kmt2a |
|---|---|
| partner a uniprot | Q03164 |
| partner a length AA | 3969 |
| partner b | KRR1 |
| partner b uniprot | Q13601 |
| partner b length AA | 381 |
| verdict | DECLINED |
| adjudicated by | Q129, cycle 92, 2026-09-23 |
| source publication pmid | 33961781 |
| source dataset | BioPlex 3.0 |
| detection method | anti-tag co-immunoprecipitation (AP-MS) |
| experimental role partner | bait |
| experimental role kmt2a | prey |
| kmt2a was never the bait | true |
| expansion method | spoke expansion |
| intact miscore | 0.35 |
| replicated in second bioplex line | false |
| reciprocally confirmed | false |
| reciprocal confirmation possible | false |
| n publications | 1 |
| n methods | 1 |
| orthogonal method exists | false |
| partner nucleolar evidence code | ECO:0000269 (experimental) |
| intact row count partner | 464 |
| druggability | Not assessable. No residue, domain or region is mapped on either side, and no structure of the pair exists. |
| absence class | never-measured |
| absence class note | No experiment has tested KMT2A against KRR1 with KMT2A as the bait, in any cell, and none has tested the pair in a haematopoietic cell. Closing it needs wet work. A secondary, cheaper gap is measured-not-analysed: nucleolar-fraction proteomics of KMT2A-rearranged leukaemia cells would say whether KMT2A has any nucleolar pool at all, and this project has never asked that question of any deposited dataset. |
| shown for fusion | NO - nobody has looked in any KMT2A fusion. Full-length KMT2A as a prey in embryonic kidney cells; no fusion protein present. |
| epigenetic readout | none - no chromatin, histone-mark, rDNA or occupancy measurement exists for this pair. |
| nucleolar hypothesis role | The best-annotated of the three nucleolar candidates; adjudicated together with ABT1 and UTP3 in ana-q129-the-nucleolar-triple-survives-none-of-the-four-pre-registered-discriminators. |
host cell lines
[ "HEK293T embryonic kidney" ]
records
[ "EBI-54612384" ]
ribosomal block in same screen
[ "ABT1", "KRR1", "RPL36", "RPLP0", "RPS9", "Rpl13a", "RPL3/RPL3L group", "PIH1D1 (R2TP, counted separately)" ]
interface residues a
[]
structures
[]
ligands
[ "drug-revumenib", "drug-ziftomenib" ]
- Ribosome biogenesis is required by every dividing cell. If KRR1 or the SSU processome were a genuine KMT2A dependency, targeting it would have essentially no therapeutic window against normal haematopoietic progenitors, which is the same objection that sinks most ribosome-directed proposals in leukaemia.
- Declining this row does not close the nucleolus. If KMT2A does work in the nucleolus, every menin-inhibitor readout in current use - chromatin occupancy, H3K79me2, HOXA9/MEIS1 expression - is blind to it, and a nucleolar escape route would present clinically as resistance with no assay pointing at it.
- Does KMT2A have a nucleolar pool in any cell? This graph carries subnuclear localisation as an explicit PI dimension and has never measured or searched for a nucleolar KMT2A signal; the question is entirely open in both directions.
- Would a nucleolar-fraction proteomic dataset from a KMT2A-rearranged leukaemia line, already deposited somewhere in PRIDE, settle the compartment question without any wet work? No archive query was run in this beat.
- Are the seven ribosomal BioPlex baits that recovered KMT2A also recovering other large nuclear proteins at the same rate? That is the control that would convert this entry's block argument from a pattern into a measurement, and it needs the full BioPlex prey matrix rather than IntAct.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
IntAct curates one record of a KMT2A-KRR1 association, EBI-54612384, from PMID 33961781 (BioPlex 3.0), annotated to Supplementary table 1. Detection method anti-tag co-immunoprecipitation, interaction type 'association', expansion method SPOKE EXPANSION, host Homo sapiens HEK293T embryonic kidney cell. KRR1 (Q13601) is the BAIT and KMT2A (Q03164) is the PREY. intact-miscore 0.35. Of the 24 distinct BioPlex baits that recover KMT2A as prey, the miscores are WDR5 0.96, RBBP5 0.95, ASH2L 0.91, CSNK2A2 0.79, CSNK2B 0.64, HCFC2 0.62, CSNK2A1 0.53, and 0.35 for all seventeen others including ABT1, KRR1, RPL36, RPLP0, RPS9, Rpl13a, the RPL3/RPL3L group, PIH1D1, ANOS1, FGFBP1, OASL, PRKRA, SET, ASB7, SRSF4, SRSF5 and SRSF6. All 31 BioPlex rows touching Q03164 have KMT2A as prey.
IntAct molecular interaction database, REST findInteractions on Q03164 (KMT2A), 376 records retrieved 2026-09-23 by Q129 · IntAct (EMBL-EBI) 2026What this evidence cannot show
Read live from IntAct REST on 2026-09-23, not from PSICQUIC (rule 93). The BioPlex supplementary table was not opened, so the row rests on IntAct's curation of it. intact-miscore is a pair-level score driven mainly by how many publications and methods support a pair, so 0.35 largely restates 'one record of one kind' rather than grading this experiment; it is quoted to make the stratification legible, not as an independent opinion. The ribosomal/nucleolar grouping of the seven baits is this entry's classification of them, made from their UniProt function lines, and another reader could draw the boundary differently - PIH1D1 in particular is a chaperone for snoRNP and RNA-polymerase assembly and is counted separately above.
Evidence that argues against this (1)
- The floor score is shared by SET, SRSF4, SRSF5 and SRSF6, and SET is a genuine nucleosome-assembly and histone-chaperone protein whose association with a chromatin methyltransferase is not absurd. So 0.35 is not a synonym for 'false' even within this list, and the ribosomal block argument is a statement about which reading is better supported, not a proof. IntAct (EMBL-EBI) 2026
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
In human HEK293T and HCT116 epithelial cell lines, BioPlex 3.0 affinity-purified 10,128 tagged human proteins giving 118,162 interactions among 14,586 proteins, plus 5,522 immunoprecipitations in HCT116, scored by CompPASS-Plus against a GFP control at 1% protein-level FDR with additional entropy-based dataset-level filtering. The paper states that about 60% of interactions in both networks are confirmed by replication, reciprocal recovery or co-fractionation, and that proteins average just 8 partners.
Dual proteome-scale networks reveal cell-specific remodeling of the human interactome. · Huttlin EL, Bruckner RJ, Navarrete-Perea J, Cannon JR, Baltier K, Gebreab F, et al. (28 authors), Gygi SP · Cell 2021What this evidence cannot show
Full text read 2026-09-23 through the PMC HTML route (PMC8165030, 221,127 stripped characters) after Europe PMC fullTextXML returned HTTP 500. The strings 'KMT2A' and 'nucleolar' appear zero times in the paper's text, so it does not report this interaction in prose; the row exists only in a supplementary table that was not opened. The paper's one mention of the ribosome is in a bait-selection note - replicate baits were chosen so that large complexes such as the ribosome and proteasome would not be over-represented - which concerns bait sampling, not prey background, and is not evidence about this row either way.
Evidence that argues against this (1)
- BioPlex recovered WDR5, RBBP5 and ASH2L on KMT2A at 0.96, 0.95 and 0.91, so this dataset demonstrably sees true KMT2A biology and the screen is not being dismissed wholesale. Any general argument that AP-MS cannot be trusted would also delete the MLL1 complex from this graph. IntAct (EMBL-EBI) 2026
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
UniProtKB Q13601 is 381 aa, primary gene name KRR1, recommended name 'KRR1 small subunit processome component homolog'. Subcellular location comments: Nucleus, nucleolus [ECO:0000269]; Nucleus [ECO:0000269]; Cytoplasm [ECO:0000269]. Function: part of the small subunit (SSU) processome, the first precursor of the small eukaryotic ribosomal subunit, assembled in the nucleolus on nascent pre-rRNA. KMT2A (Q03164) is 3,969 aa with one ECO:0000269 Nucleus comment plus two uncoded comments for the N320 and C180 taspase1 cleavage products.
UniProtKB entry Q13601 (KRR1): primary gene name, sequence length, subcellular location and function, retrieved 2026-09-23 · UniProt Consortium 2026What this evidence cannot show
UniProt location comments are curated pointers, not measurements (rule 91). The cited localisation papers were not opened. Note that this evidence item argues FOR the plausibility of the pair's compartment, not against it - KRR1's nucleolar annotation is experimentally coded, which is why the decline rests on the screen-level block rather than on compartment.
Evidence that argues against this (1)
- KRR1's compartment is compatible with KMT2A's, and this project has never measured whether KMT2A has a nucleolar pool. UniProt's KMT2A record carries three location comments of which two are uncoded, so the absence of a nucleolar line for KMT2A is an absence in the annotation, not a measured exclusion. UniProt Consortium 2026
- E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
Two of the 24 BioPlex baits that recovered KMT2A at the same 0.35 floor score are experimentally annotated as extracellular proteins: ANOS1 (P23352, 680 aa) is Cell membrane and Secreted at ECO:0000269, and FGFBP1 (Q14512, 234 aa) is Secreted/extracellular space and Cell membrane at ECO:0000269. KRR1 carries 464 IntAct interaction records in total.
UniProtKB entries P23352 (ANOS1) and Q14512 (FGFBP1): subcellular location with evidence codes, retrieved 2026-09-23 · UniProt Consortium 2026What this evidence cannot show
Using two implausible pairs in the same score band as an internal negative control assumes that the score band is homogeneous in reliability, which BioPlex does not claim - the intact-miscore is IntAct's, not BioPlex's, and BioPlex's own per-pair CompPASS-Plus probabilities were not retrieved. A stronger version of this control would compare the authors' own scores for KRR1, ANOS1 and FGFBP1.
Evidence that argues against this (1)
- Secreted proteins pass through the ER and can be detected in whole-cell lysate AP-MS without being biologically paired, but they are also over-expressed from a tag construct here, which can mislocalise a protein and create genuine-but-artefactual cytoplasmic pools. So 'secreted therefore impossible' is slightly too strong; the correct statement is that an interaction with an over-expressed secreted protein in a lysate is uninterpretable, which is also the position taken for KRR1. Huttlin EL et al. 2021
- related to →KMT2A / MLL1 — the gene that breaks
adjudicates the IntAct KMT2A-KRR1 row; DECLINED - one of seven ribosomal baits at the 0.35 floor in one screen
same screen, same floor score, same ribosome-biogenesis block; ABT1's nucleolar annotation is ECO:0000250 where KRR1's is ECO:0000269
internal negative control: ANOS1 sits at the identical 0.35 floor score, in the identical screen, as the nucleolar candidates
adjudicates KRR1 as two visible members of a seven-member ribosomal block, all at the 0.35 floor against WRAD at 0.91-0.96