Re-bound at full text, where methods live: 20 records pair a WRAD subunit with a proximity-labelling method and a KMT2A term, 14 pair one with a tiling-CRISPR term, and none is a proximity, degron or tiling study of WRAD in a KMT2A-fusion line
also called Q115 WRAD re-bound
Whether the four-protein module that partners the fusion has ever been mapped by proximity labelling, degraded on a timer, or scanned domain by domain in this disease is a question about methods, and methods are described in the body of a paper rather than its summary. Searching bodies finds no such study.
The same thing, for a scientist
This is the abstract-level absence with the strongest prior of being overturned at full text, because **BioID, dTAG and tiling-CRISPR designs are named in Methods and almost never in an abstract**. Run: `(WDR5 OR RBBP5 OR ASH2L OR DPY30) AND (BioID OR TurboID OR "proximity labeling" OR "proximity labelling" OR APEX2 OR miniTurbo)` returns **74**; adding a seven-term KMT2A clause gives **20**; the degron form with the KMT2A clause gives **92**; `(WRAD) AND ("tiling CRISPR" OR "CRISPR tiling" OR "tiling screen" OR "domain-focused")` gives **14** [E1]. Operands pass (WRAD alone 5,307; proximity methods alone 16,966). **All 20 and all 14 titles were read. None is the study the entry says does not exist.** The near misses are informative and all three are adjacent rather than on target: **PMID 39922194** (*Mol Cell* 2025) uses condensate proteomics on **NUP98** oncofusions, not KMT2A; **PMID 40341256** (*Leukemia* 2025) does domain-focused work on the **SETD1B** catalytic domain in MLL-rearranged leukaemia, which is a COMPASS-family catalytic subunit and not a WRAD subunit; **PMID 40830799** (*Genome Biol* 2025) places **KAT6A** chimeras with NURF and MLL/COMPASS [E2]. The entry's own caveat already anticipated the first of these.
Why the full-text run was the right test and what it cost
The entry's proximity query returned 0 at abstract level. At full text the same subunits against the same methods return 74, and 20 once scoped to KMT2A. That is a seventy-four-fold change in recall and the verdict did not change - which is the most useful thing a re-bound can report, because it converts a weak zero into a screened zero.
The degron arm is the one that is still not properly bounded. WRAD x degron vocabulary x KMT2A returns 92 and only the first 60 were retrieved. The WDR5 degrader chemistry literature is large and the entry already says its 15 abstract-level hits were all WDR5 degrader chemistry with no KMT2A-fusion-association design. This beat did not read the 92, so that arm remains an upper bound rather than a screened zero.
The bounded rejections
Read and put down, with the reason:
- PMID 39922194, The phenylalanine-and-glycine repeats of NUP98 oncofusions form condensates that selectively partition transcriptional coactivators, Mol Cell 2025 - a proximity/partition study of an oncofusion that recovers coactivators, but the fusion is NUP98 and the design is condensate partitioning, not WRAD-in-KMT2A.
- PMID 40341256, Regulation of H3K4me3 breadth and MYC expression by the SETD1B catalytic domain in MLL-rearranged leukemia, Leukemia 2025 - domain-focused, in the right disease, on the wrong subunit.
- PMID 35499757, MOZ and Menin-MLL Complexes Are Complementary Regulators of Chromatin Association and Transcriptional Output in Gastrointestinal Stromal Tumor, Cancer Discov 2022 - menin-MLL complex chromatin work, wrong tissue and not WRAD-directed.
- PMID 34555496, Defining Proximity Proteome of Histone Modifications by Antibody-mediated Protein A-APEX2 Labeling - the method paper, mark-directed rather than subunit-directed.
- PMID 38989615, BOD1L mediates chromatin binding and non-canonical function of H3K4 methyltransferase SETD1A - appears in both the proximity and tiling sets; SETD1A, not WRAD.
| script | none - Europe PMC BODY: queries, 2026-09-22 |
|---|---|
| status | run |
| verdict | SURVIVES for the proximity and tiling arms (screened zeros); the degron arm remains an UPPER BOUND, unscreened |
| entry re bounded | absence-no-proximity-degron-or-crispr-study-of-wrad-in-kmt2a-fusion-leukaemia |
| instrument before | three abstract-level method-word queries; the arm most likely to be overturned because methods do not reach abstracts |
| instrument after | four full-text intersections; 20 and 14 read in full, 92 not screened |
| recall change | the proximity query moved from 0 records at abstract level to 74 at full text, and the verdict did not change |
| absence class | never-measured |
| absence class basis | UNCHANGED. No archive was queried for this one and none obviously applies: a proximity-labelling experiment that was never performed leaves no deposit. If PMID 39922194 turns out to include a KMT2A-fusion arm, its ProteomeXchange deposit would change the class to measured-not-searched, and that check was not run. |
queries — the numbers behind the chart
[
{
"q": "WRAD group AND proximity methods",
"hits": 74,
"read": 60,
"screened": false
},
{
"q": "WRAD group AND proximity methods AND KMT2A clause",
"hits": 20,
"read": 20,
"on_topic": 0,
"proceedings": 5
},
{
"q": "WRAD group AND degron vocabulary AND KMT2A clause",
"hits": 92,
"read": 60,
"screened": false,
"note": "UPPER BOUND ONLY - this arm is not re-bounded"
},
{
"q": "WRAD group AND tiling-CRISPR vocabulary",
"hits": 14,
"read": 14,
"on_topic": 0,
"proceedings": 1
},
{
"q": "WRAD group alone",
"hits": 5307,
"role": "operand control"
},
{
"q": "proximity methods alone",
"hits": 16966,
"role": "operand control"
}
]wrad subunits
[ "WDR5", "RBBP5", "ASH2L", "DPY30" ]
proximity terms
[ "BioID", "TurboID", "proximity labeling", "proximity labelling", "APEX2", "miniTurbo" ]
proximity terms NOT tried
[ "BASU", "AirID", "split-TurboID", "PUP-IT", "biotin ligase" ]
degron terms
[ "dTAG", "auxin-inducible degron", "degron", "PROTAC", "degrader" ]
degron terms NOT tried
[ "IAA7", "FKBP12F36V", "SMASh", "AID2" ]
bounded rejections
[
{
"pmid": "39922194",
"why": "NUP98 oncofusion condensate proteomics, not KMT2A, not WRAD-directed - and the most likely refutation"
},
{
"pmid": "40341256",
"why": "domain-focused work in MLL-rearranged leukaemia but on SETD1B, a catalytic subunit, not WRAD"
},
{
"pmid": "40830799",
"why": "KAT6A chimeras with NURF and MLL/COMPASS - a different chimera"
},
{
"pmid": "35499757",
"why": "menin-MLL complex chromatin association in GIST - wrong tissue"
},
{
"pmid": "34555496",
"why": "Protein A-APEX2 proximity proteome of histone MARKS, not of a subunit"
},
{
"pmid": "38989615",
"why": "BOD1L and SETD1A, appears in both the proximity and tiling sets, not WRAD"
}
]- A WDR5 degrader is already in the chemistry literature. If nobody has ever degraded a WRAD subunit in a KMT2A-fusion line, then the assumption that WRAD matters for the FUSION rather than for wild-type KMT2A has never been tested with the one tool that could separate them - and a clinical WDR5 programme would inherit that assumption untested.
- Does PMID 39922194 include a KMT2A-fusion comparator, and does its coactivator partition list contain WDR5, RBBP5 or ASH2L? One read decides whether the proximity arm of this absence stands.
- The 92-record degron intersection was never screened. Reading those titles is the unfinished half of this re-bound and is an afternoon's work.
- Would BASU, AirID or split-TurboID find anything the six-term list missed? The method-word list is the bound and it is short.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
Europe PMC full text returns 74 records for WRAD subunits crossed with proximity-labelling methods, 20 when scoped to a KMT2A clause, 92 for the degron form scoped to KMT2A and 14 for the tiling-CRISPR form; operands are 5,307 for the WRAD group and 16,966 for the proximity methods; no title in the 20 or the 14 is a proximity, degron or tiling study of a WRAD subunit in a KMT2A-fusion line.
Europe PMC RESTful search, BODY: WRAD subunits x proximity/degron/tiling methods x KMT2A clause, 2026-09-22 · Europe PMC 2026What this evidence cannot show
Europe PMC BODY: reaches only the 12,362,145 full texts the index holds. A congress abstract book is one document containing thousands of unrelated abstracts and therefore matches almost any conjunction; across this beat's 18 fully-read intersections 83 of 321 records (25.9%, range 0-50%) are proceedings. Screening was at TITLE level unless a paper is listed as opened. The 92-record degron arm was retrieved only to 60 records and was NOT title-screened, so it is an upper bound and not a screened zero; that arm of the absence is therefore weaker than the proximity and tiling arms. Five of the 20 proximity hits and one of the 14 tiling hits are congress abstract books.
Evidence that argues against this (1)
- The method name list is the bound. A study using BASU, AirID, split-TurboID, PUP-IT, a SpyTag proximity system or simply 'biotin ligase' would not match; nor would a degron study writing 'IAA7', 'FKBP12F36V' or 'SMASh' without the word degron. Rule 70 applies to a method-word list exactly as it does to a noncoding-element name list, and this one has six proximity terms. Europe PMC 2026
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
The three nearest records are a NUP98-oncofusion condensate proteomics study, a SETD1B catalytic-domain study in MLL-rearranged leukaemia, and a KAT6A chimera study placing it with NURF and MLL/COMPASS - each adjacent on one axis and off on another.
The phenylalanine-and-glycine repeats of NUP98 oncofusions form condensates that selectively partition transcriptional coactivators. · Molecular Cell 2025What this evidence cannot show
TITLE-CHECKED ONLY; none of the three was opened. Titles were taken from full Europe PMC records rather than truncated listings. The disqualifying feature quoted for each is stated in its own title.
Evidence that argues against this (1)
- PMID 39922194 is the single most likely refutation: an oncofusion condensate proteomics paper that recovers transcriptional coactivators may well recover WDR5, RBBP5 or ASH2L, and if it also includes a KMT2A-fusion comparator arm the entry's proximity claim falls. The entry's own caveat already flags a NUP98-PHF23 BioID study as invisible to its abstract query; this may be the same observation or a newer one, and neither has been read. Molecular Cell 2025
- supports →No proximity, degron or tiling-CRISPR study has ever been run on WRAD in KMT2A-fusion leukaemia
SURVIVES for the proximity and tiling arms after a 74-fold recall increase; the degron arm is an unscreened upper bound of 92
- related to →WDR5 — the partner that is not a partner: it binds the half of KMT2A the fusion throws away
no proximity, degron or tiling-CRISPR study of this subunit in a KMT2A-fusion line at full text
answers that entry's first open question - whether WRAD matters for the fusion at all - with the finding that the experiment that would decide it still does not exist