Re-bound in the matching instrument: the Antibody Registry holds no junction-specific antibody for any KMT2A fusion under five spellings, while the same query template returns the synovial sarcoma clone as a positive control and turns up two leukaemia-fusion junction reagents nobody here had found
also called Q115 junction antibody re-bound
The claim that no antibody exists against the join of a KMT2A fusion protein was based on a search of paper summaries, which is the wrong place to look for a laboratory reagent. Searching the catalogue of antibodies instead confirms the claim and, in the same two minutes, turns up two such antibodies for other leukaemia fusions.
The same thing, for a scientist
The cycle-65 census named this entry as an instrument mismatch: *a reagent's existence is a catalogue fact* and the instrument used was PubMed `[tiab]`. Run in the catalogue. **The Antibody Registry returns 0 records for `MLL-AF4`, `MLL-AF9`, `MLL-ENL`, `KMT2A-MLLT3` and `KMT2A-AFF1`, while `KMT2A` and `MLL` return many antibodies all directed at the protein itself** [E1]. The positive control travels the identical `/api/fts-antibodies?q=` template: `SS18-SSX` returns exactly one record, **RRID:AB_3665243, the E9X9V clone this entry already cites**, and the negative control `NOTAREALANTIGENXYZ` returns 0 [E1]. **The absence survives, now bounded by the instrument that can see the object.** Two findings came out of the same sweep. `junction specific` returns **RRID:AB_490962**, *Bcr-Abl (b2a2 Junction Specific)*, clone L99H4, western blot only; and `ETV6-RUNX1` returns **RRID:AB_3713372**, clone TEL 6F2, raised against a 12-mer spanning the fusion site and used for ChIP-on-chip [E2]. Both are leukaemia fusions and neither is cited anywhere in this graph. The full-text arm adds a clean zero: `BODY:` on the three antibody phrases intersected with a seven-term KMT2A clause returns **3 records, all soft-tissue-tumour diagnostics**, and the phrase `junction epitope` intersected with KMT2A returns **0** [E3].
What the entry should now say
Its headline is unchanged and is now better supported. Two sentences in it need amendment, and both are in the summary_lay and summary_expert:
- "One exists for a different cancer - synovial sarcoma" - two exist for leukaemia fusions, and one of them is a paediatric B-ALL fusion antibody validated for ChIP. That is a materially stronger argument for the entry's own recommendation.
- The implied technical objection is breakpoint diversity. The ETV6::RUNX1 precedent shows a 12-mer junction peptide was a sufficient immunogen; this entry's own computation puts the KMT2A::AFF1 requirement at six 16-mers sharing an eight-residue acceptor. Nothing about the immunogen is harder here.
The reagent facts are written up separately in concept-junction-specific-antibody-to-a-leukaemia-fusion-tel-6f2; this entry is the re-bound of the absence and does not restate them.
The catalogue has its own trap and it is worth recording
The registry's full-text search tokenises, so short alphanumeric strings are unusable as operands: q=b2a2 returned 25 records dominated by names containing [Biotin] and clone 2B4, none of them BCR-ABL. The junction reagent was found through q=junction specific and q=BCR-ABL, not through the transcript code.
So a zero from this endpoint is only meaningful for a token the tokeniser preserves, and every returned name must be read. The five KMT2A-fusion spellings tested are all hyphenated multi-token strings of the same shape as SS18-SSX, which returned its record - which is why the zeros are interpretable.
| script | none - Antibody Registry API and Europe PMC BODY: queries, 2026-09-22 |
|---|---|
| status | run |
| verdict | SURVIVES, and the instrument mismatch named by the cycle-65 census is now closed |
| entry re bounded | absence-no-junction-specific-antibody-exists-for-any-kmt2a-fusion |
| instrument before | PubMed [tiab] - cannot see a catalogue object |
| instrument after | SciCrunch Antibody Registry /api/fts-antibodies, plus Europe PMC full text |
| absence class | never-measured |
| absence class basis | UNCHANGED, and now bounded in the catalogue rather than the literature. No such reagent exists, so nothing could have been measured with it. The ETV6::RUNX1 precedent means the class is never-measured for want of a campaign, not for want of a feasible design. |
| absence class assigned | 2026-09-22 (Q115), re-affirming the 2026-09-21 assignment |
registry queries — the numbers behind the chart
[
{
"q": "MLL-AF4",
"records": 0
},
{
"q": "MLL-AF9",
"records": 0
},
{
"q": "MLL-ENL",
"records": 0
},
{
"q": "KMT2A-MLLT3",
"records": 0
},
{
"q": "KMT2A-AFF1",
"records": 0
},
{
"q": "KMT2A",
"records": "many, all anti-KMT2A/MLL protein"
},
{
"q": "MLL",
"records": "many, all anti-MLL protein"
},
{
"q": "SS18-SSX",
"records": 1,
"role": "POSITIVE CONTROL - same URL template",
"found": "RRID:AB_3665243 clone E9X9V"
},
{
"q": "NOTAREALANTIGENXYZ",
"records": 0,
"role": "NEGATIVE CONTROL - same URL template"
},
{
"q": "junction specific",
"records": 4,
"on_topic": 1,
"found": "RRID:AB_490962 Bcr-Abl b2a2"
},
{
"q": "ETV6-RUNX1",
"records": 1,
"found": "RRID:AB_3713372 clone TEL 6F2"
},
{
"q": "EWSR1-FLI1",
"records": 0
},
{
"q": "PML-RARA",
"records": 2,
"on_topic": 0,
"note": "both are anti-PRAM1, a PML-RARA TARGET GENE product, not the fusion"
},
{
"q": "NPM-ALK",
"records": 17,
"on_topic": 0,
"note": "all are anti-ALK or anti-NPM reagents used to detect the fusion, none junction-raised"
},
{
"q": "b2a2",
"records": 25,
"on_topic": 0,
"note": "TOKENISATION TRAP: matched [Biotin] and clone 2B4 names; the BCR-ABL record is not among them"
}
]full text queries — the numbers behind the chart
[
{
"q": "(fusion-specific OR breakpoint-specific OR junction-specific OR neoepitope) antibody phrases",
"hits": 234
},
{
"q": "those phrases AND a 7-term KMT2A clause",
"hits": 3,
"on_topic": 0
},
{
"q": "KMT2A clause AND junction-epitope language",
"hits": 0
},
{
"q": "BODY:\"SS18-SSX\" AND BODY:\"fusion-specific antibody\"",
"hits": 37,
"role": "positive control"
},
{
"q": "BODY:\"QTYSNEVH\" (the computed shared acceptor octamer)",
"hits": 0
}
]endpoint notes
[ "/api/antibodies/search -> HTTP 422; /api/antibodies (no params) -> HTTP 400; the working search is /api/fts-antibodies?q=&page=&size=", "the human-facing /AB_<id> and /antibody/AB_<id> pages return HTTP 200 with a 1,147-byte JS shell and no record content" ]
- A reagent that exists only as a non-commercial hybridoma is a single point of failure. The ETV6::RUNX1 clone is thirteen years old, held in one institute, with zero recorded citations - so 'it has been done' and 'it can be obtained' are different claims, and a programme that assumes the first will discover the second late.
- Would a KMT2A::AFF1 junction 16-mer raise a usable monoclonal? The ETV6::RUNX1 campaign is the protocol and the six epitopes are already computed; nobody has tried, and a failed immunisation would itself be worth publishing.
- Does PMID 34750023, a review of IHC surrogates for fusion oncoproteins in soft tissue, tabulate junction antibodies by fusion? It is the one place a missed KMT2A reagent would be listed and it was not opened.
- How many junction-specific reagents exist across all fusion cancers? The registry can be swept systematically by fusion name and nobody has done it; the answer would tell this field whether it is behind or whether nobody does this.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
The Antibody Registry full-text search returns 0 records for MLL-AF4, MLL-AF9, MLL-ENL, KMT2A-MLLT3 and KMT2A-AFF1; many anti-KMT2A/MLL antibodies for KMT2A and MLL; exactly one record for SS18-SSX (RRID:AB_3665243, clone E9X9V) as the positive control; and 0 for the nonsense negative control.
Antibody Registry full-text search endpoint /api/fts-antibodies, queries run 2026-09-22 · SciCrunch Antibody Registry 2026What this evidence cannot show
The endpoint was located by reading /api/openapi.json, because /api/antibodies/search returns HTTP 422 and /api/antibodies without parameters returns HTTP 400 - two wrong-looking failures that are parameter errors, not empty results. The response object carries no `total` field, so a zero here is 'no items returned on page 1 of size 25', which for these queries is the whole result. The registry indexes vendor and community submissions; a reagent made in a laboratory and never registered would be invisible.
Evidence that argues against this (1)
- TEL 6F2 is itself the proof that a non-commercial hybridoma CAN be in this registry, which strengthens the zeros - but it was registered because someone deposited it, and the ETV6::RUNX1 field is small. A KMT2A junction antibody sitting unregistered in a laboratory freezer is exactly the object this instrument cannot see, and PMID 24121163's own abstract shows such a reagent can be described in print without the word 'junction' appearing anywhere in the paper. SciCrunch Antibody Registry 2026
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
The same sweep returns two junction-specific reagents against leukaemia fusions: RRID:AB_490962 (Bcr-Abl b2a2 junction, clone L99H4, western blot only) and RRID:AB_3713372 (ETV6-RUNX1, clone TEL 6F2, FACS/WB/ChIP, defining citation PMID 24121163).
Antibody Registry records 490962 and 3713372 · SciCrunch Antibody Registry 2026What this evidence cannot show
Both records were fetched individually through /api/antibodies/{id} and read field by field. The human-facing pages for these records return HTTP 200 with a 1,147-byte client-rendered shell containing neither reagent name, so the API is the only route that returns content - a 200 that is not an answer, the same family as the 13-character DOI redirect stub.
Evidence that argues against this (1)
- Neither record carries an `epitope` field, so 'junction-specific' rests on the product NAME for AB_490962 and on the defining paper's methods for AB_3713372. A vendor name is a marketing claim; only the ETV6::RUNX1 case has a published immunisation design behind it. Blood Cancer Journal 2013
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
In Europe PMC full text, the three antibody phrases intersected with a seven-term KMT2A clause return 3 records - all soft-tissue-tumour diagnostic papers - and a KMT2A clause intersected with junction-epitope language returns 0, against 234 for the antibody phrases alone and 37 for the SS18-SSX positive control.
What this evidence cannot show
Europe PMC BODY: reaches only the 12,362,145 full texts the index holds, so a paywalled paper is still invisible; and a congress abstract book is indexed as one document containing thousands of unrelated abstracts, so a fraction of every hit list here is proceedings (measured across this beat's 18 fully-read intersections: 83 of 321 records, 25.9%, range 0-50%). Screening was at TITLE level unless a paper is listed as opened. The three hits are PMID 34750023 (Semin Diagn Pathol, IHC as a surrogate for fusion oncoproteins in soft tissue), PMID 36983010 (Int J Mol Sci, molecular diagnosis of soft tissue tumours) and PMID 32744500 (eLife, CHD4 in rhabdomyosarcoma); none was opened.
Evidence that argues against this (1)
- PMID 34750023 is a review of immunohistochemistry as a surrogate for fusion oncoproteins across soft tissue tumours. If it tabulates junction antibodies by fusion, it is the single best place a KMT2A entry would appear and it was not opened. Its being soft-tissue-scoped is a reason to expect no KMT2A row, not evidence of one. Seminars in Diagnostic Pathology 2022
SURVIVES in the Antibody Registry under five fusion spellings with the E9X9V clone as a same-template positive control; two leukaemia-fusion junction reagents found in the process
the reagent this re-bound found; written up separately rather than folded into the absence