Run a fusion-specific and a total-AFF1 readout side by side in the same KMT2A::AFF1 nuclei, and report the two target sets rather than one
Measure where the fusion protein is using a method that can only see the fusion, and at the same time with the usual antibody that also sees the normal partner protein, in the same cells. The difference between the two answers is the thing nobody has measured.
The same thing, for a scientist
**Design.** In a KMT2A::AFF1 line with an endogenously tagged fusion allele (tag knocked into the KMT2A 5' exon so that only the fusion and the wild-type N-terminal fragment carry it, or into the AFF1 3' side of the junction so that only the fusion carries it), run in the same nuclei preparation, on the same day, with the same protocol: (1) anti-tag CUT&RUN, (2) anti-AFF1 CUT&RUN with **both** MyBioSource MBS190886 and Thermo PA5-77068, (3) KMT2A N-terminal and C-terminal CUT&RUN, (4) ENL, (5) IgG, (6) **a spike-in**, because this deposit's central limitation is that every number in it is a within-sample share. **Primary readout.** The fraction of total-AFF1 peaks that carry the tag, and the fraction of tagged peaks that carry total AFF1. Stated as two numbers, not one - they are different questions and the corpus conflates them. **Secondary readout.** Repeat the ENL-versus-AFF1 discrimination contrast measured here in primary material (1.90-2.78 versus 1.03-1.45) in the tagged line, to see whether the line reproduces the primary ordering. If it does not, the line is the wrong model for this question and that is worth knowing before any further cell-line occupancy work. **Controls that this deposit lacked and that make the experiment worth doing.** A spike-in; a third AFF1 reagent or a knockdown arm; matched IgG in every batch; and both antibodies reported separately at every stage rather than pooled. **Cost tier: medium.** The tagged allele is the expensive part and tagged KMT2A::AFF1 lines already exist in several laboratories.
The primary-sample version, which is harder and more valuable
No tag is possible in uncultured patient material, so the fusion-specific arm there has to be either a junction-spanning antibody (none is commercially available for KMT2A::AFF1) or the N/C ratio, which is what GSE252378 already provides. The realistic primary-sample experiment is therefore: add AFF1, ENL, DOT1L, CDK9, ELL and menin to the existing N/C panel in the seven banked KMT2A::AFF1 infants in GSE252378, with a spike-in. That is a reagent addition to a protocol already validated on this exact material by this exact group.
| design | Same-nuclei comparison of a fusion-specific (tag or N/C) readout against total-AFF1 with two antibodies, plus ENL, DOT1L, IgG and a spike-in, in a tagged KMT2A::AFF1 line; primary-sample version adds the same panel to the seven banked KMT2A::AFF1 infants in GSE252378. |
|---|---|
| readout | Fraction of total-AFF1 peaks carrying the fusion-specific mark, and fraction of fusion-specific peaks carrying total AFF1, reported as two numbers; plus the ENL/AFF1 discrimination contrast. |
| controls | Spike-in; IgG in every batch; both AFF1 antibodies reported separately; KMT2A N and C in the same nuclei. |
| cost tier | med |
| timeline | 3-6 months in a line; longer for the primary-sample panel |
- If the experiment shows most AFF1 peaks are fusion-free, every AFF1-derived target list in the field shrinks, including lists used to nominate therapeutic targets; the inconvenience of that result is a reason to pre-register the readout.
- Which existing tagged KMT2A::AFF1 line carries the tag on the AFF1 side of the junction rather than the KMT2A side? Only the former is fusion-specific.
- E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset
The measurement this experiment would make does not exist: no dataset separates fusion-bound from free AFF1 on chromatin in any KMT2A::AFF1 system.
GSE252378 - AutoCUT&RUN and AutoCUT&Tag profiles of 36 MLL-rearranged leukemia samples (171 GSMs, hg38) · Janssens DH, Henikoff S (Fred Hutchinson Cancer Center / Van Andel Institute) · NCBI Gene Expression Omnibus 2024What this evidence cannot show
Bounded by this deposit and by the source paper, not by an exhaustive literature search. A tagged-allele AFF1 comparison could exist in a paper not reached here.
Evidence that argues against this (1)
- Degron studies of MLL::AF9 with HA-tagged fusion alleles do separate fusion from partner in a related system, and a reader could argue the question is already answered in principle by those; the answer would then be imported across fusion partners, which is the transfer this corpus repeatedly finds unsafe. Janssens DH et al. 2024
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset
In primary infant KMT2A::AFF1 material ENL discriminates fusion from wild-type sites at 1.90-2.78-fold and AFF1 at 1.03-1.45, giving the secondary readout its pre-specified comparison values.
GSE252378 - AutoCUT&RUN and AutoCUT&Tag profiles of 36 MLL-rearranged leukemia samples (171 GSMs, hg38) · Janssens DH, Henikoff S (Fred Hutchinson Cancer Center / Van Andel Institute) · NCBI Gene Expression Omnibus 2024What this evidence cannot show
Promoter-resolution, within-sample shares, two patients.
Evidence that argues against this (1)
- If the tagged line reproduces neither value, the experiment cannot distinguish 'the line is a poor model' from 'the promoter-window design was the wrong instrument'; a peak-level reanalysis of GSE252378 should be run first so the comparison is like-for-like. Janssens DH et al. 2024
separates fusion-bound from free AFF1 in the same nuclei
- related to →KMT2A::AFF1 (MLL-AF4), t(4;11)(q21;q23)
fusion-specific occupancy readout
the primary-sample version closes this absence in the same run