H3K4me3 and wild-type MLL1 are the negative controls for the partition, and both come out flat
If splitting genes into 'fusion sat here' and 'fusion did not' made every measurement look different, the split would be telling us about the split and not about the fusion. Two things that should not change when the fusion is destroyed - a histone mark laid down by the normal protein, and the normal protein itself - do not change, which is what makes the menin result mean something.
The same thing, for a scientist
Negative controls for the HA-occupancy partition, all at 3 h after 500 nM dTAG-VHL. **H3K4me3**: median dVHL/DMSO TSS ratio 0.974 at the authors' 200 target genes vs 1.047 at the bottom half of HA occupancy - a gap of 7.3 points that becomes **-2.6 points** under matching on H3K4me3's own DMSO signal decile (6 deciles used). **Wild-type MLL1, C-terminal antibody** (which does not recognise the fusion): 0.703 occupied vs 0.625 non-occupied - the gap is **negative** (-7.9 points raw, -5.5 matched, 6 deciles). Neither control moves with fusion occupancy. Against these, MENIN's gap is 47.1 raw / 28.5 matched and DOT1L's is 45.5 raw / 47.0 matched. The partition is therefore not a generic property of the stratification.
Why these two and not others
H3K4me3 is deposited by wild-type MLL1 and the SET domain is lost from the fusion, so it should not respond to fusion degradation - the authors make the same argument (Figure S4E) 1. The MLL1 C-terminal antibody recognises only wild-type MLL1, so it directly tests whether a compensatory wild-type protein is redistributing 2. The N-terminal MLL1 antibody recognises both proteins and behaves like a partial fusion channel (0.479 occupied vs 0.888 non-occupied), which is the expected intermediate and is a useful internal check on antibody specificity rather than a control.
What a flat control rules out and what it does not
It rules out the explanation that any channel measured at fusion-occupied genes will show a stratum gap because those genes are busy. It does not rule out the narrower version of that explanation confined to high-signal channels; for that, see the signal-decile matching entry, where MENIN's gap shrinks by 39% and the controls go to zero or below.
Scale and limits
No spike-in exists anywhere in GSE173599, so every number is a within-sample share, and the treated arms are scaled by a 21-gene housekeeping panel whose own raw 3 h H3K27ac ratio averages 0.809 and is forced to 1. A housekeeping normalisation DEFINES the global change to be zero, so only the DIFFERENCE between strata is measurable; a uniform genome-wide loss of any of these proteins would be invisible here. n = 1 per arm for every channel except HA, which has two independent 3 h batches. The cells are HCB1, transformed by KMT2A::MLLT3 (MLL::AF9) - NOT the index fusion of this project, KMT2A::AFF1 - and a prior beat in this corpus found that per-gene ordering transfers between MLLT3 and AFF1 lines while magnitude does not. The MLL1 C-terminal sheet contains 1,249 genes and only 69 of the authors' 200 target genes survive into it (34.5%), so that control's occupied arm is small.
| script | analyses/menin-retention-by-ha-occupancy/analyze_controls.py |
|---|---|
| status | run |
| result summary | Negative controls flat: H3K4me3 raw gap +0.073 / matched -0.026; wild-type MLL1 C-terminal raw -0.079 / matched -0.055. MENIN raw +0.467 / matched +0.285; DOT1L raw +0.455 / matched +0.470. |
| peak width ratio note | Signal instrument. The one peak-based figure in this family, H3K79me2, has a treated/control median peak width ratio of 0.96, below the 1.3 trigger. |
inputs
[ "GSE173599_TSS_gene_body_signal_Figure3.xlsx", "GSE173599_ChIPseq_TSS_Figure3_revisions.xlsx" ]
- A drug that removes menin from chromatin removes it from wild-type MLL1 complexes too - menin-MLL1 is the natural interaction the inhibitor blocks - so an intervention validated against a flat wild-type-MLL1 control under DEGRADATION says nothing about the wild-type liability under menin INHIBITION, where the drug targets the shared interface.
- Does H3K4me3 at fusion targets change by 24 h or 4 d, when the transcriptional programme has collapsed? The deposit has a 24 h arm (0.908 at target genes) but no later timepoint for this mark.
- Would an MLL1 C-terminal ChIP with more sequencing depth recover the 131 target genes its sheet currently lacks, and does the negative gap survive there?
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
H3K4me3 at promoters does not change 3 h after MLL::AF9 degradation at either target or non-target genes.
MLL::AF9 degradation induces rapid changes in transcriptional elongation and subsequent loss of an active chromatin landscape · Olsen SN, Godfrey L, Healy JP, Choi YA, Kai Y, Hatton C, Perner F, Haarer EL, Nabet B, Yuan GC, Armstrong SA · Molecular Cell 82(6):1140-1155.e11 2022What this evidence cannot show
The authors state this as an argument against wild-type MLL1 compensation; it is n=1 per arm and housekeeping-normalised, like every other channel here.
Evidence that argues against this (1)
- H3K4me3 is a slow-turnover mark; a 3 h window may be too short to see a change even if the depositing activity were lost, so its flatness is weaker evidence about MLL1 activity than about the partition being non-generic. Olsen SN et al. 2022
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset
Recomputed from the deposited per-gene tables, H3K4me3 gives a 7.3-point raw and -2.6-point matched stratum gap, and wild-type MLL1 (C-terminal) gives -7.9 raw and -5.5 matched.
GSE173599_ChIPseq_TSS_Figure3_revisions.xlsx - per-gene MLL1 N-terminal and C-terminal TSS signal, DMSO vs dTAG-VHL 3 h, plus the 21-gene housekeeping normalisation panel and the published MLL::AF9 target-gene lists · Olsen SN et al. · NCBI GEO supplementary file 2022What this evidence cannot show
1,249 genes in the MLL1 C-terminal sheet; 69 of the 200 target genes survive the authors' floor. n=1 per arm. The MLL1 C-terminal library was sequenced in a separate batch (210928) from every other 3 h channel.
Evidence that argues against this (1)
- A flat control at 3 h does not exclude a change at a later timepoint; H3K4me3 falls to 0.908 at target genes by 24 h in the same deposit, so 'flat' is timepoint-specific. Olsen SN et al. 2022
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset
MENIN and DOT1L, measured identically in the same partition, give raw gaps of 47.1 and 45.5 points.
GSE173599_TSS_gene_body_signal_Figure3.xlsx - per-gene housekeeping-normalised TSS and gene-body ChIP signal, DMSO vs dTAG-VHL 3 h and 24 h, 16 channels · Olsen SN et al. · NCBI GEO supplementary file 2022What this evidence cannot show
Same normalisation and same n=1 limitation as the controls; the comparison between channels is the informative quantity, not any single level.
Evidence that argues against this (1)
- MENIN and DOT1L are also the two channels whose sheets are most aggressively floored, so part of their larger gap could reflect a narrower and higher-signal gene set rather than a larger biological effect. Olsen SN et al. 2022
two pre-registered negative controls are flat, so the stratum gap is not a generic property of the partition
- related to →KMT2A / MLL1 — the gene that breaks
wild-type MLL1 C-terminal channel does not redistribute after fusion degradation
- related to →Plain-English: chromatin and epigenetics — the sticky notes on the DNA, and which ones matter here
H3K4me3 at promoters is unchanged 3 h after fusion degradation in both strata