start / graph / ana-q136-the-menin-inhibitor-clears-menin-off-every-canonical-target-locus-and-three-dot1l-inhibitors-do-not

AnalysisstrongQ136-menin-vs-dot1l-inhibitors

First deposit to put both drug classes on the MEN1 channel in one experiment: VTP-50469 leaves ZERO of the 19 DMSO menin peaks at 15 canonical menin-MLL target loci, while EPZ5676 keeps 14, compound 10 keeps 14 and compound 11 keeps 8 - strength-matched permutation p = 0.0003, and the zero is unchanged at every flank from 0 to 100 kb

also called GSE149183 MEN1 ChIP head-to-head · cycle 83 threat, menin channel

Two kinds of leukaemia drug were tested side by side in the same experiment, and we asked where the protein menin still sits on the DNA afterwards. The menin-blocking drug removed menin completely from all fifteen of the genes this cancer depends on, while all three of the enzyme-blocking drugs left menin sitting on most of them.

The same thing, for a scientist

GSE149183 (Perner et al., Blood 2020) is the only deposit this project has found that assays **MEN1 occupancy under a menin inhibitor and under three DOT1L inhibitors in the same experiment**: MOLM13, 4 d, DMSO / VTP-50469 / EPZ5676 / compound 10 / compound 11, GSM4491621-GSM4491625 [E1][E2]. All ten factor ChIPs were peak-called by the depositors against **one shared input BAM**, MACS2 2.1.1, BAMPE, q < 0.01, hg38 [E1]. **The site set is external and pre-specified**: fifteen canonical menin-MLL target genes (HOXA5/7/9/10, MEIS1, PBX3, MEF2C, JMJD1C, CDK6, SENP6, MYB, FLT3, BCL2, SKIDA1, MYC), with hg38 coordinates fetched from MyGene.info rather than typed, gene body +/- 5 kb. Nineteen distinct DMSO MEN1 peaks touch that set. | MEN1 arm | depth (fragments) | peaks called | DMSO target peaks kept | p, strength-matched | |---|---:|---:|---:|---:| | DMSO | 13,980,511 | 2,362 | 19/19 | - | | **VTP-50469** | 13,844,955 | 1,101 | **0/19** | **0.0003** | | EPZ5676 (pinometostat) | 15,172,064 | 1,626 | 14/19 | 0.9988 | | compound 10 | 13,577,277 | 1,779 | 14/19 | 0.9989 | | compound 11 | 12,441,785 | 1,273 | 8/19 | 0.6466 | **The MEN1 arm is depth-matched.** VTP-50469 is within **1.0%** of the DMSO library (13.84 M vs 13.98 M fragments) and the other three are within 11%, so the zero cannot be a sequencing-depth artefact [E1]. **The null is restricted with the population.** Target-locus peaks are strong peaks and strong peaks survive more often, so the null draws, for each of the 19 target peaks, a non-target DMSO peak of matched fold-enrichment (+/-15%) and asks how often it is recovered in that same treatment sample; 20,000 permutations, seed 20260923. Each arm carries its own genome-wide recovery rate into the null (VTP 0.265, EPZ 0.341, compound 10 0.333, compound 11 0.303) - so VTP's globally lower recovery is already inside the null and the 0/19 is significant **against it** [E5]. **It is not a flank artefact.** Repeating at 0, 2 kb, 5 kb, 25 kb and 100 kb flanks, the target peak count runs 15 / 18 / 19 / 23 / 23 and **VTP-50469 recovers zero at every one of them**, while EPZ5676 and compound 10 recover 11 / 13 / 14 / 16 / 16 [E5]. **VTP's MEN1 library is not dead.** It calls 1,101 peaks, 67 of them above 5-fold, and its single strongest peak (fold-enrichment 88.19) is the **strongest MEN1 peak in the whole deposit**, above DMSO's 58.49. The antibody worked; menin is simply not at the fusion's target genes [E5]. **Fold-enrichment at the individual loci**, DMSO to VTP to EPZ to compound 10 to compound 11: HOXA9/HOXA10 3.72, none, 3.30, 5.55, 3.16; MEIS1 5.66, none, 4.78, 7.35, 6.96; PBX3 6.20, none, 4.12, 7.88, 5.73; JMJD1C 4.57, none, 5.27, 6.81, 4.91; SENP6 5.12, none, 4.03, 5.24, 4.36; MYB 3.93, none, 3.40, 3.99, none; SKIDA1 4.07, none, 3.79, 3.94, 3.71 [E5].

Go deeper

What this does and does not settle about cycle 83's threat

Cycle 83's standing threat is stated about the MLL1-N channel: at MLL1's own top sites in murine MLL-AF9 the share ratio falls to 0.3012 under SGC0946 and 0.3298 under VTP-50469, so a DOT1L inhibitor moves that channel as much as a menin inhibitor does (ana-q122-curated-the-mll1-channel-is-not-inert-at-its-own-sites-and-a-dot1l-inhibitor-moves-it-as-much-as-a-menin-inhibitor).

GSE149183 contains no MLL1 ChIP and no tagged-fusion ChIP, so it cannot re-measure that channel and the threat survives as stated. What it does is test the inference drawn from it - that neither drug's effect is diagnostic of the mechanism it was chosen to test - on the channel the menin inhibitor is actually named for. On the MEN1 channel the two drug classes are completely separable: 0/19 against 14/19, 14/19 and 8/19. Menin displacement IS diagnostic of menin-interface engagement; the MLL1 channel's non-specificity is a fact about the MLL1 channel, not about the drug class.

Why compound 11 is the weakest of the three DOT1L arms, and why that is not a counter-example

Compound 11 keeps 8/19 rather than 14/19. It also has the lowest depth of any MEN1 arm (12.44 M, 89% of DMSO) and the lowest genome-wide DMSO-peak recovery of the three DOT1L arms (0.303). Against its own strength-matched null the 8/19 gives p = 0.6466 - it is within what that library recovers for peaks of that strength anywhere in the genome (the matched expectation is slightly above 8). It is not a partial menin-eviction phenotype; it is a slightly shallower library. VTP's 0/19 at p = 0.0003 is the only value that departs from its own baseline.

What no metric here can say

share(s) = absolute(s)/G is an exact identity, and every number above is a contrast, not an absolute. MACS fold-enrichment is the ChIP pileup over the same input scaled to that ChIP's own depth, so it reports signal-to-background shape. It is blind to a uniform global scaling: had menin left every site by the same factor with the library sequenced to the same depth, fold-enrichment would not move at all. The dm6 spike-in table deposited with this series covers only the H3K79me2 samples - the MEN1 and DOT1L ChIPs have no spike-in, so G is unknown and no gain/loss statement on an absolute scale is available from this deposit. What is available, and is what is claimed, is that menin's signal-to-background contrast at these fifteen loci goes from detectable to undetectable under VTP-50469 and stays detectable under all three DOT1L inhibitors.

Facts and figures

script	analyses/menin-inhibitor-against-dot1l-inhibitors-in-one-deposit/run.py
preregistration	analyses/menin-inhibitor-against-dot1l-inhibitors-in-one-deposit/PREREGISTRATION.md
status	run
geo	GSE149183
cell line	MOLM13
fusion	KMT2A::MLLT3
genome build	hg38
timepoint	4 days
replicates	1
peak caller	MACS2 2.1.1.20160309, BAMPE, q<0.01, --slocal 1500, one shared input
n target dmso men1 peaks	19
n permutations	20000
seed	20260923
vtp depth ratio to dmso	0.99
rows dropped by any filter	0
shown for fusion	KMT2A::MLLT3 in MOLM13 only. NOT shown in KMT2A::AFF1 - RS4;11 is in this same series but carries H3K79me2 only, with no VTP-50469 arm and no MEN1 ChIP.
absolute scale available	false

target peaks kept

p strength matched

genome wide recovery rate

padding invariance target peaks

padding invariance vtp kept

depth fragments

peaks called

vtp library alive check

samples

[
  "GSM4491621",
  "GSM4491622",
  "GSM4491623",
  "GSM4491624",
  "GSM4491625"
]

target peaks kept — the numbers behind the chart

{
  "VTP50469": 0,
  "EPZ5676": 14,
  "comp10": 14,
  "comp11": 8
}

p strength matched — the numbers behind the chart

{
  "VTP50469": 0.0003,
  "EPZ5676": 0.9988,
  "comp10": 0.9989,
  "comp11": 0.6466
}

genome wide recovery rate — the numbers behind the chart

{
  "VTP50469": 0.265,
  "EPZ5676": 0.341,
  "comp10": 0.333,
  "comp11": 0.303
}

padding invariance target peaks — the numbers behind the chart

{
  "0bp": 15,
  "2kb": 18,
  "5kb": 19,
  "25kb": 23,
  "100kb": 23
}

padding invariance vtp kept — the numbers behind the chart

{
  "0bp": 0,
  "2kb": 0,
  "5kb": 0,
  "25kb": 0,
  "100kb": 0
}

depth fragments — the numbers behind the chart

{
  "DMSO": 13980511,
  "VTP50469": 13844955,
  "EPZ5676": 15172064,
  "comp10": 13577277,
  "comp11": 12441785
}

peaks called — the numbers behind the chart

{
  "DMSO": 2362,
  "VTP50469": 1101,
  "EPZ5676": 1626,
  "comp10": 1779,
  "comp11": 1273
}

vtp library alive check — the numbers behind the chart

{
  "n_peaks": 1101,
  "n_peaks_fe_above_5": 67,
  "max_fold_enrichment": 88.19,
  "dmso_max_fold_enrichment": 58.49
}

How this could bite us back

If menin displacement at target loci is the pharmacodynamic endpoint, this deposit says it is achievable and drug-class-specific - but it is measured at 4 d in a cell line at a saturating exposure. A clinical menin inhibitor at trough concentration may clear menin from the strongest loci and not the weakest, and this analysis has no dose arm to say where that boundary is.
MEN1 is a broadly expressed tumour suppressor with many KMT2A-independent partners. Complete clearance of menin from 19 sites is not clearance from the genome - the VTP sample still calls 1,101 peaks - so an occupancy endpoint built on the target loci says nothing about on-target toxicity at menin's other addresses.
Compound 10 and compound 11 leaving menin in place is the selectivity claim working as designed, but it also means a DOT1L inhibitor cannot be expected to deliver the menin-eviction pharmacodynamic. Combination logic that assumes the two drug classes converge on one chromatin endpoint is not supported by this experiment.
The result rests on a single antibody (Bethyl A300-105A) in a single cell line at a single timepoint with n = 1. Everything about the effect size is provisional; only the presence/absence contrast is defended.

Still open

Does the same separation hold in the index fusion? RS4;11 (KMT2A::AFF1) is in this very series and has no MEN1 ChIP and no VTP-50469 arm. One MOLM13-sized experiment in RS4;11 would answer it, and nothing in the corpus currently does.
Is menin's disappearance at these loci eviction, or a masked epitope? A MEN1 ChIP is an antibody assay and VTP-50469 occupies the menin pocket that KMT2A's MBM1 binds; an orthogonal readout (tagged MEN1 CUT&RUN, or a second antibody against a distant epitope) would separate the two, and no deposit in this corpus does.
What is G? Every number here is a contrast because these ChIPs carry no spike-in, while the H3K79me2 samples in the SAME series do. Repeating the MEN1 ChIP with the dm6 protocol the depositors already run would convert the whole comparison to an absolute scale for the cost of one added chromatin aliquot.

Every source behind this entry (5)

E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.dataset
GSE149183 deposits MEN1 ChIP in MOLM13 at 4 d under DMSO, VTP-50469, EPZ5676, compound 10 and compound 11 (GSM4491621-GSM4491625) as MACS2 peak tables against a single shared input BAM, hg38, q<0.01, with per-sample fragment counts in the file header.
NCBI GEO GSE149183 - MEN1, DOT1L and H3K79me2 ChIP-seq in MOLM13 and RS4;11 under DMSO, VTP-50469, EPZ5676, compound 10 and compound 11; 31 samples, Homo sapiens; SubSeries of GSE149185, BioProject PRJNA627612, SRA SRP258055 · NCBI Gene Expression Omnibus 2020
What this evidence cannot show
n = 1 per condition; every factor file is REP1, so there is no within-condition variance and no noise floor can be measured in this arm. No floor is imported: the same-condition floor in this project spans ~20x across arms of a single deposit, so an imported one would be meaningless.
Evidence that argues against this (1)
- Peak counts are sensitive to library depth and to MACS's per-sample fragment-size estimate, which ranges 822-1112 bp across these ten samples. Peak count alone is therefore never interpreted here; the MEN1 arm happens to be depth-matched, which is why the MEN1 comparison is the one that carries the claim. NCBI Gene Expression Omnibus 2020
E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
The deposit accompanies Perner et al., Blood 2020;136(17):1983-1988, which reports compound 10 and compound 11 as novel DOT1L inhibitors from Novartis alongside EPZ5676.
Novel inhibitors of the histone methyltransferase DOT1L show potent antileukemic activity in patient-derived xenografts. · Perner F, Gadrey JY, Xiong Y, Hatton C, Eschle BK, Weiss A, Stauffer F, Gaul C, Tiedt R, Perry JA, Armstrong SA, Krivtsov AV · Blood 2020
PMID 32575123 DOI
What this evidence cannot show
A Blood brief report; the PubMed record carries no abstract, and the chemical structures of compound 10 and compound 11 are not given in the GEO deposit. Four authors are Novartis employees and shareholders (declared).
Evidence that argues against this (1)
- No independent group has re-measured compound 10 or compound 11; their selectivity over the menin-KMT2A interface is asserted by class membership, not shown in this deposit. Perner F et al. 2020
E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
MOLM-13 carries KMT2A::MLLT3 (MLL-AF9), not the index fusion.
Cellosaurus CVCL_2119 (MOLM-13): Sequence variation - Gene fusion KMT2A + MLLT3 (MLL-AF9); FLT3-ITD heterozygous · Cellosaurus (SIB) 2026
What this evidence cannot show
It also carries a heterozygous FLT3-ITD and retains a wild-type KMT2A allele, so a shared-epitope MLL1 antibody would not be fusion-specific here - not an issue for MEN1 ChIP, but part of the reason this deposit cannot be used to re-measure the MLL1 channel.
Evidence that argues against this (1)
- None found; the fusion is supported by three independent PubMed references in the Cellosaurus record. Cellosaurus (SIB) 2026
E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
The pivotal menin-inhibitor paper reports global loss of menin binding under VTP-50469 in human KMT2A-rearranged lines, while MLL1 is NOT globally displaced and specifically retains occupancy at the HOXA cluster, MYB and GAPDH.
A Menin-MLL Inhibitor Induces Specific Chromatin Changes and Eradicates Disease in Models of MLL-Rearranged Leukemia. · Krivtsov AV, Evans K, Gadrey JY, et al. · Cancer Cell 2019
PMID 31821784 DOI
What this evidence cannot show
A different series (GSE127508), 3 d not 4 d, and a different cell panel. Cited here as the independent prediction this analysis confirms for MEN1, not as a replicate.
Evidence that argues against this (1)
- Krivtsov et al. report MLL1n RETAINED at HOXA and MYB under VTP-50469; this analysis reports MEN1 ABSENT at HOXA9/HOXA10 and MYB under VTP-50469. These are not in conflict - they are two different channels - but the pair is the clearest statement in the corpus that menin and MLL1 part company at those loci under drug, and any reading that treats the two channels as interchangeable is wrong. Krivtsov AV et al. 2019
E5strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset
Re-analysis of the ten deposited MACS2 peak tables: VTP-50469 recovers 0/19 DMSO MEN1 peaks at the 15 canonical loci (strength-matched permutation p = 0.0003, 20,000 draws, seed 20260923) against EPZ5676 14/19, compound 10 14/19, compound 11 8/19; zero is invariant across flanks of 0, 2, 5, 25 and 100 kb; no row was dropped by any filter.
analyses/menin-inhibitor-against-dot1l-inhibitors-in-one-deposit (PREREGISTRATION.md, run.py, results.json) - this repository, re-analysis of the ten deposited GSE149183 MACS2 peak tables · Oncofusion knowledge graph 2026
What this evidence cannot show
Recovery is a binary peak-call overlap, so it inherits MACS's q<0.01 threshold; a site with real but sub-threshold menin scores as lost. The strength-matched null controls for this by asking how often peaks of the same DMSO strength are recovered elsewhere in the same library. n = 19 target peaks for MEN1 and 5 for DOT1L; the DOT1L n is small and its p is correspondingly weak (0.0432).
Evidence that argues against this (1)
- On the WHOLE DMSO peak set rather than the pre-specified loci, the same data reads the opposite way: VTP loses 73.7% of DMSO MEN1 peaks and compound 11 loses 69.6%, within 6% of each other, which scores as cycle 83's threat SURVIVING. That metric is shown to be uninformative in ana-q136-two-different-factors-in-untreated-cells-overlap-as-much-as-any-drug-pair, and the disagreement between the two metrics on one dataset is itself a result. Oncofusion knowledge graph 2026

Connected to

← refutesSearching 32 GEO series and 2,312 samples for a menin readout under DOT1L inhibition: exactly ONE distinct experiment exists, MOLM13_MEN1-ChIP_EPZ5676_4d at 100 nM, and it is the sample Q136 already used - so the deflationary reading of Q136 is blocked not by a missing assay but by a ~50-fold DOSE GAP, and Q136's drug-class conclusion rests on a single unreplicated sample in a single cell line
bounds that result: its EPZ5676 arm is the ONLY menin ChIP under a DOT1L inhibitor in 32 series and 2,312 samples - one unreplicated sample, one cell line, one dose
← supportsA full 2x2 on the MLL1 channel in one run - DMSO, EPZ5676 1 uM, VTP-50469 100 nM and both - closes cycle 83's thirty-cycle-old threat: at the fusion's strongest MLL1 sites the menin inhibitor removes 58.3% against the DOT1L inhibitor's 17.5%, a 3.33-fold separation at TEN TIMES the DOT1L dose Q136 could test, with sequencing depth running opposite to the effect in every arm
closes the half Q136 could not reach: Q136 tested the MEN1 channel at 100 nM EPZ, this tests the MLL1 channel at 1 uM and finds a 3.33-fold separation in the same direction
tests →MEN1 / menin — the adaptor the fusion cannot work without
MEN1 ChIP occupancy at 15 canonical target loci under four drugs, MOLM13 4 d
supports →Revumenib (SNDX-5613, Revuforj) — first approved menin–KMT2A interface inhibitor
VTP-50469, the preclinical analogue of revumenib, clears menin from 19/19 target peaks - the pharmacodynamic this drug class is named for
refutes →Pinometostat (EPZ-5676) — the DOT1L inhibitor, and the most instructive failure in this field
EPZ5676 (pinometostat) leaves menin on 14/19 target peaks, so a DOT1L inhibitor does NOT phenocopy menin displacement
related to →KMT2A::MLLT3 (MLL-AF9), t(9;11)(p21.3;q23.3)
the whole comparison is in MOLM13, a KMT2A::MLLT3 line
supports →The pivotal menin-inhibitor paper reports that the drug does NOT globally displace MLL1 or the fusion from chromatin - menin goes, the fusion largely stays
confirms the menin half of Krivtsov's geometry in an independent deposit: menin goes at HOXA9/HOXA10 and MYB where MLL1n stays
refutes →CURATED: Q120 called the MLL1-N channel 'inert to three decimals'. At MLL1's OWN top sites in mouse it falls to 0.3012 under a DOT1L inhibitor and 0.3298 under a menin inhibitor - so 'inert' was population-specific, and a drug that does not touch menin moves the MLL1 channel as much as one that does
the INFERENCE that neither drug is diagnostic does not hold on the MEN1 channel (0/19 vs 14/19); the MLL1-channel observation itself is untouched - this deposit has no MLL1 ChIP
related to →A DOT1L inhibitor, which does not touch the menin-KMT2A interface, removes 70% of the fusion's enrichment from its own strongest sites - two thirds of what a menin inhibitor removes - so fusion chromatin residency is partly downstream of H3K79 methylation rather than only of menin
same shape of question on a different channel: Q117 measured the FLAG-tagged fusion in murine MLL-AF9, this measures MEN1 in human MOLM13
related to →HOXA9 — required for survival of established MLL-rearranged leukemia, and dispensable for starting it
MEN1 fold-enrichment 3.72 in DMSO, no peak called under VTP-50469, retained under all three DOT1L inhibitors
related to →MEIS1 — rate-limiting for MLL leukemia stem cells, a surviving co-dependency, and embryonic-lethal when removed
MEN1 fold-enrichment 5.66 in DMSO, absent under VTP-50469, 4.78/7.35/6.96 under EPZ5676/compound 10/compound 11
related to →PBX3 - the TALE cofactor the field says is non-redundant, and the experiment that says HOXA9 is the redundant one
MEN1 fold-enrichment 6.20 in DMSO, absent under VTP-50469, retained under all three DOT1L inhibitors
related to →MYB — rank 17 of 17,728 for KMT2A selectivity, and the only node in this beat where partial, transient suppression eradicated leukaemia without stopping normal blood production
MEN1 3.93 in DMSO and absent under VTP-50469 - the locus where Krivtsov et al. report MLL1n RETAINED, so menin and MLL1 part company here
← supportsIn the same deposit, the MENIN inhibitor removes DOT1L from all five DMSO DOT1L peaks at canonical target loci (p = 0.0432) while all three DOT1L inhibitors keep 5/5 - and the VTP DOT1L library still calls ACTB, GAPDH and EEF1A1, which is the internal positive control the MEN1 arm lacks
a second, independently immunoprecipitated library gives the same 0-at-target-loci result for VTP-50469 (0/5, p = 0.0432) with an internal housekeeping control
← supportsThe menin inhibitor in this head-to-head is at 50 nM and all three DOT1L inhibitors are at 100 nM - the comparison is dose-biased AGAINST the menin inhibitor - and compound 10's target engagement at exactly 100 nM is demonstrated inside the same series by a spike-in-normalised H3K79me2 collapse of 5.6-fold by day 3
the menin inhibitor is at 50 nM against the DOT1L inhibitors' 100 nM, so dose bias runs AGAINST the headline result
← supportsThe whole-peak-set overlap metric cannot tell a MEN1 ChIP from a DOT1L ChIP in the same untreated cells (Jaccard 0.357, inside the 0.310-0.487 range of all 45 sample pairs) - so on the same data it scores cycle 83's threat as SURVIVING while the pre-specified locus set scores it as FALLING
shows the bulk-overlap metric that would have given the opposite answer cannot distinguish two antibodies in untreated cells
← related toCORRECTS A DISPATCH PREMISE: GSE149183's menin-versus-DOT1L-inhibitor head-to-head exists in MOLM13 (KMT2A::MLLT3) ONLY. RS4;11 - the index fusion - is in the series but carries H3K79me2 alone, with no VTP-50469 arm, no MEN1 ChIP and no DOT1L ChIP, and the two arms are on different genome builds
bounds the headline result to KMT2A::MLLT3; the index fusion is absent from the comparison
← related toThe deposited dm6 spike-in table covers the eighteen H3K79me2 samples and nothing else - the ten MEN1 and DOT1L ChIPs in the same series have no fly chromatin, so every occupancy number from them is a within-sample contrast and no gain-or-loss statement on an absolute scale is available
bounds every occupancy number in the headline result to a within-sample contrast - no absolute gain or loss is available
← supportsA tempting second finding that does NOT survive its control: about half of each treated sample's peaks are absent from DMSO, but the two independent VTP-50469 libraries share only 10.5% of their gained sites - against 7.7%, 8.1% and 10.6% for the DOT1L-inhibitor arms - so the gained peaks are threshold noise and there is no evidence of redistribution
the gained peaks are noise, so nothing in this deposit undercuts the loss result by reinterpreting it as relocation
← depends onResolving cycle 83's threat: the MLL1 channel moves under both drug classes and is therefore not diagnostic, but the MEN1 channel separates them completely - so 'menin displacement' remains a specific readout of menin-interface engagement and the corpus must stop treating the two channels as one
the MEN1-channel separation is the measurement this adjudication rests on
tests →Novartis 'compound 10' - a DOT1L methyltransferase inhibitor from Perner et al. (Blood 2020), active at 100 nM, whose chemical structure is not given in the GEO deposit and whose target engagement at that dose IS demonstrated in the same series by a 5.6-fold spike-in-normalised H3K79me2 collapse by day 3
compound 10 leaves menin on 14/19 target peaks at a demonstrably target-engaging dose - the arm that carries the class claim
tests →Novartis 'compound 11' - a DOT1L methyltransferase inhibitor from Perner et al. (Blood 2020), active at 100 nM, whose chemical structure is not given in the GEO deposit and whose target engagement is not measured anywhere in the series
compound 11 leaves menin on 8/19, which is exactly its own genome-wide rate for peaks of that strength (p = 0.6466)
tests →VTP-50469 - the preclinical menin-KMT2A interface inhibitor that almost every chromatin experiment in this field actually used, and which had no entry of its own in this graph
VTP-50469 at 50 nM leaves 0/19
← supportsA per-sample parameter nobody reads: MACS's fragment-size estimate varies 822-1112 bp across ten ChIPs from one sequencing run, and within the DOT1L arm peak count correlates +0.7 with fragment size and -0.8 with library depth - so the obvious confound is not the operating one, and the non-obvious one is
the MEN1 arm is the tight one on both depth (1.22x) and fragment size (1.10x), which is why the headline claim rests on it