start / graph / ana-q116-the-h3k9-repressive-module-converges-on-the-kmt2a-promoter-in-mcf7-and-none-of-it-is-at-the-breakpoint

AnalysismoderateQ116-h3k9-axis-at-the-locus

THE POSITIVE RESULT: the writer G9a, the reader HP1a and the scaffold KAP1 all sit on the KMT2A PROMOTER in MCF-7 at the 87.6th-98.9th percentile of 2,000 ACTIVE promoters - surviving the control that says an active promoter is ChIP-hot for everything - while none of the three is above the 75th percentile at the exon-11 anchor 52 kb away

also called the promoter module · G9a + HP1a + KAP1 at the KMT2A promoter · the H3K9 machinery is at the wrong end of the gene

Three proteins that work together to silence DNA all sit on the switch that turns KMT2A on, more strongly than they sit on almost any other switched-on gene. None of them is at the spot 52 kilobases away where the gene breaks in this leukaemia.

The same thing, for a scientist

**The claim.** In MCF-7, **EHMT2/G9a** (the H3K9me1/me2 writer), **CBX5/HP1a** (the H3K9me3 reader) and **TRIM28/KAP1** (the corepressor scaffold that binds HP1) all occupy the *KMT2A* promoter well above what an active promoter gets by default, and **none of the three is at the exon-11 CTCF anchor** [E1]. **The control that had to be passed first, and nearly was not.** Against 56,171 same-cell-type CTCF summit windows, **every protein track in this deposit reads 87.6-99.9 at the *KMT2A* promoter** and the whole-panel median at that interval is **96.87** over 28 C1-passing tracks. That is the signature of an active promoter being ChIP-hot for everything, and it made the CTCF-anchor percentile useless for this question. The pre-registered C5 panel baseline caught it; a third pass was added (deviation D9) asking the right question instead: **is the *KMT2A* promoter high AMONG ACTIVE PROMOTERS?** [E2] **It is, for all three, and by a clear margin:** | antigen | condition | *KMT2A* promoter mean | **percentile of 2,000 highest-expressed genes' promoters** | x median active promoter | |---|---|---:|---:|---:| | EHMT2/G9a | -E2 | 0.9704 | **87.60** | 1.44x | | EHMT2/G9a | +E2 | 0.7584 | **88.65** | 1.34x | | CBX5/HP1a | -E2 | 1.3309 | **91.05** | 1.53x | | CBX5/HP1a | +E2 | 1.1970 | **94.50** | 1.71x | | **TRIM28/KAP1** | -E2 rep1 | 3.2128 | **98.90** | **3.02x** | | **TRIM28/KAP1** | -E2 rep2 | 4.1610 | **98.80** | **3.85x** | | **TRIM28/KAP1** | +E2 | 3.1699 | **99.35** | **2.98x** | **TRIM28 is the strongest and the most reproducible: two replicates of the -E2 condition agree to 0.10 percentile points (98.90 and 98.80) and the +E2 track gives 99.35 - three tracks, two conditions, one replicate pair, all above the 98.8th percentile of active promoters and all near 3x the median active promoter.** **AND THE NULL DISCRIMINATES - the proof is KDM4B.** Against CTCF anchors KDM4B reads **99.37** at this promoter, indistinguishable from TRIM28's 99.80. Against **active promoters** it reads **58.60** and **1.05x** the median active promoter - background. **The two nulls disagree completely about one antigen, and the active-promoter null is the one that separates the panel:** TRIM28 98.8-99.4 > CBX5 91.1-94.5 > EHMT2 87.6-88.7 > KDM4C 78.4 > **KDM4B 58.6** > H3K9me3 0.0-61.1. So this is not a claim that everything in the deposit sits at the *KMT2A* promoter; it is a claim about the three antigens at the top of that ordering [E2]. **And it is not just signal.** The deposit's two HOMER peak files call a **KAP1 peak directly on the promoter in both hormone conditions**, and that peak overlaps by **267 bp** the ENCODE TRIM28 conservative-IDR peak `ana-q110-h3k9-writer-and-reader-occupancy-at-the-kmt2a-locus` found in **K562** [E3]. **Nothing is at the breakpoint.** At the 146 bp exon-11 anchor the same tracks give: G9a **43.01 / 23.83 / 8.36 / 69.15** (median 33.42), CBX5 **47.89 / 0.00**, TRIM28 **90.62 / 0.00**. By the pre-registered rule all three are **ORDINARY or DEPLETED**, none is PRESENT, and the nearest KAP1 called peak to the exon-11 CTCF summit is **52,230 bp** away [E1][E3]. **So the H3K9 repressive machinery at this locus is at the promoter, in two cell lineages, and the breakpoint 52 kb away has none of it.** That is the first coherent positive statement this project has about where the H3K9 layer actually is at *KMT2A*, and it points away from the element four beats have been aiming at.

Go deeper

Why this is a module and not three separate facts

The three proteins are a known physical unit and the deposit's own paper shows it: mass spectrometry in these same MCF-7 cells identifies KAP1 as a major HP1-interacting partner 4. HP1a's chromodomain reads H3K9me2/me3, the mark G9a's first two methylation steps create, and KAP1 is the scaffold that couples HP1 to the silencing machinery. Finding the writer, the reader and the scaffold at the same 2 kb, each in the top decile of active promoters, is one observation about one complex, not three coincidences - although see the counter-evidence: the co-occupancy base rate for these three across all promoters was not computed.

And the two lineages agree on the address to within a few hundred base pairs. Q110's K562 numbers: G9a's strongest KMT2A-region ENCODE peak at chr11:118,434,740-118,435,115 (signalValue 218.5), three more bracketing the TSS, TRIM28 at chr11:118,436,029-118,436,453. MCF-7's KAP1 peak: chr11:118,436,186-118,436,464. Erythroleukaemia and breast carcinoma, two labs, two peak callers, the same promoter.

The exon-11 anchor, and what four beats have now established there

protein	cell type	verdict at the 146 bp exon-11 window	source
SUV39H1	MCF-7	NEITHER (median 79.93, but the whole gene body is 82-87)	Q114
SUV39H1	SU-DIPG13 / 36	NOT OCCUPIED (explicit 0.0000 and 74.17th)	Q114
SETDB1	K562	no called peak in the gene or within 50 kb, 3 peak sets, 2 labs	Q110
CBX5, CBX1, CBX3	K562, GM12878	no called peak in the gene or within 50 kb	Q110
EHMT2/G9a	MCF-7	ORDINARY, median 33.42, 0 of 4 tracks above the 75th	this beat
CBX5/HP1a	MCF-7	ORDINARY, 47.89 and 0.00	this beat
TRIM28/KAP1	MCF-7	ORDINARY at signal level; ZERO called peaks, nearest 52,230 bp away	this beat
H3K9me3	MCF-7	DEPLETED - an explicit 0.0000 in five of six tracks	Q114, this beat

Five proteins, five cell types, two assay formats, three labs. Nothing in the H3K9 layer is at this element above the 75th percentile of same-cell-type CTCF anchors, anywhere it has been looked for. That is now a strong negative and it should change what gets commissioned next.

Facts and figures

script	analyses/q116-h3k9-axis-at-the-locus/run116.py, promoter_specificity.py and kap1_peaks.py
prereg	analyses/q116-h3k9-axis-at-the-locus/PREREGISTRATION.md, section 8 and deviations D8 and D9
status	run
run date	2026-09-23
run by	Q116-h3k9-axis-at-the-locus
complete run	true
assembly	GRCh38 throughout; each bedGraph's build established from its own maximum coordinate per chromosome in both directions before any interval was read
cross cell type agreement	K562 (Q110, ENCODE peak level): G9a strongest KMT2A-region peak chr11:118,434,740-118,435,115, TRIM28 chr11:118,436,029-118,436,453. MCF-7 (this beat): KAP1 peak chr11:118,436,186-118,436,464, overlapping by 267 bp.
localization	chromatin at chr11q23.3 - the KMT2A promoter (chr11:118,435,456-118,437,456 and TSS +/- 500 bp), against the promoters of the 2,000 highest-expressed genes in the same cells; and the 146 bp exon-11 CTCF anchor 52 kb away
subnuclear localization	not measured
cell type	MCF-7 (ER+ breast adenocarcinoma) for the signal; K562 (CML erythroleukaemia) for the peak-level replication. NEITHER is KMT2A-rearranged and neither is a progenitor.
cell state	hormone-deprived culture and 40 min 17-beta-oestradiol. Oestradiol changes none of these promoter numbers materially.
epigenetic readout	co-occupancy of the H3K9 writer (EHMT2), reader (CBX5) and scaffold (TRIM28) at the KMT2A promoter and at the exon-11 CTCF anchor
writer of that mark	EHMT2/G9a for H3K9me1/me2
shown for fusion	NO - neither cell line carries a KMT2A fusion, and no H3K9-machinery occupancy has ever been measured at this locus in a KMT2A-rearranged cell of any lineage
absence class	measured-not-analysed
absence class note	Every track used here was deposited publicly in 2019 or 2021. The promoter module was never reported by the depositors, whose paper is about oestrogen-responsive promoters and never mentions KMT2A.
trim28 is the strongest and most reproducible	98.90, 98.80 and 99.35 percentile of 2,000 active promoters across three tracks and two conditions, with the two -E2 replicates agreeing to 0.10 percentile points - the tightest replicate agreement anywhere in this beat, against exon-11 replicate spreads of 47.9 to 90.6 points.
panel median provenance	C5/C5b PANEL MEDIANS ARE QUOTED AT THE FINAL 28-TRACK COUNT (2026-09-23). Earlier drafts of this entry quoted them at intermediate counts (12, 18 or 25 tracks) while the run was still going. The gene-body panel median moved 61.80 -> 61.39; the PROMOTER panel median moved 82.36 -> 96.87, i.e. UP, and the direction matters: at 96.87 the CTCF-anchor percentile at the KMT2A promoter is even less informative than the first draft said, and the promoter claim rests ENTIRELY on the active-promoter null of deviation D9.
active promoter null discriminates	THE ACTIVE-PROMOTER NULL DISCRIMINATES, AND THE PROOF IS KDM4B. Against 56,171 CTCF summit windows KDM4B reads the 99.37th percentile at the KMT2A promoter - indistinguishable from TRIM28's 99.80 and above the 96.87 panel median. Against the promoters of the 2,000 highest-expressed MCF-7 genes it reads the 58.60th percentile and 1.05x the median active promoter, i.e. BACKGROUND. So the two nulls disagree completely about one antigen, the active-promoter null is the one that separates the panel, and the ordering it gives is TRIM28 98.8-99.4 > CBX5 91.1-94.5 > EHMT2 87.6-88.7 > KDM4C 78.4 > KDM4B 58.6 > H3K9me3 0.0-61.1. The module claim rests on the three antigens at the top of that ordering and is not a claim that everything in the deposit is at this promoter.

results

[
  "analyses/q116-h3k9-axis-at-the-locus/results.json",
  "analyses/q116-h3k9-axis-at-the-locus/results_promoter_specificity.json",
  "analyses/q116-h3k9-axis-at-the-locus/results_scored.json",
  "analyses/q116-h3k9-axis-at-the-locus/results_kap1_peaks.json"
]

promoter specificity the statistic that decides this

{
  "window": "TSS +/- 500 bp; KMT2A TSS 118,436,463 (+), GENCODE v29",
  "null": "the promoters of the 2,000 highest-expressed MCF-7 protein-coding genes (ENCFF251DYL, TPM 63.74-10,242.87)",
  "EHMT2_minusE2": {
    "mean": 0.9704,
    "pct_of_active_promoters": 87.6,
    "x_median_active": 1.44
  },
  "EHMT2_plusE2": {
    "mean": 0.7584,
    "pct_of_active_promoters": 88.65,
    "x_median_active": 1.34
  },
  "CBX5_minusE2": {
    "mean": 1.3309,
    "pct_of_active_promoters": 91.05,
    "x_median_active": 1.53
  },
  "CBX5_plusE2": {
    "mean": 1.197,
    "pct_of_active_promoters": 94.5,
    "x_median_active": 1.71
  },
  "TRIM28_minusE2": {
    "mean": 3.2128,
    "pct_of_active_promoters": 98.9,
    "x_median_active": 3.02
  },
  "TRIM28_minusE2_rep2": {
    "mean": 4.161,
    "pct_of_active_promoters": 98.8,
    "x_median_active": 3.85
  },
  "TRIM28_plusE2": {
    "mean": 3.1699,
    "pct_of_active_promoters": 99.35,
    "x_median_active": 2.98
  },
  "plus1_nucleosome_pct_of_active_promoters": {
    "EHMT2_minusE2": 80,
    "EHMT2_plusE2": 69.45,
    "CBX5_minusE2": 60.3,
    "CBX5_plusE2": 52.05,
    "TRIM28_minusE2_rep1": 93.9,
    "TRIM28_minusE2_rep2": 97.55,
    "TRIM28_plusE2": 57.5
  },
  "KDM4C_minusE2": {
    "pct_of_active_promoters": 78.35,
    "x_median_active": 1.61
  },
  "KDM4B_minusE2": {
    "pct_of_active_promoters": 58.6,
    "x_median_active": 1.05,
    "note": "BACKGROUND on this null, against 99.37 on the CTCF-anchor null - the discriminating case"
  },
  "H3K9me3_for_contrast": {
    "pct_of_active_promoters_shCTL": [
      56.7,
      0
    ],
    "pct_of_active_promoters_shKDM4B_shKDM4C": [
      20.55,
      61.05
    ],
    "note": "the mark itself is not elevated at this promoter on this null"
  }
}

C5b panel baseline that forced it

{
  "kmt2a_promoter_percentile_of_ctcf_anchors_all_protein_tracks": "87.55 to 99.91",
  "panel_median_all_C1_passing_tracks": 96.87,
  "reading": "the CTCF-anchor percentile cannot distinguish a protein address from an active promoter in this deposit"
}

exon11 verdicts

{
  "EHMT2": {
    "percentiles": [
      43.01,
      23.83,
      8.36,
      69.15
    ],
    "median": 33.42,
    "verdict": "ORDINARY"
  },
  "CBX5": {
    "percentiles": [
      47.89,
      0
    ],
    "verdict": "ORDINARY"
  },
  "TRIM28": {
    "percentiles": [
      90.62,
      0
    ],
    "verdict": "ORDINARY at signal level; ZERO called peaks, nearest 52,230 bp away"
  }
}

C5 final 28 track values

{
  "panel_median_kmt2a_gene_percentile": 61.39,
  "panel_median_kmt2a_promoter_percentile": 96.87,
  "per_antigen_median_promoter_percentile_vs_CTCF_anchors": {
    "TRIM28": 99.8,
    "CBX5_mut4E": 99.75,
    "KDM4B": 99.37,
    "KDM4C": 98.47,
    "CBX5_mut4A": 96.76,
    "CBX5": 96.65,
    "EHMT2": 95.83,
    "H3K9me3": 29.43
  },
  "reading": "every PROTEIN antigen reads 95.8-99.8 at the KMT2A promoter against CTCF anchors and only the histone mark (29.43) does not - so that statistic cannot discriminate between proteins at all, and every promoter claim in this beat rests on the ACTIVE-PROMOTER null instead"
}

How this could bite us back

If the H3K9 repressive module sits on the KMT2A PROMOTER rather than the breakpoint, then a G9a/GLP inhibitor or a KAP1 disruptor given prophylactically - the intervention risk-the-h3k9me3-exoneration-of-g9a-inhibition-is-epithelial-only concerns - would act on KMT2A's own expression, not on the stability of the region 52 kb away that the rearrangement mechanism names. Raising wild-type KMT2A expression in a cell at risk of a KMT2A translocation is an effect with no established sign, and nothing in this corpus predicts it.
The module is defined here by occupancy in two cancer cell lines with no input track and no IgG. A promoter that is hot for every antibody in a deposit is exactly the setting in which a shared artefact looks like a shared complex, and the active-promoter null re-bases that artefact rather than removing it. Anyone acting on this before an input-controlled repeat is acting on a ChIP artefact hypothesis that has not been excluded.

Still open

Do G9a, HP1a and KAP1 co-occupy most active promoters in MCF-7, or only some? Without the genome-wide three-way overlap there is no base rate and 'converges' is unearned. Q110 named this same one-command control for K562 and did not run it; two beats have now flagged it and neither has executed it. It is the cheapest remaining step in this whole axis.
Is the promoter module present at the KMT2A promoter in a KMT2A-rearranged cell? Two non-leukaemic lineages agree; nothing in blood with a fusion has ever been measured. exp-q116 is the design.
What is the H3K9me2 state of the KMT2A promoter? G9a is there at the 87th-89th percentile of active promoters and its product has never been measured at this locus in any cell type. A writer with no mark measured is half an observation.
Does the module do anything to KMT2A transcription? Nothing here is functional. KMT2A is expressed at 32.94 TPM in these cells with the full repressive module on its promoter, which is either a poised state or an irrelevance, and occupancy cannot tell the two apart.

Every source behind this entry (4)

E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset
In MCF-7, the KMT2A promoter (TSS +/-500 bp) reads the 87.60th and 88.65th percentile of the 2,000 highest-expressed genes' promoters for EHMT2/G9a, the 91.05th and 94.50th for CBX5/HP1a, and the 98.90th, 98.80th and 99.35th for TRIM28/KAP1 (two -E2 replicates and one +E2) - 1.34x-3.02x the median active promoter. At the 146 bp exon-11 anchor the same antigens read a median 33.42 (G9a, 4 tracks), 47.89 and 0.00 (CBX5), and 90.62 and 0.00 (TRIM28) of same-cell-type CTCF summit windows, so none is PRESENT by the pre-registered rule.
GSE139196: ChIP-seq of G9a (EHMT2), HP1a (CBX5) and KAP1 (TRIM28) in MCF-7 +/- estradiol, GSM4133792/94, GSM4133826/27, GSM4133828/30, hg38 bedGraph · Yang F, Tanasa B, Micheletti R, Ohgi KA, Aggarwal AK, Rosenfeld MG · NCBI GEO 2019
What this evidence cannot show
MCF-7 only, one ER+ breast adenocarcinoma line, not KMT2A-rearranged and not haematopoietic. No input track, no IgG, no antibody clone stated for any target - so occupancy is bounded by the antibody in every case. The active-promoter set is ranked by ENCODE MCF-7 RNA-seq (ENCFF251DYL), a different culture in a different laboratory from the ChIP. The promoter-specificity pass is deviation D9: its window geometry and gene sets were declared in q2_promoter_control.py's header before any track was reopened, but the pass itself was added mid-run because the C5 panel baseline forced it. n is one track per antigen per condition for CBX5 and TRIM28 and two for EHMT2, so no antigen here has a within-condition replicate in the promoter-specificity arm.
Evidence that argues against this (1)
- The exon-11 numbers for CBX5 and TRIM28 include a 0.00 in the +E2 track of each, and GSM4133830 (TRIM28 +E2) is a 61,995,812-byte file, 2.2-6x smaller than every other track in the panel. A zero from the shallowest library in the deposit is a depth statement as much as an occupancy statement, and the ORDINARY verdicts for CBX5 and TRIM28 rest on two numbers each. Only the G9a verdict has four tracks behind it. source
E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset
Against 56,171 same-cell-type CTCF summit windows, every protein track in GSE139196 reads between the 87.55th and 99.91st percentile at the KMT2A promoter and the whole-panel median across all C1-passing tracks of all antigens is 96.87 - so the CTCF-anchor percentile cannot distinguish a protein address from an active promoter, and the active-promoter comparison in E1 is the statistic that can. Against the ACTIVE-PROMOTER null the same panel separates: TRIM28 98.80-99.35, CBX5 91.05-94.50, EHMT2 87.60-88.65, KDM4C 78.35, KDM4B 58.60 (1.05x the median active promoter, i.e. background) and H3K9me3 0.00-61.05 - so the active-promoter null discriminates where the CTCF-anchor null does not.
analyses/q116-h3k9-axis-at-the-locus/results_scored.json, controls C5 and C5b (deviation D8) · Q116-h3k9-axis-at-the-locus · oncofusion-knowledge-graph 2026
What this evidence cannot show
C5 was pre-registered on the 90 kb gene-body interval; its extension to the promoter interval is deviation D8 and is labelled as an extension, not as a pre-registered control. The panel median moves as tracks are added and is quoted at the track count recorded in results_scored.json. The panel is one deposit, one submitter and one pipeline, so it controls for that pipeline's promoter behaviour and not for ChIP-seq in general.
Evidence that argues against this (1)
- An active-promoter null does not remove the hyper-ChIPable artefact, it only re-bases it. Highly expressed genes' promoters differ enormously in accessibility and Pol II load, and the KMT2A promoter could be in the top decile of the artefact rather than in the top decile of real binding. THE CONTROL THAT WOULD SETTLE THIS IS AN INPUT TRACK, AND NEITHER GSE139196 NOR GSE225028 DEPOSITS ONE. That is the single most important thing missing from this analysis and it is why exp-q116 exists. source
E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset
The two HOMER peak files deposited with GSM5660804 and GSM5660805 call a KAP1 peak at chr11:118,436,186-118,436,464 (-E2) and chr11:118,436,251-118,436,545 (+E2), overlapping by 267 bp and 202 bp the TRIM28 ENCODE conservative-IDR peak at chr11:118,436,029-118,436,453 that Q110 found in K562. Both files call zero peaks in the exon-11 window, the CTCF element or the BCR core, with the nearest chr11 peak 52,230 and 52,149 bp from the exon-11 CTCF summit.
GSM5660804 and GSM5660805 HOMER peak files, and ana-q110-h3k9-writer-and-reader-occupancy-at-the-kmt2a-locus · Yang F et al.; Q110-h3k9me3-writer-layer · NCBI GEO / oncofusion-knowledge-graph 2021
What this evidence cannot show
The MCF-7 peak files' declared positive control (TRIM28 peaks in the ZNF274 gene body) FAILED - zero peaks in both, against 2 in Q110's ENCODE set - so their absences are weakly bounded. The promoter peak is 11.8 and 12.2 normalized tags against a top peak of 1,525.5, i.e. the 1,512th of 8,278 and 1,062nd of 5,884 genome-wide. See ana-q116-kap1-has-a-called-peak-on-the-kmt2a-promoter-in-mcf7-and-none-within-52-kb-of-the-anchor for the full peak-level treatment.
Evidence that argues against this (1)
- The second address Q110 found in K562 - a shared EHMT2/TRIM28 element at chr11:118,415,7xx-118,416,1xx, 20 kb upstream - does NOT reproduce: the MCF-7 KAP1 peak files call zero peaks there. So the two cell types agree on one of the two addresses and disagree on the other, and 'the module converges on the promoter' is a claim about the promoter only. source
E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
KAP1/TRIM28 is a major HP1-interacting partner in MCF-7 by mass spectrometry, and HP1a's chromodomain-H3K9me3 interaction is regulated by phosphorylation of its N-terminal extension. HP1a's recognition of methylated H3 Lys-9 is the founding reader interaction for this mark.
Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domain · Bannister AJ, Zegerman P, Partridge JF, Miska EA, Thomas JO, Allshire RC, Kouzarides T · Nature 2001
PMID 11242054 DOI
What this evidence cannot show
The Bannister paper is cited for the reader chemistry at abstract level; the MCF-7 mass-spectrometry statement is from PMID 34194047 (Extended Data Fig. 11a-b, legend read via PMC8439151 on 2026-09-23) and the companion founding paper Lachner M et al., Nature 2001, PMID 11242053, DOI 10.1038/35065132, establishes the same interaction independently. Neither paper is about KMT2A.
Evidence that argues against this (1)
- A physical complex in solution does not make co-occupancy at one promoter a module. The genome-wide overlap of the G9a, HP1a and KAP1 peak sets in MCF-7 was NOT computed here, so there is no base rate: if these three co-occupy most active promoters, the KMT2A promoter is unremarkable among them and the word 'converges' is doing work the data does not support. Q110 asked exactly this question of K562 and also did not run it - two beats have now named the same one-command control and neither has executed it. source

Connected to

related to →EHMT2 / G9a - the H3K9me1/2 writer that sets whether the KMT2A LOCUS ITSELF rearranges, and whose loss delays MLL-AF9 AML in a supplementary figure nobody can find
G9a at the KMT2A promoter, 87.6th-88.7th percentile of 2,000 active promoters, 1.34x-1.44x the median active promoter
related to →CBX5 / HP1-alpha (P45973, 191 aa) - the H3K9me3 reader that recruits the writer, the writer's destroyer and the second writer; and the one member of that circuit with a published pro-leukaemic role, as part of an MLL leukaemia stem-cell programme that immortalises myeloid progenitors WITHOUT Hoxa or Meis
HP1a at the KMT2A promoter, 91.1st-94.5th percentile of active promoters - the first CBX5 occupancy at this locus that is not a zero
related to →TRIM28 / KAP1 / TIF1beta (Q13263) - the KRAB-ZFP corepressor scaffold in the pan-MLL-fusion core, with the weakest index-fusion p-value but a leukaemic role in another fusion family
KAP1 at the KMT2A promoter, 98.9th percentile of active promoters and 3.02x the median, plus a called peak in both hormone conditions
supports →RUN: ten ENCODE peak sets and fourteen signal tracks read at the KMT2A locus in two blood cells - not one called peak for SETDB1, CBX5, CBX1, CBX3 or TRIM28 anywhere in the gene, while G9a/EHMT2 and TRIM28 both sit on the KMT2A PROMOTER and neither touches the breakpoint cluster region
the K562 G9a-plus-TRIM28 promoter address reproduces in MCF-7 at signal and peak level, while the shared upstream element 20 kb away does not
refutes →No H3K9me3 writer and no H3K9me3 reader has ever been localised at the KMT2A locus IN THE PUBLISHED LITERATURE, in any cell type: 20 queries, 5 alias spellings, 0 on-topic records - and CORRECTED 2026-09-22: the SUV39H1 half is no longer unanswerable from archives. ENCODE still holds zero SUV39H1 experiments (verified by three routes with controls), but GEO holds two deposits, and Q114 read both at this locus and found the writer ABSENT
three H3K9-machinery proteins ARE localised at the KMT2A locus in a second cell type - at the promoter. The absence holds only for the breakpoint and the exon-11 anchor
supports →RUN: G9a/EHMT2 reads the 95.8th percentile of 56,171 same-cell-type CTCF anchors at the KMT2A promoter in all four MCF-7 ChIP-seq tracks, and a median 33.42nd at the exon-11 anchor - so the cheapest test of hyp-q114's third prediction REFUTES it
places the G9a promoter reading in its panel context and supplies the active-promoter control that the CTCF-anchor percentile could not
supports →RUN: KAP1/TRIM28 has a called peak overlapping the KMT2A promoter in MCF-7 in both hormone conditions, 267 bp of it inside the K562 TRIM28 peak Q110 found - and its nearest chr11 peak to the exon-11 CTCF summit is 52,230 bp away
the peak-level half of the same module; the signal-level KAP1 number is the 98.9th percentile of active promoters
related to →H3K9me3 - measured in KMT2A::AFF1 cells, in a supplementary figure, and it rises GENERALLY where H3K27ac falls SPECIFICALLY in the same experiment
the reader and the scaffold for this mark are at the KMT2A promoter while the mark itself reads an explicit 0.0000 at the exon-11 anchor in five of six MCF-7 tracks
related to →The CTCF site in KMT2A exon 11 - a conserved insulator element sitting exactly on the functional cut point of the breakpoint cluster region, whose occupancy decides whether the locus rearranges
five H3K9-layer proteins in five cell types are now ORDINARY or absent at this element; none exceeds the 75th percentile of same-cell-type CTCF anchors anywhere it has been looked for
related to →The experiment this beat could not do: G9a, HP1a, KAP1 and H3K9me3 at the KMT2A promoter and exon-11 anchor in a KMT2A-rearranged B-lineage cell and a CD34+ progenitor, WITH an input arm, an IgG arm and a same-cell CTCF arm
the experiment that would decide whether this module is real or an input-uncontrolled promoter artefact
← supportsRUN: endogenous HP1a/CBX5 is DEPLETED at the exon-11 anchor (median 23.94th percentile) and has the lowest KMT2A gene-body reading of any protein in the panel (30.16) - while the phosphomimetic Bio-ChIP mutant 4E reads higher than the phospho-compromised 4A at all seven intervals AND that difference falls inside the genome-wide interquartile range of the same comparison, so it carries no information
supplies the HP1a arm of that module: 91.05th and 94.50th percentile of 2,000 active promoters at the KMT2A promoter, against DEPLETED at the anchor
← proposed byThe H3K9 machinery's address at KMT2A is the PROMOTER, not the exon-11 breakpoint anchor: G9a, HP1a and KAP1 all read the top decile of active promoters there in MCF-7 and KAP1's peak replicates a K562 peak to 267 bp, while five proteins across five cell types are ORDINARY or absent at the anchor - so every H3K9-directed intervention this graph has discussed is aimed 52 kb from where the machinery is
the analysis that generates this hypothesis