start / graph / ana-q116-g9a-sits-on-the-kmt2a-promoter-in-mcf7-and-is-ordinary-at-the-exon-11-anchor

AnalysismoderateQ116-h3k9-axis-at-the-locus

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

also called the G9a answer · EHMT2 at the KMT2A promoter in a breast line · hyp-q114 prediction 3, run

The enzyme G9a sits firmly on the switch that turns the KMT2A gene on, in every one of four measurements, but it is unremarkable at the small spot 52 kb away where this gene breaks. So G9a is not the missing enzyme at the breakpoint either.

The same thing, for a scientist

**The question.** `hyp-q114-the-h3k9me3-at-the-kmt2a-anchor-is-not-put-there-by-suv39h1` states as its third prediction that *"EHMT2/G9a occupancy at the exon-11 anchor in MCF-7 (GSE139196, four deposited samples, never analysed) is at or above the 75th percentile of MCF-7 CTCF anchors - i.e. higher than SUV39H1's"*. That prediction is now run, on the four samples it names, against the same null Q114 used [E1]. **The answer is no, twice over.** At the 146 bp exon-11 window, G9a reads the **43.01st, 23.83rd, 8.36th and 69.15th** percentile of **56,171 MCF-7 CTCF summit windows** - median **33.42**, **zero of four tracks above the 75th**, and **below** SUV39H1's median of 79.93 in the same cells rather than above it. By the pre-registered rule the verdict is **ORDINARY**, and prediction 3 is **REFUTED** [E1]. **And the positive result is 52 kb away and it is clean.** At `kmt2a_tss_2kb` (chr11:118,435,456-118,437,456) G9a reads **95.82 / 95.83 / 87.55 / 97.76** - **four of four above the 87th percentile**, median 95.82, folds of **1.73x / 2.41x / 1.26x / 1.61x** the random-tile mean. This is the tightest agreement any protein in this panel shows at any interval, against a **60.8-percentile** replicate spread at the anchor in the same four files [E1]. **It is also a cross-cell-type replication of a result this graph already holds.** `ana-q110-h3k9-writer-and-reader-occupancy-at-the-kmt2a-locus` found G9a's strongest *KMT2A*-region ENCODE peak at chr11:118,434,740-118,435,115 in **K562**, with three more bracketing the TSS and none in the BCR or the element. A second cell type, a second lab, a second assay format and a continuous rather than a peak-level instrument agree: **G9a is at the *KMT2A* promoter and not at the breakpoint** [E2]. **What stops this becoming a bigger claim.** The whole 90 kb gene body reads **63.88 / 69.24 / 60.92 / 70.10** - below the 75th in all four tracks, so G9a does **not** show the locus-wide elevation SUV39H1 showed (81.65-86.79). That is the useful half: the promoter number is specific within the locus, not a locus-wide artefact. But the chr2 gene desert in these same tracks reads **1.16-1.41x** the random-tile mean, so a gene-body fold of 1.03-1.08 is indistinguishable from background and only the promoter fold clears it. **ADDED LATER THE SAME DAY, and it is a correction to the strength of the promoter claim above.** The pre-registered C5 panel baseline, extended to the promoter interval (deviation D8), found that **every protein track in this deposit reads the 87.6th-99.9th percentile of CTCF summit windows at the *KMT2A* promoter**, panel median **96.87** - the signature of an active promoter being ChIP-hot for everything. **So the 95.8th percentile above is NOT by itself evidence that G9a binds this promoter.** A third pass (deviation D9) asked the right question against the right null: against the promoters of the **2,000 highest-expressed MCF-7 genes**, G9a at the *KMT2A* promoter reads the **87.60th and 88.65th percentile, 1.44x and 1.34x the median active promoter**. **The claim survives, and it survives on that statistic and not on the one in the table above.** See `ana-q116-the-h3k9-repressive-module-converges-on-the-kmt2a-promoter-in-mcf7-and-none-of-it-is-at-the-breakpoint`.

Go deeper

The numbers, per track, with the denominators printed

All four tracks pass the pre-registered C1 dynamic-range gate, and every denominator is non-zero and is printed - the failure mode that made Q114's gate report inf >= 5 -> passes does not arise here:

track	condition	C1 ratio	C1 numerator	C1 denominator	N2 zero fraction
GSM4133792	-E2	35.56x	16.751	0.4711	17.0%
GSM4133793	-E2	66.31x	23.677	0.3571	16.4%
GSM4133794	+E2	21.08x	9.802	0.4649	15.4%
GSM4133795	+E2	20.91x	11.514	0.5508	28.1%

All seven intervals, percentile of width-matched MCF-7 CTCF summit windows:

interval	width	GSM4133792	GSM4133793	GSM4133794	GSM4133795	median
exon11	146	43.01	23.83	8.36	69.15	33.42
ctcf_element	1,047	74.33	20.43	35.99	62.08	49.04
ctcf_summit_1kb	1,000	74.10	20.35	38.05	61.73	49.89
dcal_cut_1kb	999	16.49	18.32	27.09	42.40	22.71
bcr_core	7,465	55.08	56.85	48.37	54.50	54.79
kmt2a_tss_2kb	2,000	95.82	95.83	87.55	97.76	95.82
kmt2a_gene	90,376	63.88	69.24	60.92	70.10	66.56

The kmt2a_gene row is scored against 5,000 subsampled width-matched anchors; every other row against all 56,171.

Modal raw-interval run length over chr11:118,480,000-118,500,000 is 83, 22, 83 and 83 bp - all below the 146 bp window, so the exon-11 number is a real measurement in every track and not one bin.

Why an ORDINARY verdict here is worth more than it looks

G9a writes H3K9me1 and H3K9me2, not me3 3. That makes it the enzyme named by mechanism-h3k9me1-2-ctcf-balance-gates-kmt2a-rearrangement, which is the arm of this graph's H3K9 dispute that argues lowering the mark is protective - the opposite therapeutic direction from mark-h3k9me3. So a G9a positive at the anchor would have connected to a different existing claim than a SUV39H1 positive would have, and would have put a G9a/GLP inhibitor in the frame at this element.

It does not fire. Combined with Q114's SUV39H1 result and Q110's SETDB1 peak-level zeros in K562, no H3K9 methyltransferase has now been placed at the exon-11 anchor above the 75th percentile of same-cell-type CTCF anchors in any cell type in which one has been measured - three enzymes, three cell types, two assay formats.

And the estradiol arm adds nothing at this locus: -E2 gives 95.82 and 95.83 at the promoter, +E2 gives 87.55 and 97.76. The paper this deposit belongs to reports G9a declining at oestrogen-responsive promoters after 40 min of E2 2; KMT2A is not one of those promoters and shows no such change.

A 60.8-percentile replicate spread at the anchor, and a 10.2-point one at the promoter

The same 146 bp window reads 8.36 and 69.15 in the two +E2 samples of one antibody in one cell line in one submission. Q114 measured a 44.5-percentile spread across four SUV39H1 samples of the same deposit and called it the corpus's only replicate estimate of that uncertainty; this is wider.

The promoter window in the same four files spans 87.55 to 97.76 - 10.2 points. So the spread is not a property of the tracks; it is a property of reading a 146 bp window in a ChIP-seq track whose modal interval is 22-83 bp. Any single-track percentile at a sub-200 bp window in this deposit should be read as carrying tens of percentile points of uncertainty, and this analysis reports medians across four tracks for that reason.

Facts and figures

script	analyses/q116-h3k9-axis-at-the-locus/run116.py
prereg	analyses/q116-h3k9-axis-at-the-locus/PREREGISTRATION.md
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 was established from that file's OWN maximum coordinate per chromosome against GRCh38 and hg19 lengths in both directions before any interval was read; nothing was lifted.
verdict by the preregistered rule	ORDINARY at exon11 (median 33.42, in [25,75], 0 of 4 tracks above 75). NOT locus-wide elevated (kmt2a_gene median 66.56, below 75). PRESENT at the KMT2A promoter is not a verdict the pre-registered rule defines, because the rule scores exon11; the promoter result is reported descriptively and its strength is that 4 of 4 tracks exceed the 87th percentile.
hyp q114 prediction 3	REFUTED. The prediction was 'at or above the 75th percentile, i.e. higher than SUV39H1's'. Measured: median 33.42, zero of four tracks above 75, and below SUV39H1's 79.93 median in the same cells.
exon11 median percentile	33.42
exon11 replicate spread percentile points	60.79
kmt2a tss 2kb median percentile	95.82
kmt2a tss 2kb spread percentile points	10.21
null	N1 - 56,171 windows of each interval's own width centred on the point-source summit of every ENCFF942TCG optimal-IDR CTCF peak (ENCSR000AHD, Myers/HAIB, MCF-7 - the SAME cell type as the target tracks). 32 of 56,203 peaks dropped for falling on non-primary scaffolds (0.057%), counted not silent.
localization	chromatin at chr11q23.3 - the KMT2A promoter (chr11:118,435,456-118,437,456), the 7,465 bp breakpoint cluster core, the 146 bp exon-11 CTCF anchor and the 90 kb gene body; plus chr19q13 and a chr2 gene desert as controls
subnuclear localization	not measured. A ChIP signal value is a linear-genome address and carries no compartment information.
cell type	MCF-7 (ER+ breast adenocarcinoma). NOT haematopoietic, NOT KMT2A-rearranged, NOT a CD34+ progenitor.
cell state	hormone-deprived culture, and 40 min 17-beta-oestradiol. No damage arm, no knockdown arm of G9a itself.
epigenetic readout	occupancy of EHMT2/G9a, the H3K9me1/me2 writer, at the KMT2A locus
mark written	H3K9me1 and H3K9me2 (the enzyme's products; neither mark is measured in this deposit)
writer of that mark	EHMT2/G9a, Q96KQ7
shown for fusion	NO - nobody has looked in any KMT2A fusion. MCF-7 carries no KMT2A rearrangement and the deposit has no leukaemia arm.
absence class	measured-not-analysed
absence class note	The four G9a samples were deposited publicly in 2019 and nobody had read them at this locus. The absence this analysis closes is the analysis, not the measurement.
added after first write	This entry was written when four G9a tracks were in and before the panel baseline had enough tracks to bite. The promoter paragraph was amended the same day rather than rewritten, and the original numbers are unchanged.
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.

results

[
  "analyses/q116-h3k9-axis-at-the-locus/results.json"
]

exon11 percentiles

[
  43.01,
  23.83,
  8.36,
  69.15
]

kmt2a tss 2kb percentiles

[
  95.82,
  95.83,
  87.55,
  97.76
]

kmt2a tss 2kb fold vs random tile mean

[
  1.727,
  2.405,
  1.261,
  1.608
]

kmt2a gene percentiles

[
  63.88,
  69.24,
  60.92,
  70.1
]

c1 dynamic range ratios

[
  35.56,
  66.31,
  21.08,
  20.91
]

c1 denominators printed

[
  0.4711,
  0.3571,
  0.4649,
  0.5508
]

c3 chr2 desert fold vs random tile mean

[
  1.2,
  1.2,
  1.16,
  1.41
]

modal raw interval run bp

promoter claim bounded by C5b and rescued by D9

{
  "problem": "every protein track in GSE139196 reads 87.55-99.91 percentile of CTCF summit windows at the KMT2A promoter; panel median 96.87",
  "right_statistic": "percentile among the promoters of the 2,000 highest-expressed MCF-7 genes",
  "EHMT2_minusE2": 87.6,
  "EHMT2_plusE2": 88.65,
  "x_median_active_promoter": [
    1.44,
    1.34
  ],
  "verdict": "the G9a promoter claim survives the active-promoter control; the CTCF-anchor percentile alone would not have supported it"
}

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

This removes G9a from the frame at the exon-11 anchor and does NOT remove it from the frame at the KMT2A promoter, where it is the most strongly occupied H3K9 enzyme measured at this locus in any cell type. A G9a/GLP inhibitor given prophylactically - the intervention risk-the-h3k9me3-exoneration-of-g9a-inhibition-is-epithelial-only is about - would act on the promoter, where the enzyme demonstrably is, not on the breakpoint, where the mechanism says it matters. The risk moves along the gene rather than disappearing.
An ORDINARY occupancy verdict is not an exoneration of the enzyme. G9a's products are H3K9me1 and me2, neither of which is measured anywhere in this deposit, and a processive enzyme can hand a me2 substrate to another writer without dwelling at the site. Prediction 3 of hyp-q114 is refuted; the hypothesis's underlying question - who deposits the mark at this element - is not answered.

Still open

Does G9a occupy the KMT2A promoter in a haematopoietic cell, and does it matter 52 kb away at the breakpoint? Two cell types now agree on the promoter address and neither is blood. A CD34+ or KMT2A-rearranged G9a ChIP would settle whether this is a general property of the KMT2A promoter or an epithelial one.
Is the promoter address specific, or do G9a and TRIM28 share most active promoters in MCF-7? Q110 asked exactly this of K562 and did not compute it; this run did not compute it either. Without the genome-wide overlap of the two, a shared promoter is a coincidence with an unknown base rate.
H3K9me1 and H3K9me2 - G9a's actual products - are in NO track of this deposit. The enzyme has now been placed at the KMT2A promoter in two cell types and its mark has never been measured there in either. What is the me2 state of the KMT2A promoter in MCF-7?
Why is the exon-11 replicate spread 60.8 percentile points while the promoter spread is 10.2 in the same four files? If the answer is simply that a 146 bp window is below the resolution of this library, then every sub-200 bp percentile this corpus holds from ChIP-seq needs an uncertainty band it does not currently carry.

Every source behind this entry (3)

E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset
In four MCF-7 G9a/EHMT2 ChIP-seq tracks of GSE139196, the KMT2A exon-11 146 bp window reads the 43.01st, 23.83rd, 8.36th and 69.15th percentile of 56,171 same-cell-type CTCF summit windows (median 33.42, 0 of 4 above the 75th), while the KMT2A TSS +/-1 kb window reads the 95.82nd, 95.83rd, 87.55th and 97.76th (4 of 4 above the 87th). All four tracks pass a dynamic-range gate of 20.9x-66.3x over a printed, non-zero denominator.
GSE139196: ChIP-seq of G9a (EHMT2) in MCF-7 with and without estradiol, GSM4133792-95, hg38 bedGraph · Yang F, Tanasa B, Micheletti R, Ohgi KA, Aggarwal AK, Rosenfeld MG · NCBI GEO 2019
What this evidence cannot show
MCF-7 is an ER+ breast adenocarcinoma, not a haematopoietic cell and not KMT2A-rearranged. The deposit gives no antibody clone for G9a. There is no IgG or knockout arm, so occupancy is bounded by the antibody. The null is MCF-7's own optimal-IDR CTCF anchor set (ENCFF942TCG, ENCSR000AHD) but the C4 instrument control on that geometry FAILED in Q114 at 72.97 against a required 90th percentile, so no verdict here is labelled a class-matched positive; the diagnosis there was a defect in the control's geometry, not in the locus, and both MCF-7 and SK-N-SH carry a score-1000 CTCF peak directly over the anchor. 32 of 56,203 peaks fall on non-primary scaffolds and were dropped and counted (0.057%). HOMER makeUCSCfile normalises each bedGraph to a fixed tag total, so absolute means are depth-normalised and only the within-track percentile is comparable between samples.
Evidence that argues against this (2)
- The chr2 gene desert (chr2:35,000,000-36,000,000), a declared negative control, reads 1.20x, 1.20x, 1.16x and 1.41x the random-tile mean in these same four tracks. A gene-body fold of 1.03-1.08 at KMT2A is therefore indistinguishable from a desert, and only the promoter fold of 1.26-2.41x clears that floor. The percentile statistic is the one that carries the promoter result, not the fold. source
- The exon-11 percentile spans 8.36 to 69.15 across four samples of one antibody in one cell line. A median of 33.42 from four numbers with that spread is a weak point estimate, and a reader who wanted to argue that G9a IS at the anchor could point at 69.15 and be quoting a real measurement. The ORDINARY verdict rests on the pre-registered rule (median in [25,75] and fewer than 60% of tracks above 75), not on the spread being small. source
E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
GSE139196 is a sub-series of GSE139199, whose paper reports that G9a binding DECLINES at oestrogen-responsive promoters after 40 min of E2, and that the H3K9me3 writer SUV39H1 is dismissed while KDM4B and KDM4C are recruited. KMT2A is not among the promoters that paper analyses, and the word KMT2A does not appear in it.
Shape of promoter antisense RNAs regulates ligand-induced transcription activation · Yang F, Tanasa B, Micheletti R, Ohgi KA, Aggarwal AK, Rosenfeld MG · Nature 2021
PMID 34194047 DOI
What this evidence cannot show
Full text read via the PMC HTML route (PMC8439151) on 2026-09-23, not abstract only. The load-bearing sentence is 'The H3K9me3 writer SUV39H1 was dismissed in response to E2, whereas the erasers of this H3K9me3 mark, KDM4B and KDM4C, were recruited at increased levels to the promoters after E2 treatment', and separately 'G9a, an H3K9me2 methyltransferase, also showed a decline in binding to the promoters following E2 treatment'. Searched for 'KMT2A' and 'MLL' in the full text: zero hits for each. The paper is about promoter antisense RNA and Pol II pause release; every occupancy statement in it is about ER-responsive promoters, so it neither supports nor contradicts anything at KMT2A.
Evidence that argues against this (1)
- Q110 placed G9a at the KMT2A promoter in K562 at peak level, but with a single ENCODE peak set (ENCSR175EOM/ENCFF373EMV, 14,904 peaks) and one peak inside the gene against 0.464 expected under uniform placement. One peak against a 0.46 expectation is a weak individual observation; what makes the pair worth reporting is that two independent systems put the same protein at the same 2 kb of the genome. source
E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
G9a/EHMT2 (Q96KQ7, 1,210 aa) mono- and dimethylates H3 Lys-9 in euchromatin; UniProt records catalytic activities for the unmethylated-to-me1 and me1-to-me2 steps at H3K9 and no me2-to-me3 step, and G9a is the dominant euchromatic H3K9 methyltransferase in mouse genetics.
G9a histone methyltransferase plays a dominant role in euchromatic histone H3 lysine 9 methylation and is essential for early embryogenesis · Tachibana M, Sugimoto K, Nozaki M, Ueda J, Ohta T, Ohki M, Fukuda M, Takeda N, Niida H, Kato H, Shinkai Y · Genes Dev 2002
PMID 12130538 DOI
What this evidence cannot show
Abstract-level plus the live UniProt Q96KQ7 record fetched 2026-09-23; the 2002 paper's full text was not read. The me1/me2 restriction is an enzymology fact about the purified enzyme and about bulk mouse chromatin, and does not by itself exclude a G9a contribution to H3K9me3 in cells via a me2 substrate pool handed to another enzyme - which is exactly the escape route hyp-q114 was relying on and which this analysis does not close.
Evidence that argues against this (1)
- An open question already in this corpus, on ana-gse210480-h3k9me3-at-the-kmt2a-bcr, asks whether the G9a-knockdown-alone arm lowers H3K9me3 at the BCR - i.e. whether G9a influences me3 in cells despite being annotated me1/me2-only. That question is unanswered, and if the answer is yes then an ORDINARY G9a occupancy result at the anchor still would not exonerate G9a from the me3 state there. source

Connected to

refutes →The H3K9me3 this graph argues about at the KMT2A exon-11 anchor is not deposited by SUV39H1: in SU-DIPG13 the mark sits at the 57.8th percentile of class-matched CTCF anchors while the writer reads an EXPLICIT ZERO in the same nuclei, and genome-wide the two barely track each other (Spearman 0.148, 0.078)
runs that hypothesis's own prediction 3 on the four samples it names and refutes it: G9a at the exon-11 anchor is median 33.42nd percentile, not above the 75th, and below SUV39H1's 79.93
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
first signal-level occupancy of G9a/EHMT2 at the KMT2A locus in any cell type: 95.8th percentile at the promoter, ordinary at the anchor
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
replicates Q110's K562 peak-level G9a-at-the-KMT2A-promoter address in a second cell type, a second lab and a continuous instrument
tests →The deposit this graph filed as 'four SUV39H1 ChIP-seqs' is actually 77 samples in MCF-7 carrying the ENTIRE H3K9 axis in one cell type and one pipeline - writer SUV39H1, writer G9a/EHMT2, reader HP1-alpha/CBX5 (wild-type AND two phospho-mutants), co-repressor KAP1/TRIM28, erasers KDM4B and KDM4C, the mark H3K9me3, and an shKDM4B+shKDM4C eraser-knockdown arm
the G9a arm of that experiment entry, run
related to →The H3K9me1/2-CTCF brake: a KDM3B-G9a balance keeps the KMT2A breakpoint cluster region methyl-poor and CTCF-bound, and losing either end makes the locus amplify and rearrange - in epithelium, in primary cord-blood progenitors and in mice
the me1/me2 writer that mechanism names is at the KMT2A promoter and ordinary at the exon-11 anchor in MCF-7 - the first occupancy measurement of that enzyme at this locus
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
G9a occupancy measured at this element, median 33.42nd percentile of 56,171 same-cell-type CTCF anchors
← related toRUN: 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 same 2 kb of the KMT2A promoter carries G9a at the 95.8th signal percentile and a KAP1 called peak, in the same cell line - the K562 pairing Q110 reported, in a second lineage
← related toANSWERED from the paper's own methods, not from the corresponding author: the `dsgf` token that splits GSE139196's four SUV39H1 tracks into the two high and the two low ones is DSG-plus-formaldehyde double crosslinking - so the 44.5-percentile spread Q114 reported is a protocol variable, not replicate noise
the paper's methods name G9a as the DSG-crosslinked antigen, so the four G9a tracks read here may carry the same protocol difference the deposit's filenames do not record
← testsThe 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 tell whether the MCF-7 promoter reading is a protein address or an active-promoter artefact
← supportsTHE 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
places the G9a promoter reading in its panel context and supplies the active-promoter control that the CTCF-anchor percentile could not
← refutesA panel of many antigens in one cell type is the input track the deposit did not provide: eight H3K9-axis proteins in MCF-7 all read the whole KMT2A locus elevated (six-non-anchor medians 49.8-72.4) and all read the KMT2A promoter at the 87.6th-99.9th percentile - so the shared component has to be subtracted before any single antigen's number means anything
removed the stated basis of that entry's promoter claim: every protein track in the deposit reads 87.6-99.9 at the same promoter