start / graph / ana-q114-in-mcf7-h3k9me3-is-depleted-across-the-kmt2a-locus-while-the-writer-looks-elevated

AnalysismoderateQ114-suv39h1-at-the-locus

In MCF-7 the two signals point in opposite directions over the same 90 kb: H3K9me3 sits at the 15.8th and 10.3rd percentile of 56,171 same-cell CTCF anchors across the KMT2A gene body (0.82x and 0.65x the random-tile mean) while SUV39H1 sits at the 81.7th-86.8th in all four of its tracks - and the 147 bp anchor carries ZERO H3K9me3 in the minus-estradiol track

also called mark down, writer up · Q114 MCF-7 mark arm

In breast cancer cells the silencing tag is unusually scarce right across the KMT2A gene - at the exact 147-letter spot this project cares about there is none at all in one sample - while the enzyme that is supposed to make that tag looks, if anything, plentiful there. Enzyme and product disagree about the same piece of DNA.

The same thing, for a scientist

**Both signals were read from the same deposit, the same cells and the same pipeline, against the same 56,171 same-cell-type CTCF anchor windows** [E1][E2]. | interval | H3K9me3 −E2 | pctile | H3K9me3 +E2 | pctile | SUV39H1 pctile, all four | |---|---:|---:|---:|---:|---| | **exon11 (147 bp)** | **0.0000** | **0.00** | 0.3900 | 50.03 | 90.81 / 69.05 / 96.29 / 51.78 | | ctcf_element | 0.5877 | 41.53 | 0.3900 | 5.34 | 86.08 / 75.45 / 91.38 / 34.46 | | bcr_core | 0.6534 | 44.21 | 0.4281 | 22.67 | 82.71 / 81.25 / 82.66 / 72.76 | | kmt2a_tss_2kb | 0.5840 | 25.87 | 0.4004 | 32.99 | 78.01 / 90.74 / 89.69 / 69.73 | | **kmt2a_gene (90 kb)** | 0.6546 | **15.77** | 0.4620 | **10.30** | 85.51 / 85.31 / 86.79 / 81.65 | **H3K9me3 is depleted across the whole locus** - 0.82x and 0.65x the random-tile mean over the 90 kb gene body, the 15.8th and 10.3rd percentile of class-matched anchors - and the 147 bp exon-11 window carries **exactly zero** H3K9me3 in the minus-estradiol track and exactly the file's minimum value (0.39, one tag's worth after HOMER's 10-million-tag normalisation) in the plus-estradiol track. **And in the same cells the writer reads 82nd-87th percentile over the same 90 kb.** Whatever produces the SUV39H1 elevation there, **it is not producing H3K9me3.** That is the single clearest statement this beat can make about writer-mark dissociation, because it is a within-deposit, within-cell-type, opposite-direction comparison rather than a correlation. **The mark tracks pass the control that the writer tracks fail.** The pre-registered heterochromatin positive control gives **3.05x and 4.15x** at the 1 Mb chr19q13 KRAB-ZNF window and **3.79x and 5.41x** at ZNF274 3' for the two H3K9me3 tracks - past the pre-registered 3x - while the four SUV39H1 tracks give 0.93-1.15x. **So the depletion reported here comes from tracks that demonstrably resolve constitutive H3K9me3, and the elevation it is contrasted with comes from tracks that do not resolve anything canonical.** The asymmetry is the reason this entry leads with the mark and treats the writer number as the thing in question. **Lineage first: MCF-7 is an ER-positive breast adenocarcinoma.** Nothing here reaches a blood cell, a *KMT2A*-rearranged cell, or the damage state.

Go deeper

How this sits with the corpus's methyl-poor-BCR picture

This graph holds the KMT2A breakpoint cluster as an H3K9-methyl-poor broad domain - H3K9me1 at 0.54x the chr11 mean over the BCR against 1.28x in its flanks in RPE, replicated in K562 as a property of the 7.5 kb BCR and not of the 1 kb element. This beat added a third lineage for H3K9me3 in glioma (7.78th and 28.52nd percentile of class-matched anchors over the BCR core).

MCF-7 makes a fourth, and it is broader than the BCR: the depletion here covers the whole 90 kb gene body (15.8th and 10.3rd percentile), with the BCR core at 44.2nd and 22.7th - i.e. the gene body is more depleted than the BCR in this cell type. So the pattern is a low-H3K9me3 gene, not specifically a low-H3K9me3 breakpoint region, which is a weaker and more ordinary claim: KMT2A is a large, broadly transcribed gene, and transcribed gene bodies are H3K9me3-poor as a class.

That reading deflates the corpus's breakpoint-specific framing and it should. A matched null of 90 kb windows inside other broadly transcribed gene bodies would separate 'methyl-poor because breakpoint' from 'methyl-poor because transcribed'. It was not run here and is the cheapest way this whole line of argument could be deflated.

What the zero at the anchor is and is not

In GSM4585206 (H3K9me3, −E2) the 147 bp exon-11 window reads 0.0000. Two facts about that number:

It is a converted bigWig, written by this beat from a HOMER bedGraph. The per-file NaN probe (mandatory in the pre-registration) shows these converted files return NaN for genuinely uncovered positions and 0.0 only when fillna=0 is passed - unlike the GEO-supplied DIPG bigWigs, which carry explicit zero records and agree across all three fill modes. So this zero means no bedGraph interval covers the window, i.e. no tag, rather than a written zero.
In the +E2 track the same window reads 0.3900, which is the file's value_min - exactly one tag's depth after normalisation. So both tracks put this element at the floor of what the assay can report.

A single-tag or zero-tag readout over 147 bp is at the limit of a normalised ChIP-seq bedGraph, and no strong statement should be built on the difference between 0 and one tag. What survives is the direction, which the 90 kb gene-body number carries with far more support.

Facts and figures

script	analyses/q114-suv39h1-at-the-kmt2a-locus/run_mcf7.py
status	run
run date	2026-09-22
run by	Q114-suv39h1-at-the-locus
result summary	H3K9me3 over the 90 kb KMT2A gene body in MCF-7 is at the 15.77th (-E2) and 10.30th (+E2) percentile of 56,171 same-cell CTCF anchor windows, 0.821x and 0.648x the random-tile mean; the 147 bp exon-11 window reads 0.0000 and 0.3900 (the file minimum). In the same cells SUV39H1 over the same 90 kb is at the 81.65th-86.79th percentile in all four of its tracks.
nan probe note	the converted MCF-7 bigWigs return NaN for genuinely uncovered positions (default) and 0.0 only with fillna=0, unlike the GEO-supplied DIPG bigWigs which carry explicit zero records; the zero at exon11 in GSM4585206 therefore means no covering bedGraph interval
cell type	MCF-7, ER-positive luminal breast adenocarcinoma
cell state	hormone-deprived versus 100 nM 17-beta-estradiol 40 min
lineage	breast epithelium - NOT haematopoietic, NOT KMT2A-rearranged
shown for fusion	NO - nobody has looked in any KMT2A fusion system
genome build	GRCh38, from each file's own maximum coordinate per chromosome

writer vs mark spearman over 56171 anchors

frac top1pct writer anchors with mark below median

inputs

[
  "GSM4585206",
  "GSM4585207",
  "GSM4133857",
  "GSM4133858",
  "GSM4133859",
  "GSM4133860",
  "ENCFF942TCG"
]

h3k9me3 intervals

{
  "minusE2_GSM4585206": {
    "exon11": [
      0,
      0
    ],
    "ctcf_element": [
      0.5877,
      41.53
    ],
    "bcr_core": [
      0.6534,
      44.21
    ],
    "kmt2a_tss_2kb": [
      0.584,
      25.87
    ],
    "kmt2a_gene": [
      0.6546,
      15.77
    ]
  },
  "plusE2_GSM4585207": {
    "exon11": [
      0.39,
      50.03
    ],
    "ctcf_element": [
      0.39,
      5.34
    ],
    "bcr_core": [
      0.4281,
      22.67
    ],
    "kmt2a_tss_2kb": [
      0.4004,
      32.99
    ],
    "kmt2a_gene": [
      0.462,
      10.3
    ]
  }
}

h3k9me3 fold over random tile mean

{
  "kmt2a_gene_minusE2": 0.821,
  "kmt2a_gene_plusE2": 0.648
}

suv39h1 kmt2a gene percentiles

[
  85.51,
  85.31,
  86.79,
  81.65
]

positive control passes for the mark and fails for the writer

{
  "H3K9me3_chr19q13_1Mb": [
    3.05,
    4.15
  ],
  "H3K9me3_znf274_3prime_10kb": [
    3.79,
    5.41
  ],
  "SUV39H1_znf274_3prime_10kb": [
    0.93,
    1.15,
    1.04,
    1.11
  ],
  "prereg_threshold": 3
}

writer vs mark spearman over 56171 anchors — the numbers behind the chart

{
  "GSM4133857_vs_GSM4585206": 0.1017,
  "GSM4133858_vs_GSM4585206": 0.0968,
  "GSM4133859_vs_GSM4585207": 0.0713,
  "GSM4133860_vs_GSM4585207": 0.0942
}

frac top1pct writer anchors with mark below median — the numbers behind the chart

{
  "GSM4133857": 0.416,
  "GSM4133858": 0.411,
  "GSM4133859": 0.482,
  "GSM4133860": 0.457
}

How this could bite us back

A locus that is H3K9me3-poor in epithelium, in a blood line and now in two more cell types is a locus whose mark cannot easily be lowered further by a drug - which cuts against the safety concern in mark-h3k9me3 at this element specifically, while saying nothing about the fusion's target genes, where that concern actually lives.
The writer-up/mark-down pattern is the shape an antibody artefact makes. Before it is used as an argument that SUV39H1 is non-catalytic here, the SUV39H1 tracks need the knockout control that neither deposit has.

Still open

Is the KMT2A gene body H3K9me3-poor because it is a breakpoint region or because it is a large broadly transcribed gene? A null of 90 kb windows inside other broadly transcribed gene bodies of matched expression would separate the two, and it is the cheapest way to deflate this graph's breakpoint-specific framing. Not run.
Does the writer-mark anti-correlation at this locus hold at domain scale? The gene body is 90 kb and the comparison here is one window; a profile across the locus in 1 kb bins in both tracks would show whether the writer's elevation and the mark's depletion are spatially interleaved or uniform.
Why does the +E2 H3K9me3 track read exactly the file minimum at four of the seven intervals? Either the locus is genuinely at the floor in that sample, or the track's depth is too low for a 1 kb readout at all. The SRA runs would give library complexity and were not retrieved.

Every source behind this entry (2)

E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset
The four GSE139196 SUV39H1 ChIP-seq samples in MCF-7 (two minus estradiol, two at 40 min estradiol) and the matched H3K9me3 pair (GSM4585206 -E2, GSM4585207 +E2) were read over 56,171 same-cell-type CTCF summit windows and 200,000 random tiles, all by one code path, build established from each file's own maximum coordinate per chromosome as GRCh38.
GSE139196: SUV39H1 (KMT1A) and H3K9me3 ChIP-seq in MCF-7 with and without estradiol - GSM4133857-60, GSM4585206-207, HOMER bedGraph on hg38 · Yang F, et al. · NCBI GEO 2019
What this evidence cannot show
No antibody clone stated for any target; no input or IgG track identified in the sub-series. The four SUV39H1 samples differ in depth (44.3-49.8 million bedGraph intervals) and two of them carry an unexplained 'dsgf' token in their titles. HOMER normalises to 10 million tags, so the minimum value in each file is one tag's worth - 0.39 to 0.55 depending on the file - which is why several intervals read exactly the file minimum.
Evidence that argues against this (1)
- Replicate concordance depends on depth and on the aligner's handling of duplicates ('1 copy/read per genome position' here), so a low correlation between replicates can be a processing artefact rather than a statement about the assay. Without the raw runs this cannot be separated, and they were not retrieved. source
E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
The comparison set is 56,171 primary-assembly 147 bp windows centred on MCF-7's own CTCF point-source summits, from ENCODE's optimal-IDR peak file for the same cell type.
ENCSR000AHD - CTCF ChIP-seq in MCF-7; optimal IDR thresholded peaks ENCFF942TCG (GRCh38), default_analysis preferred_default · Richard Myers, HudsonAlpha · ENCODE Portal 2026
What this evidence cannot show
Same cell line, different lab, different decade from the SUV39H1 samples; MCF-7 sublines drift and CTCF site usage could differ. 32 of 56,203 peaks on non-primary scaffolds were dropped and counted.
Evidence that argues against this (1)
- Using a cell type's own CTCF anchors as the null assumes CTCF anchors are the right comparison class for an H3K9 writer. They are the class this project's earlier positives died against, which is why they are used - but a heterochromatin enzyme's natural comparison class might be H3K9me3 domains instead, and that null was not built. source

Connected to

tests →H3K9me3 - measured in KMT2A::AFF1 cells, in a supplementary figure, and it rises GENERALLY where H3K27ac falls SPECIFICALLY in the same experiment
H3K9me3 at the 15.8th and 10.3rd percentile of same-cell CTCF anchors over the 90 kb KMT2A gene body
refutes →The MCF-7 arm, and the clause that saved it: SUV39H1 at the exon-11 anchor reaches the 96.3rd percentile of 56,171 same-cell-type CTCF anchors in one replicate - but all six non-anchor KMT2A intervals are elevated too (medians 73-87), so the pre-registered FIRES clause does not fire, the verdict is NEITHER, and four replicates of one antibody in one cell line span 51.8 to 96.3 at the same 147 bp
the writer's locus-wide elevation is not producing mark: over the same 90 kb the mark is at the 10th-16th percentile
related to →In patient-derived H3K27M glioma, baseline H3K9me3 over the 7,466 bp KMT2A breakpoint cluster core sits at the 7.8th and 28.5th percentile of 59,821 width-matched CTCF-anchor windows (0.56x and 0.90x the random-tile median) - the methyl-poor-BCR pattern this graph found in epithelium and in a blood line, now in a third lineage and for the trimethyl state
a fourth cell type with a methyl-poor KMT2A region - but the depletion here is gene-wide rather than breakpoint-specific, which deflates the framing
supports →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)
second system, opposite arrangement, same dissociation: writer up and mark down over the same 90 kb
supports →The rarest thing in the deposit, measured: across 59,821 CTCF anchors in the SAME nuclei, SUV39H1 occupancy and H3K9me3 correlate at Spearman 0.148 and 0.078 - and a quarter to two-fifths of the enzyme's own strongest sites carry no more of its mark than the median anchor
replicates the weak writer-mark coupling in a second cell type and assay: Spearman 0.071-0.102 over 56,171 anchors
← refutesTHE DEFLATING NULL, RUN: matching on gene length AND MCF-7 expression roughly HALVES the H3K9me3 depletion over the KMT2A gene body - from the 7.6th-15.0th percentile of class-matched CTCF anchors to the 18.2nd-23.4th of 385 matched transcribed gene bodies - so the methyl-poor breakpoint domain is partly a fact about transcription and does not dissolve into one
runs the null that entry named and did not compute: matching on length and MCF-7 expression roughly halves the depletion, 7.6-15.0th percentile becoming 18.2-23.4th. The finding survives, at about half the effect size