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)

[
  "In any human cell type, SUV39H1 occupancy at the 147 bp exon-11 anchor is below the 75th percentile of same-cell-type CTCF anchors while H3K9me3 at the same window is at or above the 50th.",
  "A SUV39H1 degron or knockout in a human cell lowers H3K9me3 at the 147 bp exon-11 anchor by less than 1.3-fold, while lowering it at pericentromeric or KRAB-ZNF control regions by more than 2-fold in the same experiment.",
  "REFUTED 2026-09-23. The prediction was: \"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.\" MEASURED by the cycle-68 beat across the four deposited EHMT2 tracks and re-checked by the king in its committed results.json: exon-11 percentiles against 56,171 MCF-7 CTCF summit windows are **43.01, 23.83, 8.36 and 69.15, median 33.42, and 0 of 4 are above the 75th**. SUV39H1 reads 79.93 at the same window, so G9a is not merely below the threshold - it is BELOW SUV39H1, which is the opposite of what this prediction staked. The hypothesis that something other than SUV39H1 puts H3K9me3 here survives, but G9a is no longer its candidate: the same beat found NOTHING fires at the anchor across 8 antigens and 28 tracks."
]

[
  "exp-q114-a-suv39h1-cutandrun-in-a-cd34-progenitor-and-a-kmt2a-rearranged-line",
  "exp-q114-gse139196-holds-the-whole-h3k9-axis-in-one-cell-type"
]

[
  "ana-q114-suv39h1-is-not-at-the-kmt2a-locus-in-either-patient-derived-glioma-line",
  "ana-q114-the-h3k9me3-writer-and-its-own-mark-barely-coincide-in-the-same-nuclei"
]

[
  "A SUV39H1 degron or knockout, in ANY human cell line, that lowers H3K9me3 at the 147 bp exon-11 anchor (GRCh38 chr11:118,488,614-118,488,760) by MORE THAN 1.5-fold relative to a matched control, in a CUT&RUN or ChIP with a spike-in and with a pericentromeric or KRAB-ZNF positive control showing at least 2-fold loss in the same samples, kills this hypothesis.",
  "SUV39H1 occupancy at the same 147 bp window ABOVE the 75th percentile of same-cell-type CTCF anchors in a haematopoietic cell (CD34+, SEM or RS4;11), with a C4 instrument control that PASSES (a same-cell CTCF track placing the window above its own 90th percentile), kills the locus-specific form of this hypothesis.",
  "A writer-mark Spearman correlation ABOVE 0.5 over the same class of anchor windows in any cell type where SUV39H1 and H3K9me3 are measured together with an IgG control would show that 0.148 and 0.078 are a property of this deposit rather than of the enzyme, and would remove the genome-wide support for this hypothesis.",
  "ALREADY PARTLY TESTED, AND THE RESULT CUTS BOTH WAYS (cycle 66): in SU-DIPG36, the second line of the same deposit, SUV39H1 at this window is at the 74.17th percentile - just below the 75th threshold named in falsifier 2 and far from the zero seen in SU-DIPG13. The hypothesis survives on a technicality in one of the two lines it was generated from, and that is recorded here rather than left out."
]

{
  "verdict": "REFUTED",
  "measured": "exon-11 vs 56,171 MCF-7 CTCF summit windows: 43.01 / 23.83 / 8.36 / 69.15, median 33.42, 0 of 4 above the 75th percentile",
  "suv39h1_same_window": 79.93,
  "source": "analyses/q116-h3k9-axis-at-the-locus/results.json, re-read by the king",
  "what_survives": "the hypothesis itself. Nothing in the H3K9 panel fires at the anchor, so the question of what deposits the mark there is open and now has no named candidate at all",
  "and_a_label_error_in_this_entry": "predictions 1 and 3 both say '147 bp exon-11 anchor'. The stored bounds chr11:118,488,614-118,488,760 are 146 bp half-open and 147 inclusive, and the beat that tested this derived 146 from the interval while q114's design.py hard-coded ANCHOR_W = 147 for its null windows. Which convention is intended is not recorded anywhere; the label is inconsistent with the arithmetic either way."
}