Oncofusion

start / graph / exp-q118-a-tagged-kmt2a-aff1-chip-in-rs4-11-under-vtp-50469

Experimentmoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q118-menin-at-the-human-index-fusion

Tag the endogenous KMT2A::AFF1 allele in RS4;11 and ChIP the tag with and without VTP-50469 alongside MENIN and MLL1-N in the same chromatin - the one experiment that converts every occupancy claim in this literature from a reagent statement into a fusion statement

also called the tagged index fusion experiment · the experiment that resolves the 4-fold

Put a small molecular handle on the cancer fusion protein in the standard human cell line for this disease, then pull it down with and without the drug. That single experiment would say whether the drug actually takes the fusion off the DNA, which nobody currently knows for this fusion.

The same thing, for a scientist

**Design.** CRISPR knock-in of a 3xFLAG (or HA, or a dTAG/FKBP12-F36V degron plus tag) at the **3' end of the AFF1 portion of the endogenous der(11) fusion allele** in RS4;11, which is heterozygous for t(4;11) and retains a wild-type KMT2A allele. Tagging the fusion allele specifically - not KMT2A's N-terminus, which would tag both species - is the whole point. Single-cell clone, junction-sequenced, tag expression confirmed by immunoblot against the untagged parental line. **Then, from ONE chromatin preparation per condition, in parallel:** | ChIP | reagent | what it answers | |---|---|---| | anti-tag | anti-FLAG M2 or anti-HA | **the fusion's own occupancy** - the quantity nobody has in a human KMT2A::AFF1 line | | anti-MLL1 **N**-terminus | Bethyl A300-086A | reproduces the deposited GSE127507 measurement in the same cells | | anti-MLL1 **C**-terminus | a C-terminal antibody | **wild-type MLL1 only**, because the C-terminus is lost in the fusion | | anti-Menin | Bethyl A300-105A | reproduces the deposited menin measurement | | input | none | **in BOTH conditions** - the gap in GSE127507 | **Conditions:** DMSO and 330 nM VTP-50469 (or revumenib, for translational relevance) for 3 days, matching the deposited series exactly so the numbers are comparable, plus a 7-day arm to test whether the retained material eventually leaves. **n = 3 biological replicates per condition per ChIP.** GSE127507 is n = 1 throughout and this beat's every number is therefore spreadless; a previous beat in this project measured a 44.5-percentile spread across four replicates of one antibody in one line, and no number here is bounded against that. **A Drosophila or yeast spike-in, at a recorded ratio, with the scale factor deposited.** The depositors of GSE127507 ran a spike-in, judged it small and did not report it, so no cross-condition fold change is recoverable from the founding dataset of this drug class. **Reporting the scale factor is the cheapest single improvement over the existing data** and it is what would let this experiment report an amount rather than only a distribution. **The readouts, pre-specifiable now:** 1. **The fusion's within-sample share ratio at its own top 0.1% of blacklist-filtered 2 kb hg19 bins.** Above 0.6 supports decoupling; below 0.35 supports eviction; between them is the informative middle. With a spike-in, the same quantity as a true fold change. 2. **The Jaccard of the tag-held set with the MLL1-N and MLL1-C held sets**, which apportions the MLL1-N signal between the fusion and wild-type MLL1 - the number every claim in this field currently lacks. 3. **The 53 gained bins from this beat, scored at the tag.** Fusion-positive means redistribution is real and small. 4. **HOXA9 and MEIS1 at the tag**, against this beat's MLL1-N figures of 0.831 and 0.245 over floor and menin's 0.145 and 0.010. **Cost tier: medium.** One knock-in line (6-10 weeks), one ChIP-seq experiment with five targets x 2 conditions x 3 replicates plus inputs (~36 libraries), and the analysis is already written - this beat's `binsB.py` runs on it unchanged. **Why it has not been done.** The tagged-fusion route exists in mouse (GSE260742's FLAG-MLLAF9, which is why Q117's numbers are attributable) and has never been published for a human KMT2A::AFF1 line, which is the disease this lab studies. This graph already records `exp-q117-spike-in-normalised-tagged-fusion-chip-under-a-menin-inhibitor-in-a-human-kmt2a-aff1-line` as the same proposal reached from the murine side; **this entry is the version specified against the measurements that now exist**, with the comparator numbers, the missing input, the missing C-terminal arm and the missing replicates all named.

Facts and figures
designCRISPR knock-in of 3xFLAG or HA (optionally plus a dTAG degron) at the 3' end of the AFF1 portion of the endogenous der(11) KMT2A::AFF1 allele in RS4;11; single-cell clone, junction-sequenced, parental line compared for viability and drug sensitivity.
readoutChIP-seq from one chromatin prep per condition with five targets in parallel - anti-tag, MLL1 N-terminus (Bethyl A300-086A), MLL1 C-terminus, anti-Menin (Bethyl A300-105A) and input - at DMSO and 330 nM VTP-50469 for 3 days plus a 7-day arm, n = 3 biological replicates, with a Drosophila or yeast spike-in whose per-sample scale factor is DEPOSITED.
controlsinput in BOTH conditions (the gap in GSE127507); the untagged parental RS4;11 line for the anti-tag ChIP as a specificity control; an IgG arm (GSE127507 has none in any line); the MLL1 C-terminal arm as the wild-type-only track that allows the MLL1-N signal to be apportioned.
cost tiermed
timeline6-10 weeks for the knock-in line, 3-4 weeks for the ChIP-seq and sequencing, analysis already written (analyses/q118-menin-at-the-human-index-fusion/binsB.py runs on it unchanged)
pre specified primary readoutthe tagged fusion's within-sample share ratio at its own top 0.1% of blacklist-filtered 2 kb hg19 bins; >0.6 supports decoupling, <0.35 supports eviction
shown for fusionthe experiment is specified FOR KMT2A::AFF1 in RS4;11; the analogous experiment exists only for FLAG-tagged KMT2A::MLLT3 in a murine line (GSE260742)
comparator numbers already measured
meis1_over_floor_menin0.010meis1_over_floor_menin: 0.01hoxa9_over_floor_menin0.145hoxa9_over_floor_menin: 0.145menin_share_ratio_rs4110.200menin_share_ratio_rs411: 0.1999meis1_over_floor_mll1n0.245meis1_over_floor_mll1n: 0.245mll1n_share_ratio_at_meni…0.800mll1n_share_ratio_at_menin_held: 0.7996hoxa9_over_floor_mll1n0.831hoxa9_over_floor_mll1n: 0.831
comparator numbers already measured — the numbers behind the chart
{
  "menin_share_ratio_rs411": 0.1999,
  "mll1n_share_ratio_at_menin_held": 0.7996,
  "hoxa9_over_floor_menin": 0.145,
  "hoxa9_over_floor_mll1n": 0.831,
  "meis1_over_floor_menin": 0.01,
  "meis1_over_floor_mll1n": 0.245
}
How this could bite us back
  • If the experiment shows the fusion is retained, the menin-interface strategy this project is built around is shown to work on menin and not on the fusion, and the lab's central hypothesis needs restating rather than abandoning - the interface would still be the right thing to drug, but for a different reason and with a different pharmacodynamic endpoint.
  • If it shows the fusion is evicted, then the field's standard MLL1-N readout under-reports the drug's effect four-fold, and compounds triaged out on that readout were mis-triaged.
  • Either result invalidates a published class of claim, which is the strongest argument for doing it: there is no outcome in which the experiment is uninformative.
Still open
  1. Is the der(11) fusion allele in RS4;11 taggable at its 3' end without disrupting the AFF1 C-terminal domains that recruit the SEC? The breakpoint architecture of RS4;11's specific fusion would have to be resolved first, and this graph holds breakpoint entries that would inform it.
  2. Would a degron tag be better than an epitope tag? A dTAG arm gives the acute-degradation comparison that separates occupancy loss from protein loss, which is the ambiguity every ChIP in this field carries - at the cost of a larger insert and more perturbation risk.
  3. Is 330 nM for 3 days the right condition, or should the experiment match a clinically achievable revumenib exposure instead? Matching GSE127507 makes the numbers comparable; matching the clinic makes them relevant, and doing both doubles the arms.
Every source behind this entry (2)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset

    No tagged-fusion occupancy measurement exists for human KMT2A::AFF1 under a menin inhibitor: the deposited founding dataset GSE127507 uses an MLL1 N-terminus antibody and a menin antibody in all three of its lines, has no MLL1 C-terminal arm, no replicate, and no VTP-50469 input for RS4;11 - and the discrepancy between its two antibodies at the same bins in RS4;11 is 4.00-fold.

    What this evidence cannot show

    The claim that no such dataset exists is bounded by this beat's reading of GSE127507 and GSE127508's subseries relation, plus this graph's existing entries on menin-inhibitor occupancy data. NO systematic archive sweep for a tagged human KMT2A::AFF1 ChIP was run in this beat - not in GEO, not in SRA, not in ENCODE - so the absence is bounded by what this graph knows and not by a search. n = 1 per antibody per condition per line - there is no replicate anywhere in GSE127507 - so no number here carries a spread and none is precise.

    Evidence that argues against this (1)
    • A knock-in tag on one allele of a fusion in a cell line is itself a perturbation: it can alter fusion stability, complex assembly or drug sensitivity, and a degron tag more so. The experiment therefore measures a tagged fusion's occupancy and the untagged parental line has to be shown to behave the same way by viability and expression readouts, which adds work and is why the design above includes the parental comparison. Krivtsov AV et al. 2019
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset

    The comparator numbers this experiment would be read against already exist in the same cells: menin's within-sample share ratio at its own top 0.1% of RS4;11 2 kb bins is 0.1999 and MLL1-N's is 0.7996 at the same bins, with HOXA9 at 0.145 (menin) and 0.831 (MLL1-N) over an occupancy-matched floor and MEIS1 at 0.010 and 0.245.

    What this evidence cannot show

    NEITHER ANTIBODY IN THIS SERIES IS FUSION-SPECIFIC AND THE ANTIBODY NAME SAYS SO. MLL1n is anti-MLL1 N-terminus (Bethyl A300-086A, lots 5 or 6); the N-terminus is retained in KMT2A::AFF1, so that track is the sum of the fusion and wild-type MLL1 in unknown proportion. MENIN is Bethyl A300-105A Lot 8; menin binds the fusion, wild-type MLL1 and non-KMT2A partners. Q117 could attribute its numbers because its fusion was FLAG-tagged; nothing here can. Every occupancy statement in this beat is about MLL1-N-reactive material or menin-occupied sites, never about the fusion. No scale factor exists for any sample. The depositors state that a Drosophila spike-in was run on one batch, that fly/human ratios were 'found to be small', and that NO spike-in normalisation was applied; no number for that measurement is reported and none is deposited. Every cross-condition figure here is therefore a WITHIN-SAMPLE share or a within-sample rank, which cancel an unknown per-sample scale factor exactly, and never a fold change. n = 1 per antibody per condition per line - there is no replicate anywhere in GSE127507 - so no number here carries a spread and none is precise. The locus figures are from the deposited 6 kb TSS-window table and the genome-wide figures from 2 kb bins, so a tagged experiment should compute both to be comparable to each.

    Evidence that argues against this (1)
    • The comparator numbers come from two different instruments with two different universes - the locus figures from 6 kb TSS windows and the genome-wide figures from 2 kb bins - and a share ratio is not comparable across universes (this beat measured a 3.35-fold difference from the denominator alone). A tagged experiment read against the wrong one of these would appear to agree or disagree for reasons that have nothing to do with the fusion, which is why the design specifies computing both. Iqbal Lab oncofusion knowledge graph 2026
Connected to