Oncofusion

start / graph / ana-q120-the-mll1-n-and-menin-compartments-overlap-a-third-in-molm13-and-a-half-in-rs4-11

Analysisstrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q120-h3k79me2-second-line-and-the-mll1-epitope

The chromatin compartment that the field treats as 'the fusion's sites' is smaller and less shared in MOLM13 than in RS4;11: 467 of 1,389 MLL1-N-top bins also carry top menin (33.6%) against 692 of 1,390 (49.8%), and the MLL1-N and H3K79me2 compartments barely overlap in either line

also called the co-occupancy proportion · 33.6% against 49.8%

The places where the MLL1 antibody and the menin antibody both peak - the best available guess at where the fusion sits - are only a third of the MLL1 peaks in one cell line and half of them in the other. The compartments are not the same size and the histone tag sits somewhere else again.

The same thing, for a scientist

**Stated as proportions, because a raw intersection count says nothing without its denominator** [E1]. Both lines, top 0.1% of each track's own DMSO signal in a universe of ~1.39 million 2 kb bins: | | RS4;11 | MOLM13 | |---|---:|---:| | top 0.1% per track | 1,390 bins | 1,389 bins | | MLL1-N and MENIN both in top 0.1% | 692 = **49.8%** | 467 = **33.6%** | | MLL1-N and DOT1L both in top 0.1% | not computed | 565 = **40.7%** | | of the whole universe | 0.0498% | **0.0336%** | | fraction of the co-occupied set inside a deposited ROSE TSS window | not computed | **79.0%** | | fraction of the MLL1-N held set inside a ROSE TSS window | not computed | 86.0% | | fraction of the H3K79me2 held set inside a ROSE TSS window | - | **41.0%** | **Two things worth carrying.** **One: the proxy is weaker in MOLM13.** 'MLL1-N and menin co-occupied' is this project's standard stand-in for 'fusion-occupied' in the absence of a tag, and it covers half the MLL1-N compartment in RS4;11 and a third in MOLM13. Every cross-line comparison built on it - including this beat's headline - compares a half-set with a third-set. **Two: the mark lives somewhere else.** The MOLM13 H3K79me2 held set is **41.0%** inside a RefGene TSS window against 86.0% for the MLL1-N held set - broad gene-body domains against promoter summits. Q118 measured the same separation in RS4;11 from a different angle (Jaccard 0.0308 between the MLL1-N and H3K79me2 held sets, 21.4% CpG islands against 75.18%). **So every H3K79me2 figure in this beat and in Q118 is the mark measured at the fusion's compartment and not at the mark's own**, which is the right choice for the question and is also why a 'no change' there is compatible with large changes in the mark's own domains.

Facts and figures
scriptanalyses/q120-h3k79me2-second-line-and-the-mll1-epitope/arm1-molm13-ordering.py
statusrun
result summaryMLL1-N and MENIN co-occupancy is 33.6% of the MLL1-N held set in MOLM13 and 49.8% in RS4;11; the H3K79me2 held set is 41.0% TSS-proximal against 86.0% for MLL1-N.
cell lineMOLM13, compared with RS4;11
epigenetic readoutthe H3K79me2 compartment is 41.0% TSS-proximal and the MLL1-N compartment 86.0%, so the two marks occupy different chromatin and every H3K79me2 figure in this beat is measured at the fusion's compartment
mark writtenH3K79me2
shown for fusionKMT2A::MLLT3 and KMT2A::AFF1; the co-occupancy proxy for fusion occupancy is weaker in the former and neither reagent identifies the fusion
absence classnever-measured
absence class noteWhat the co-occupancy set is a proxy FOR - the fusion's actual binding sites in a human cell - has never been measured, so the proxy's accuracy is unknown in both lines. Needs a tagged human allele.
absence class assigned2026-09-23 (Q120)
How this could bite us back
  • If the co-occupancy proxy covers a third of the MLL1-N compartment in one line and a half in another, then 'the fusion's sites' is a different object in each line and any cross-line mechanism stated over it inherits that difference silently.
Still open
  1. Is the smaller MOLM13 overlap a biological difference or a signal-to-noise difference between a 3-day and a 4-day MENIN sample? The DOT1L intersection at 40.7% is pure 3-day and lands between the two lines' figures, which is suggestive and not decisive.
  2. What is in the two thirds of the MOLM13 MLL1-N compartment that carries no top-menin signal? Q118 measured the RS4;11 equivalent as not moving at all under drug (1.0422), which it read as the wild-type-dominated fraction. The same partition has not been computed for MOLM13.
Every source behind this entry (1)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset

    In MOLM13, 467 of 1,389 MLL1-N-DMSO top-0.1% bins are also in the MENIN-DMSO top 0.1% (33.62% of the held set, 0.0336% of the 1,389,438-bin universe) and 565 are also in the DOT1L top 0.1% (40.68%); 79.01% of the co-occupied bins, 85.96% of the MLL1-N held set and 41.04% of the H3K79me2 held set lie inside one of GSE127507's 21,461 deposited ROSE RefGene TSS windows. In RS4;11 Q118 records 692 of 1,390 (49.78%).

    What this evidence cannot show

    The two lines' universes are defined by their own inputs so the bin sets are not identical, though both are ~1.39 million bins. The MOLM13 MENIN track is a 4-day exposure and its DMSO arm defines the co-occupancy set, so a difference in DMSO signal-to-noise between a 3-day and a 4-day sample would change the intersection size; the DOT1L intersection (40.68%, pure 3-day) sits between the two figures and is the relevant control. Bethyl A300-086A, deposited as 'anti-MLL1n', an MLL1 N-TERMINUS epitope. The N-terminus is retained in every KMT2A fusion, so this track is the sum of the fusion and wild-type MLL1 in unknown proportion and CANNOT distinguish them. The immunogen residue range was not verified: three vendor routes were tried this session (fortislife.com two URL forms, bethyl.com/product/A300-086A) and all returned HTTP 404, so 'N-terminus' rests on the depositors' own target label and on the standard fact that MLL-FPs lack the C-terminal SET domain. Bethyl A300-105A anti-Menin cannot distinguish menin bound to the fusion from menin bound to wild-type KMT2A/KMT2B, nor from menin's non-KMT2A functions.

    Evidence that argues against this (1)
Connected to