CDK9 peak widths more than double after fusion degradation, firing the pre-registered re-score trigger
For one of the measured proteins, CDK9, the detected regions became two and a half times wider after treatment - a known sign that the signal has collapsed and the detection software is guessing. Any number computed from CDK9's regions has to be reported with that warning attached.
The same thing, for a scientist
The pre-registration set a re-score trigger at a treated/control median peak width ratio of 1.3. **CDK9 fires it at 2.46** (control 1,220 peaks, median width 2,102 bp; treated 646 peaks, median width 5,166 bp). No other non-HA channel comes close: MEN1 0.93, pS5 0.93, H3K79me2 0.96, Pol II 1.01, pS2 1.04, AFF4 1.05, DOT1L 1.13. The three HA channels also fire it (2.79, 1.93, 2.18), which is the expected behaviour of a destroyed protein and is why they are used only as a positive control. CDK9's peak-based figures (20.6% retention overall, 34.8% at HA-occupied sites) are therefore reported flagged and the signal measure takes precedence: on the authors' per-gene table CDK9 gives 0.687 at target genes vs 1.452 at non-occupied genes, the largest raw stratum gap of any channel (76.5 points) - which is itself suspicious and is discussed below.
Why the CDK9 signal figure is also not to be quoted alone
CDK9's non-occupied stratum median is 1.452, i.e. the housekeeping normalisation has pushed non-target CDK9 signal up by 45%. That is the signature of a channel where a large fraction of total signal left the target genes and the normalisation redistributed it, or of a library where depth differences dominate. The CDK9 libraries are the shallowest in the deposit and were sequenced alone in batch 200211. Both CDK9 readouts - peak and signal - are therefore weaker than the corresponding MEN1 and DOT1L readouts, and the beat's conclusions do not rest on CDK9.
Scale and limits
No spike-in exists anywhere in GSE173599, so every number is a within-sample share, and the treated arms are scaled by a 21-gene housekeeping panel whose own raw 3 h H3K27ac ratio averages 0.809 and is forced to 1. A housekeeping normalisation DEFINES the global change to be zero, so only the DIFFERENCE between strata is measurable; a uniform genome-wide loss of any of these proteins would be invisible here. n = 1 per arm for every channel except HA, which has two independent 3 h batches. The cells are HCB1, transformed by KMT2A::MLLT3 (MLL::AF9) - NOT the index fusion of this project, KMT2A::AFF1 - and a prior beat in this corpus found that per-gene ordering transfers between MLLT3 and AFF1 lines while magnitude does not.
| script | analyses/menin-retention-by-ha-occupancy/analyze_peaks.py |
|---|---|
| status | run |
| result summary | CDK9 treated/control median peak width ratio 2.46, above the pre-registered 1.3 trigger. Peak figures flagged. Signal figures (0.687 occupied vs 1.452 non-occupied) carry their own normalisation caveat. |
| peak width ratio note | CDK9 2.46 (FLAGGED). Comparators: MEN1 0.93, pS5 0.93, H3K79me2 0.96, Pol II 1.01, pS2 1.04, AFF4 1.05, DOT1L 1.13, HA 1.93-2.79 (all flagged). |
inputs
[ "GSE173599 MACS peak tables", "GSE173599_TSS_gene_body_signal_Figure3.xlsx" ]
- P-TEFb is required at essentially every active gene, so any intervention that removes CDK9 broadly rather than at fusion targets is a general transcription toxicity, not a targeted therapy - and this channel is the one whose measurement is least trustworthy in the deposit that would be used to argue either way.
- Is CDK9's width inflation a depth artefact? Subsampling the control library to the treated library's depth and re-calling would separate depth from biology, and the peak files carry the fragment counts needed to do it.
- E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset
CDK9's treated/control median peak width ratio is 2.46, against 0.93-1.13 for every other non-HA channel in the same deposit.
GSE173599 - Chromatin state changes induced by targeted degradation of MLL-AF9 are phenocopied by combined DOT1L and MENIN inhibition [ChIP-Seq] · Olsen SN, Hatton C (Armstrong lab, Dana-Farber) · NCBI GEO (SubSeries of GSE173600) 2022What this evidence cannot show
1,220 control and 646 treated peaks - the shallowest channel in the deposit, sequenced in its own batch (200211, hg38). A width ratio computed on few peaks is itself noisy.
Evidence that argues against this (1)
- CDK9 is a subunit of P-TEFb and is expected to be lost from target promoters when elongation stops, so a genuinely collapsing channel is the parsimonious explanation for the width inflation - which is exactly why the trigger exists: collapse and artefact look the same. Olsen SN et al. 2022
- E2weak confidence 2 of 5: Preliminary, small, indirect, or not yet peer reviewed.dataset
On the authors' per-gene table CDK9 gives 0.687 at target genes and 1.452 at non-occupied genes, the largest raw stratum gap of any channel.
GSE173599_TSS_gene_body_signal_Figure3.xlsx - per-gene housekeeping-normalised TSS and gene-body ChIP signal, DMSO vs dTAG-VHL 3 h and 24 h, 16 channels · Olsen SN et al. · NCBI GEO supplementary file 2022What this evidence cannot show
A non-occupied median of 1.45 means the normalisation raised non-target signal by 45%, which no other channel does; this figure should not be quoted without that caveat. n=1 per arm.
Evidence that argues against this (1)
- A non-occupied median of 1.452 could be a genuine redistribution of P-TEFb to other promoters when elongation stops at fusion targets, which is a published expectation for the super elongation complex rather than an artefact. Olsen SN et al. 2022
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
CDK9 is the catalytic subunit of P-TEFb and is recruited to MLL-fusion target promoters as part of the super elongation complex.
MLL::AF9 degradation induces rapid changes in transcriptional elongation and subsequent loss of an active chromatin landscape · Olsen SN, Godfrey L, Healy JP, Choi YA, Kai Y, Hatton C, Perner F, Haarer EL, Nabet B, Yuan GC, Armstrong SA · Molecular Cell 82(6):1140-1155.e11 2022What this evidence cannot show
Background claim, established in the field and restated by the authors; not measured by this analysis.
Evidence that argues against this (1)
- CDK9 is recruited to most active promoters independently of any MLL fusion, so its presence at fusion target genes is not by itself evidence of fusion-directed recruitment. Olsen SN et al. 2022
- supports →Peak-call retention tracks peak strength, not fusion occupancy, and that inverts the answer
names the one channel that fires the pre-registered width re-score trigger
CDK9 peak and signal readouts in this deposit are both weaker than MEN1 and DOT1L and are not load-bearing
P-TEFb catalytic subunit measured at fusion target promoters after degradation