The experiment this deposit is one arm short of: MEN1 ChIP after fusion degradation with a PSIP1-depletion arm, a spike-in, replicates, and timepoints past three hours - four additions to an existing protocol that would settle why menin stays
also called the missing menin arm · PSIP1 knockdown plus degron · the replicate nobody ran
We found that menin stays stuck to DNA after the cancer fusion protein is destroyed, and the likely explanation is a second anchor holding it there. The experiment that would prove it is small: repeat the existing one, but also remove the suspected anchor, add an internal measuring standard, do it more than once, and look at later timepoints.
The same thing, for a scientist
**Design.** HCB1 (or any MLL-AF9-HA-FKBP12 line), a 2x2 of **dTAG-VHL 500 nM vs DMSO** crossed with **PSIP1 shRNA/degron vs control**, with **ChIP-seq for MEN1, PSIP1 and HA** at **0.5 h, 3 h, 24 h**, in **biological triplicate**, with an **exogenous chromatin spike-in**. **Each addition closes a specific, named gap in GSE173599:** | addition | what it closes | |---|---| | **PSIP1 arm** | the authors' LEDGF proposal has never been tested; this is the whole hypothesis | | **Spike-in** | GSE173599 has none, so nothing in it is on an absolute scale and a global ChIP efficiency change is indistinguishable from a global occupancy change (`U4`) | | **Triplicate** | n = 1 per arm throughout; the only floor measurable in the whole deposit is 4.6 points from two HA libraries, and it is technical rather than biological (`U1`) | | **24 h + later** | **there is no menin ChIP beyond 3 h anywhere**, so slow off-rate cannot be distinguished from fusion-independent anchorage | | **HA in every arm** | the positive control must be internal to each condition, not borrowed from another batch | | **PSIP1 ChIP** | tests the prediction that the tether itself is fusion-independent | **Readout.** Retention scored both as peak presence and as spike-in-normalised signal, **partitioned by whether a site is HA-occupied in the matched control** — the partition this beat identified as the missing analysis and could not perform for the hypothesis because the perturbation does not exist. **Cost tier: low.** No new cell line, no new reagent class, no animal work. The degron line, the antibodies and the protocol are all published, and PSIP1 knockdown in leukaemia lines is routine. This is roughly a 16-library ChIP-seq experiment plus knockdown validation.
What each outcome would mean, fixed in advance
- Menin retention collapses under PSIP1 depletion → LEDGF is the tether. The menin-inhibitor pharmacodynamic reading changes: the drug is attacking a buffered contact, and LEDGF becomes a combination target with a mechanistic rationale rather than a hopeful one.
- Menin retention is unchanged under PSIP1 depletion → the hypothesis is refuted and the wild-type-KMT2A alternative moves to the front. That is the more uncomfortable answer, because it implies a menin inhibitor is acting substantially on normal-cell complexes.
- Menin retention collapses by 24 h in the control arm alone → this was kinetics all along, menin has a slow off-rate, and the 3 h snapshot in GSE173599 is simply too early. The genome-wide independence claim in this beat would then need to be narrowed to 'at three hours'.
- Spike-in normalisation changes the answer → every retention figure in this corpus drawn from spike-in-free ChIP is suspect in a way no re-scoring can fix, and that is a larger finding than the menin question.
The cheaper half, runnable now with no new data
Two analyses would sharpen the hypothesis before anyone runs a wet experiment, and both use files already downloaded:
- Partition the existing menin retention by HA occupancy. Score MEN1 retention separately inside and outside the matched HA peak set from the same deposit. This converts this beat's genome-wide null into the target-restricted measurement that would be directly comparable to the authors' Figure 4B.
- Intersect retained versus lost menin sites with public LEDGF/PSIP1 ChIP in a KMT2A-rearranged human line. If retained menin sites are enriched for LEDGF co-occupancy, the hypothesis earns support for the price of a download; if they are not, it loses its cheapest evidence.
Neither was done in this beat, and both should precede the experiment above.
| design | 2x2 of dTAG-VHL 500 nM vs DMSO crossed with PSIP1 depletion vs control, in an MLL-AF9-HA-FKBP12 human line, biological triplicate, with exogenous chromatin spike-in, ChIP-seq for MEN1, PSIP1 and HA at 0.5 h, 3 h and 24 h. |
|---|---|
| readout | Menin retention scored as spike-in-normalised signal AND as peak presence, partitioned by HA occupancy in the matched control arm. Primary comparison: menin retention after fusion degradation, PSIP1-depleted versus PSIP1-intact. |
| controls | HA ChIP in every arm as an internal degradation control; PSIP1 ChIP to confirm the tether is itself fusion-independent; PSIP1 knockdown validated to below 20% of protein by immunoblot; DMSO arms for both PSIP1 conditions. |
| cost tier | low |
| timeline | 6-10 weeks in a lab with the degron line already in hand |
- A positive result nominates LEDGF as a combination target, and this graph already records that the only reported direct LEDGF binder is ebselen and that no LEDGF agent has ever been tested in menin-inhibitor-resistant disease. A validated target with no tractable chemistry is a slower win than it looks.
- A negative result is more consequential than a positive one: if menin is held by wild-type KMT2A rather than LEDGF, then menin inhibitors are acting substantially on normal-cell complexes, which speaks directly to on-target toxicity rather than to efficacy.
- Should the PSIP1 arm be a knockdown or a separation-of-function IBD mutant? The knockdown is faster and the mutant is cleaner, and PSIP1's pleiotropy argues for going straight to the mutant if the reagent exists.
- Is HCB1 the right background, or would a KMT2A::AFF1 degron line - which does not exist publicly - be worth building first, given that everything in this beat is partner-imported from an MLLT3 model?
- What spike-in? Without a standard choice this project cannot compare across its own future experiments, and it currently has none specified anywhere.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.dataset
GSE173599 contains a MEN1 ChIP pair at 3 h only, with no replicate, no spike-in, no later timepoint and no PSIP1 arm - so each addition in this design corresponds to a gap that actually exists in the only relevant public deposit.
GSE173599 series supplementary file list, verified sample by sample 2026-09-23 · Olsen SN, Armstrong SA, et al. · NCBI Gene Expression Omnibus 2022What this evidence cannot show
The absence is bounded to this deposit's deposited peak tables. The GEO series overall_design claims 4-day DOT1L and MENIN ChIP in HCB1 that is not present as a peak table, so processed data may exist that was not deposited, and raw BAM-equivalent TDF tracks (23 GB) were not examined. The absence of a PSIP1 degron arm anywhere in the public record was not searched for exhaustively.
Evidence that argues against this (1)
- If a PSIP1-depletion plus MLL-fusion-degron experiment already exists in another deposit, this design is a rediscovery rather than a gap. That search was not run and should be before the experiment is commissioned. Oncofusion knowledge graph 2026
- E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
The authors themselves propose LEDGF anchorage as the explanation for menin's slower dissociation, making this the experiment their own discussion asks for.
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 2022What this evidence cannot show
A discussion-section proposal phrased as 'could explain', not a result. The authors ran no LEDGF experiment. This experiment is motivated by their hypothesis, not by their data.
Evidence that argues against this (1)
- PSIP1 depletion has pleiotropic effects - this graph records roles in R-loop suppression, DNA end resection, splicing coupling and polycomb recruitment - so a menin retention change under PSIP1 knockdown is not automatically a tether result. A separation-of-function PSIP1 IBD mutant would be the cleaner reagent and should replace the knockdown if the first result is positive. Oncofusion knowledge graph 2026
The decisive experiment, and it is four additions to a protocol that already exists: a PSIP1 arm, a spike-in, replicates, and timepoints past three hours.
Each addition in the design closes a named gap in the only relevant deposit, and the later-timepoint arm closes this one: without it, menin's persistence at 3 h cannot be told apart from a slow off-rate.
The PSIP1 arm of this design is the chromatin-occupancy counterpart to that absence: LEDGF removal has never been tested in established disease, and it has never been tested against a degraded fusion either.