The unnamed KMT2A-rearranged B-ALL line is SEM, and the deposition says KMT2A::AFF1 in a genotype field
also called GSE337426 provenance · USP22 MEIS1 cell lines · SEM MV4;11 USP22 ChIP
An entry on USP22 says the paper names a KMT2A-rearranged B-ALL line without specifying it, and does not say which fusion that line carries. The deposited sequencing names both: the line is SEM and the fusion is written out in full as KMT2A::AFF1 - the exact fusion this project is built around.
The same thing, for a scientist
`gene-usp22` carries a three-part caveat: the abstract *"names one KMT2A-rearranged B-ALL line for the screen without specifying it, does not state which KMT2A fusion that line carries, and does not name the chemical inhibitor used"* [E1]. GEO series **GSE337426** resolves the first two parts completely and leaves the third open. Ten *Homo sapiens* samples, contributed by Florian Perner, overall design *"ChIP sequencing of MEIS1 and USP22 in MV411 and SEM cell lines"*. Every sample carries `tissue: hematopoietic` and an explicit `genotype: KMT2A::AFF1`; five are `cell line: SEM, cell type: ALL` and five are `cell line: MV4;11, cell type: AML` [E2]. **This is the only deposition in this beat that states the index fusion in a structured genotype field**, and it is the second of three resolutions that land on the index system rather than an import. The chemical inhibitor is not named anywhere in the GEO metadata, so that third clause of the caveat stands [E3].
The ten samples
| GSM | Title | Cell line | Cell type | Genotype |
|---|---|---|---|---|
| GSM9855131 | MV411_input | MV4;11 | AML | KMT2A::AFF1 |
| GSM9855132-3 | MV411_MEIS1-ChIP rep1/2 | MV4;11 | AML | KMT2A::AFF1 |
| GSM9855134-5 | MV411_USP22-ChIP rep1/2 | MV4;11 | AML | KMT2A::AFF1 |
| GSM9855136 | SEM_input | SEM | ALL | KMT2A::AFF1 |
| GSM9855137-8 | SEM_MEIS1-ChIP rep1/2 | SEM | ALL | KMT2A::AFF1 |
| GSM9855139-40 | SEM_USP22-ChIP rep1/2 | SEM | ALL | KMT2A::AFF1 |
The design is unusually clean for this beat's purposes: the same two factors profiled in the same fusion across two lineages, so lineage is varied while fusion is held constant - the inverse of the confound that spoils most of the other depositions here 2.
A submitter error that a careless reader would propagate
The antibody fields are swapped relative to the sample titles in both lines. SEM_MEIS1-ChIP_rep1 and rep2 carry chip antibody: anti-USP22, Novus Biologicals, cat: NBP1-49644, lot: A-9, while MV411_USP22-ChIP_rep1 and rep2 carry chip antibody: anti-MEIS1, Abcam, cat: AB19867, lot: 1093686-2 2. Within each line, every non-input sample carries the same antibody string regardless of its title, so the field is filled per line rather than per sample.
This is the same failure class as the K563/K564/K565/K566 typographic cell types recorded in concept-resolving-an-unstated-system-from-geo, and it is the reason a GEO-sourced claim must name GEO as the authority. The cell line and genotype fields here are internally consistent and used by this entry; the antibody field is not, and this entry makes no claim from it.
| script | none - GEO accession-display text endpoint |
|---|---|
| status | run |
| result summary | 10 of 10 deposited samples are human with an explicit genotype field reading KMT2A::AFF1. The unnamed KMT2A-rearranged B-ALL line is SEM; the paired myeloid line is MV4;11. Fusion held constant across two lineages. The chemical inhibitor remains unnamed. |
inputs
[ "PMID 42418681", "GSE337426" ]
counts — the numbers behind the chart
{
"samples_total": 10,
"samples_human": 10,
"samples_kmt2a_aff1": 10,
"cell_lines": 2,
"lineages_represented": 2,
"samples_with_named_compound": 0
}entries affected
[ "gene-usp22" ]
import axes
[]
- USP22 supports H2B deubiquitination in normal B cells including during class-switch recombination, so a USP22 inhibitor in an infant B-lineage disease carries a plausible humoral-immunity toxicity that a leukaemia viability screen would not detect.
- MEIS1 protein abundance is downstream of HOXA-programme output that menin inhibitors already suppress, so a USP22 agent combined with a menin inhibitor may be redundant rather than additive, and the deposition contains no combination arm.
- Because the inhibitor is unnamed, this graph cannot predict resistance routes for it; a USP22 active-site mutation and a SAGA-module disruption would give very different escape spectra.
- Does USP22 occupancy at MEIS1 differ between the ALL (SEM) and AML (MV4;11) arms? The design holds fusion constant and varies lineage, which is the comparison this graph most often cannot make.
- Which chemical inhibitor did the paper use, and is it a SAGA-module macrocycle or a USP22-selective compound? Neither the deposition nor the abstract says, and cross-resistance predictions differ sharply between the two.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
The USP22 paper reports USP22 as a vulnerability regulating MEIS1 protein abundance and gene transcription in KMT2A-rearranged acute leukaemia, and its abstract does not specify which KMT2A-rearranged B-ALL line was used, which fusion that line carries, or the identity of the chemical inhibitor.
USP22 is a novel vulnerability regulating MEIS1 protein abundance and gene transcription in KMT2Ar acute leukemia. · Bröchtel S, Schneider C, Müller M, Villinger C, Plenge T, Gupta MK, Kesel LI, Scheich S, Wolf S, Cakir AY, Häupl B, Buettner F, Knapp S, Serve H, Stegmaier K, Perner F, Oellerich T, Cremer A · Blood 2026What this evidence cannot show
Identifiers taken from the NCBI esummary record fetched on 2026-09-19. The full text could not be read: no PMCID is present in either the PubMed esummary or the Europe PMC core record (isOpenAccess N, inEPMC N), consistent with a recent subscription Blood article. This entry therefore asserts nothing about the paper's Methods or its inhibitor.
Evidence that argues against this (1)
- USP22's H2B deubiquitinase activity within SAGA has well-characterised functions outside leukaemia, including class-switch recombination in normal B cells, so a B-ALL vulnerability attributed to USP22 needs to be separated from a general B-lineage requirement before it becomes a target. Li C et al. 2018
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.dataset
GEO series GSE337426 contains 10 Homo sapiens samples, all annotated 'tissue: hematopoietic' and 'genotype: KMT2A::AFF1', five from 'cell line: SEM, cell type: ALL' and five from 'cell line: MV4;11, cell type: AML', with the overall design stated as ChIP sequencing of MEIS1 and USP22 in MV411 and SEM cell lines.
USP22 is a novel vulnerability regulating MEIS1 protein abundance and gene transcription in KMT2Ar acute leukemia (GSE337426) · GEO submitters · NCBI Gene Expression Omnibus 2026What this evidence cannot show
Read on 2026-09-19 from the GEO accession-display text endpoint; series-level fields read from the targ=self record, which gives the title, contributor (Florian Perner), overall design and a submission date of 2026-07-02. Submitter-entered. The chip-antibody field is demonstrably mis-assigned between MEIS1 and USP22 samples in both lines, so no claim here rests on it. This series was found by a GEO keyword search on USP22 AND MEIS1, not by elink, which returns no GEO link for this PMID.
Evidence that argues against this (1)
- The swapped antibody annotations in this very series are a worked example of why submitter-entered metadata needs curation, and a systematic study of cell-line and strain fields across genomics depositions reached the same conclusion at scale. Rogic S et al. 2026
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset
The chemical inhibitor referred to in the paper's abstract is named nowhere in the GEO metadata for GSE337426, which contains no treatment or compound characteristic on any sample.
USP22 is a novel vulnerability regulating MEIS1 protein abundance and gene transcription in KMT2Ar acute leukemia (GSE337426) · GEO submitters · NCBI Gene Expression Omnibus 2026What this evidence cannot show
A bounded negative: the ten deposited samples are untreated ChIP and input, so the absence of a compound field is expected and says nothing about whether the paper used one. Published macrocyclic inhibitors of the SAGA deubiquitinating module exist and would be the natural candidates, but this entry does not assert which compound was used.
Evidence that argues against this (1)
- Potent macrocycle inhibitors of the human SAGA deubiquitinating module have been reported, so the chemical matter this paper could have used is a small and identifiable set - the identity is recoverable from the paper and is simply not recoverable from the deposition. Morgan M et al. 2022
resolves the unnamed KMT2A-rearranged B-ALL line to SEM with an explicit KMT2A::AFF1 genotype field
- related to →KMT2A::AFF1 (MLL-AF4), t(4;11)(q21;q23)
the only deposition in this beat that states the index fusion in a structured genotype field
USP22 is the SAGA H2B deubiquitinase opposing the RNF20 ligase activity resolved in the sibling entry
applies the route, and documents a second class of submitter annotation error
resolved source in the unstated-system worklist