Verdict: 'eleven cell lines, none infant-derived' is right about the screen and wrong about the material — at least ten infant KMT2A::AFF1 lines exist
also called infant model inventory · weak-spot re-adjudication
This project has been saying that no cell line of this leukaemia comes from a baby. Checking every line by its database accession shows at least ten do, including one from a nine-day-old and one that is already sitting in the collection the project's dependency data comes from.
The same thing, for a scientist
The standing weak spot reads: *'Eleven cell lines, none infant-derived, underpin every dependency conclusion.'* Re-adjudicated by resolving every relevant line through Cellosaurus by **accession**: the clause about the screened panel is **correct**, the clause about available material is **wrong**, and the panel is worse than the sentence says in a way the sentence does not capture. **1. The screened panel.** Eleven DepMap CRISPR-screened lines, ages 10Y, 65Y, 1Y, 31Y, 20Y, 5Y, 32Y, 20Y, 37Y, 5Y and *unspecified*. Ten have a recorded age and none is under one year; the eleventh (HB11;19) has no age at all. Six of the eleven are from adults aged 20-65. **And the eleven lines are nine individuals**: SEM/SEM-K2 are one patient and MOLM-13/MOLM-14 are one patient. **2. The index-fusion subset is three people.** Of the eleven, four carry KMT2A::AFF1 — MV4-11 (10Y, monocytic AML, FLT3-ITD), SEM (5Y), SEM-K2 (5Y, same patient as SEM) and RS4;11 (32Y, adult B-ALL). So this graph's index-fusion dependency evidence rests on **three individuals: a 32-year-old adult, a 10-year-old with myeloid disease, and one five-year-old counted twice.** **3. Infant material exists.** At least ten KMT2A::AFF1 lines from donors under one year are registered in Cellosaurus: PER-490 (9 days), PER-494 (11 months) and PER-485 (its relapse pair), PER-785 (2 months), ALL-PO (4 months), BLIN-3 (3 months), SCMC-L1 (4 months), KOCL-45 (8 months), KOCL-48 (5 months), KOCL-58 (5 months), KOCL-69 (<1 year) — eleven records representing about ten patients. **4. One of them is already in DepMap.** ALL-PO is DepMap ACH-002209, a CCLE collection line with methylation, exome, proteomics, drug-screening and RNA-seq data, and **no CRISPR gene-effect value**. The gap is a screening gap, not a derivation gap.
The corrected sentence
Replace:
Eleven cell lines, none infant-derived, underpin every dependency conclusion.
with:
Eleven screened cell lines representing nine individuals underpin every dependency conclusion. Ten have a recorded age at derivation and none is under one year; the eleventh has no recorded age. Only three of the nine individuals carry the index fusion, and one of those three was 32 years old. At least ten infant-derived KMT2A::AFF1 lines exist and none has been screened — including ALL-PO, which is already a DepMap collection line.
That version is longer and it is the one that tells a reader what to do next.
Why the wrong version was believable
Three mechanisms, each worth recording because each will recur.
A fusion-keyed Cellosaurus query does not return the infant lines. Searching KMT2A-AFF1 returns SEM, RS4;11, MV4-11, ALL-PO, BLIN-3, SCMC-L1, KOCL-45/48/58/69 and others — but none of the eight Perth PER lines, because Cellosaurus carries no Sequence variation: Gene fusion record for any of them 2. The single best-matched line in the world, from a nine-day-old, is invisible to the query that finds the ones we already use.
Disease-term and age fields disagree within one record. ALL-PO is AG 4M and DI Childhood B acute lymphoblastic leukemia — not the infant NCIt term the Perth lines carry. An inventory keyed on the disease term misses it; an inventory keyed on age finds it.
Catalogue availability, not biology, selected the panel. SEM, RS4;11 and MV4-11 are the only KMT2A::AFF1 lines with ATCC or DSMZ catalogue numbers. Every infant line in the list has none 3.
What this does and does not overturn
It does not overturn any dependency result. Nothing here says MEN1 is not a dependency; it says the population in which that was measured is narrower and older than the graph's phrasing implies, and it names the material that would test it.
It does sharpen the fix. 'No infant line exists' implies a derivation programme. 'Ten infant lines exist, none screened, one already in the collection' implies a screen.
It leaves one thing unresolved that could change the whole verdict. Cellosaurus records SEM's age as age at sampling, and SEM is generally described as derived at relapse. If that patient was diagnosed in infancy, the single most-used KMT2A::AFF1 line is infant disease and the weak spot is wrong in its stronger clause too. The record does not say.
| script | manual accession-resolution audit; no new script. Panel taken from analyses/depmap-kmt2a-rearranged-specificity/results.json |
|---|---|
| status | run |
| run date | 2026-09-19 |
| screened lines | 11 |
| screened individuals | 9 |
| screened index fusion lines | 4 |
| screened index fusion individuals | 3 |
| screened adult derived lines | 6 |
| screened infant derived lines | 0 |
| infant lines with published crispr screen | 0 |
| verdict | weak spot correct about the screened panel, incorrect about available material, and understates the panel's non-independence |
| result summary | Eleven screened lines are nine individuals; the index-fusion subset is three individuals, one of them a 32-year-old adult and one a 10-year-old with monocytic AML. At least ten infant-derived KMT2A::AFF1 lines exist, none screened, and one (ALL-PO, ACH-002209) is already a DepMap collection line with full omics and no gene-effect data. |
inputs
[ "https://api.cellosaurus.org/cell-line/<accession>?format=txt for CVCL_0095, CVCL_S906, CVCL_0093, CVCL_0064, CVCL_1069, CVCL_QW47, CVCL_7125, CVCL_3993, CVCL_6867, CVCL_3994, CVCL_3995, CVCL_8866, CVCL_QW66, CVCL_0081, CVCL_QW62, CVCL_A088, CVCL_AY52, CVCL_V162, CVCL_C9HE, CVCL_8227, CVCL_W986, CVCL_W984, CVCL_A8JU, CVCL_IZ03", "https://api.cellosaurus.org/search/cell-line?q=dr:<ACH id> for the eleven screened DepMap lines", "https://api.platform.opentargets.org/api/v4/graphql — depMapEssentiality screens for MEN1", "analyses/depmap-kmt2a-rearranged-specificity/results.json" ]
screened panel
[
{
"line": "MV4-11",
"accession": "CVCL_0064",
"depmap": "ACH-000045",
"age": "10Y",
"fusion": "KMT2A-AFF1",
"lineage": "monocytic AML"
},
{
"line": "OCI-AML-2",
"accession": "CVCL_1619",
"depmap": "ACH-000113",
"age": "65Y",
"fusion": "KMT2A-AFDN"
},
{
"line": "THP-1",
"accession": "CVCL_0006",
"depmap": "ACH-000146",
"age": "1Y",
"fusion": "KMT2A-MLLT3"
},
{
"line": "NOMO-1",
"accession": "CVCL_1609",
"depmap": "ACH-000168",
"age": "31Y",
"fusion": "KMT2A-MLLT3"
},
{
"line": "MOLM-13",
"accession": "CVCL_2119",
"depmap": "ACH-000362",
"age": "20Y",
"fusion": "KMT2A-MLLT3",
"same_individual_as": "CVCL_7916"
},
{
"line": "SEM",
"accession": "CVCL_0095",
"depmap": "ACH-000782",
"age": "5Y",
"fusion": "KMT2A-AFF1",
"same_individual_as": "CVCL_S906"
},
{
"line": "RS4;11",
"accession": "CVCL_0093",
"depmap": "ACH-000874",
"age": "32Y",
"fusion": "KMT2A-AFF1",
"note": "NCIt term is ADULT B-ALL"
},
{
"line": "MOLM-14",
"accession": "CVCL_7916",
"depmap": "ACH-001574",
"age": "20Y",
"fusion": "KMT2A-MLLT3",
"same_individual_as": "CVCL_2119"
},
{
"line": "SHI-1",
"accession": "CVCL_2191",
"depmap": "ACH-001647",
"age": "37Y",
"fusion": "KMT2A-AFDN"
},
{
"line": "SEMK2",
"accession": "CVCL_S906",
"depmap": "ACH-001735",
"age": "5Y",
"fusion": "KMT2A-AFF1",
"same_individual_as": "CVCL_0095"
},
{
"line": "HB11;19",
"accession": "CVCL_8227",
"depmap": "ACH-001736",
"age": "unspecified",
"fusion": "KMT2A-MLLT1"
}
]infant kmt2a aff1 lines that exist
[
{
"line": "PER-490",
"accession": "CVCL_W986",
"age": "9 days"
},
{
"line": "PER-785",
"accession": "CVCL_IZ03",
"age": "2 months"
},
{
"line": "BLIN-3",
"accession": "CVCL_QW47",
"age": "3 months"
},
{
"line": "ALL-PO",
"accession": "CVCL_1069",
"age": "4 months",
"in_depmap_collection": true
},
{
"line": "SCMC-L1",
"accession": "CVCL_7125",
"age": "4 months"
},
{
"line": "KOCL-48",
"accession": "CVCL_6867",
"age": "5 months"
},
{
"line": "KOCL-58",
"accession": "CVCL_3994",
"age": "5 months"
},
{
"line": "KOCL-45",
"accession": "CVCL_3993",
"age": "8 months"
},
{
"line": "KOCL-69",
"accession": "CVCL_3995",
"age": "<1 year"
},
{
"line": "PER-494",
"accession": "CVCL_A8JU",
"age": "11 months",
"same_individual_as": "CVCL_W984"
},
{
"line": "PER-485",
"accession": "CVCL_W984",
"age": "1 year (relapse)",
"same_individual_as": "CVCL_A8JU"
}
]- Adding infant lines to a dependency panel adds patients slowly: ten lines are about ten donors, and any selectivity statistic over that panel will be underpowered in the same way the current one is.
- Several of the infant lines carry TP53 mutations (ALL-PO homozygous R248Q, KOCL-45 R273C, KOCL-69 R181C) or NSD2 E1099K (SCMC-L1, KOCL-45), so screening them would substitute one set of co-lesion confounds for another rather than removing confounds.
- If a screen in infant lines reproduces the menin dependency, this beat will have added nothing to the target hypothesis while having cost real effort; if it does not, the project loses its central nomination instrument in its own population. Both outcomes are worth the screen and only one is comfortable.
- Was the SEM patient diagnosed in infancy? Cellosaurus stores age at sampling and SEM is described as relapse-derived; if she was an infant, the most-used index-fusion line in the world is infant material and this verdict changes again.
- What would a CRISPR screen in ALL-PO cost, and why has it not been done in a line already in the collection?
- Do any of the ten infant lines have a sequenced fusion junction? Four of the assignments are cytogenetic only, and PER-494's AFF1 was explicitly not demonstrated.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
The eleven KMT2A-rearranged lines in this project's DepMap specificity panel resolve to ages at derivation of 10Y (MV4-11, CVCL_0064), 65Y (OCI-AML-2, CVCL_1619), 1Y (THP-1, CVCL_0006), 31Y (NOMO-1, CVCL_1609), 20Y (MOLM-13, CVCL_2119), 5Y (SEM, CVCL_0095), 32Y (RS4;11, CVCL_0093), 20Y (MOLM-14, CVCL_7916), 37Y (SHI-1, CVCL_2191), 5Y (SEMK2, CVCL_S906) and unspecified (HB11;19, CVCL_8227); SEM and SEMK2 share an individual, as do MOLM-13 and MOLM-14, so eleven lines represent nine patients, and the four KMT2A::AFF1 lines represent three patients.
Cellosaurus ages and same-individual relationships resolved by DepMap cross-reference for the eleven-line KMT2A-rearranged panel · Bairoch A, et al. (SIB Swiss Institute of Bioinformatics) · Cellosaurus 2025What this evidence cannot show
Each of the eleven DepMap identifiers was queried against Cellosaurus on 2026-09-19 and the primary line name read back, so the mapping is accession-to-accession. Ages are ages at sampling as reported by depositors, not ages at diagnosis; for lines derived at relapse the two differ. The MOLM-13/MOLM-14 same-individual relationship was already established by this graph's own monocytic audit and is restated here, not newly found.
Evidence that argues against this (1)
- Cellosaurus itself flags that depositor-reported ages can be internally inconsistent: for Kid92 (CVCL_A088) the founding paper's text says 16 years and its Table 1 says 28, and the database records the discrepancy as a Caution. Any age table built from this source inherits that error class, and nothing in the eleven-line panel has been cross-checked against a founding report except SEM. Bairoch A et al. 2025
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
A Cellosaurus free-text query for KMT2A-AFF1 returns 29 records including SEM, RS4;11, MV4-11, ALL-PO (4M), BLIN-3 (3M), SCMC-L1 (4M), KOCL-45 (8M), KOCL-48 (5M), KOCL-58 (5M), KOCL-69 (<1Y), EU-8 (1Y) and AN4;11 (8M), but returns none of the eight Perth PER infant ALL lines, because those records carry no gene-fusion sequence-variation annotation; a separate query by line identifier resolves PER-490 to CVCL_W986 with age 9 days.
Cellosaurus search API: q=KMT2A-AFF1 (29 records) and per-line identifier lookups for PER-490, PER-485, PER-494, PER-785, PER-784, PER-826, PER-703, PER-910 · Bairoch A, et al. (SIB Swiss Institute of Bioinformatics) · Cellosaurus 2025What this evidence cannot show
Run on 2026-09-19. The 29-record result includes engineered derivatives (RS4;11 BCL2 variants, MV4-11 knockouts, the AND-1 hESC series, the IMSUTi002-A-2 iPSC) as well as patient-derived lines, so 29 is not a count of patients or even of independent lines. A free-text query also matches records mentioning the term for other reasons, which is why each candidate was then fetched by accession individually.
Evidence that argues against this (1)
- The absence of a fusion annotation on the Perth lines is not evidence that they were overlooked by curators — Cellosaurus annotates fusions from published molecular reports, and the Perth lines' partner assignments are cytogenetic. For PER-494 the founding paper states that AFF1 was not demonstrated by FISH. So the database's silence may be correct curation rather than a gap, and 'ten infant KMT2A::AFF1 lines' rests on cytogenetics for four of them. Cheung LC et al. 2021
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
ALL-PO (CVCL_1069, DepMap ACH-002209), derived from a 4-month-old female with KMT2A-AFF1 B-ALL, is a member of the DepMap/CCLE collection with deposited methylation, exome, drug-screening, proteomic, microarray and RNA-seq data, but returned no gene-effect value in a DepMap CRISPR screen query for MEN1 that returned 1,258 screened lines including SEM, SEM-K2, RS4;11, MV4-11 and HB11;19.
DepMap CRISPR (Chronos) gene-effect screens for MEN1 (ENSG00000133895) queried via the Open Targets Platform GraphQL API, cross-referenced to Cellosaurus CVCL_1069 · Broad Institute Cancer Dependency Map; Open Targets Platform; SIB Cellosaurus · Open Targets Platform / DepMap / Cellosaurus 2026What this evidence cannot show
Absence from one gene's screen list is strong but not conclusive evidence of absence from the screen collection, and the Open Targets release lags the DepMap portal. The five positive controls returning values from the same call are what make the negative interpretable. Run on 2026-09-19.
Evidence that argues against this (1)
- Even if ALL-PO were screened it would be n=1 and would carry a homozygous TP53 p.Arg248Gln, a genotype almost absent from primary infant KMT2A-rearranged ALL. A single screened infant line with a p53-null background would not settle an infant-versus-older question, so 'the gap is one screen away' is optimistic about what that screen could conclude. Bairoch A et al. 2025
- E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Infant-derived KMT2A::AFF1 cell lines are not merely registered but actively used: an eight-line panel generated from primary infant ALL patient samples has been characterised by RNA-sequencing and cytogenetics against matched primary specimens, screened against 62 anti-neoplastic agents, and tested in vivo in xenograft, with four of its eight lines carrying t(4;11) from three infants aged 9, 66 and 336 days at diagnosis.
Preclinical Evaluation of Carfilzomib for Infant KMT2A-Rearranged Acute Lymphoblastic Leukemia · Cheung LC, de Kraa R, Oommen J, et al. · Frontiers in Oncology 2021What this evidence cannot show
Full text read via PMC (PMC8082024); Tables 1-3 are the source for ages, karyotypes and immunophenotypes. The 62-agent screen figure comes from a later paper from the same group (PMID 39941894, abstract read). The t(4;11) assignments are cytogenetic; no sequenced fusion junction is reported for any of the four.
Evidence that argues against this (1)
- Four lines from three infants is the same arithmetic problem as the screened panel in miniature — PER-494 and PER-485 are one child at diagnosis and relapse. Adding the Perth lines to a dependency panel would add three patients, not four lines, and the field's total independent infant KMT2A::AFF1 donor count across every resource found here is roughly ten. Bairoch A et al. 2025
- refutes →The assumption the whole project rests on, tested at last: menin dependency is fusion-specific (run)
eleven screened lines are nine individuals; index-fusion subset is three, one adult
the infant line already in DepMap and never screened
youngest infant KMT2A::AFF1 line found, 9 days
inventory of material for the index disease
the screened panel's index-fusion arm is an adult, a 10-year-old with AML, and one child counted twice
- ← supportsBounded absence: no published CRISPR or dependency screen in any infant-derived KMT2A::AFF1 model
bounded absence underpinning the verdict's 'screening gap, not derivation gap'
audit trail for the verdict
the infant models catalogued there appear in no menin-inhibitor occupancy dataset