MYO1F — unconventional myosin IF, a KMT2A partner with neither a transactivation domain nor a dimerisation domain
also called unconventional myosin-If · myosin type 1F · MYO1F_HUMAN · MYOF1 (typographic variant used in the 2023 recombinome's partner list)
MYO1F is a tiny motor protein that walks along the cell's actin cables — the kind of machine immune cells use to crawl and squeeze through tissue. When it is fused to KMT2A it makes leukaemia in babies, and it is one of only two such partners that the people who discovered it said outright has no gene-switching part and no self-sticking part either.
The same thing, for a scientist
MYO1F (UniProt O00160, MYO1F_HUMAN, 1098 aa) is a class-I unconventional myosin: a myosin motor domain at aa 17-690 containing ATP-binding and actin-binding sequences (actin-binding region aa 579-589), a neck with a single IQ motif at aa 693-722, and a tail comprising TH1 at aa 728-917 and an SH3 domain at aa 1041-1098 [E1]. Its founding fusion report described a gene of at least 28 exons whose ~4 kb transcripts are expressed at roughly constant level in most tissues but in only **3 of 22 leukaemic cell lines**, and stated explicitly that the MLL-MYO1F fusion protein contains almost the entire MYO1F yet **C-terminal MYO1F has neither the transactivation domain nor the dimerisation domain found in various MLL fusion partners** [E2]. That sentence makes MYO1F the second partner in this beat, alongside MLLT11, that its own discoverers flagged as unexplained by either competing model. The fusion is not a one-off: t(11;19)(q23;p13) fusing MLL with MYO1F was subsequently reported to be **recurrent in infant acute myeloid leukaemias** [E3], and the 3401-patient recombinome records 5 patients, all AML, 4 of them infants [E4]. MYO1F is not a dependency: Chronos mean -0.130 across 1208 lines, rank 16,967 of 17,728 for KMT2A selectivity — i.e. among the least KMT2A-selective genes in the genome — with no dependency in either lineage half [E5].
This is a prediction, not a measurement. No one has seen this protein in this shape. The colours say how much the software trusts itself, and that is worth reading before anything else — the orange stretches are places the model is effectively saying this part has no fixed shape.
I have never read one of these pictures before
A protein is one long chain of amino acids that folds up. The ribbon traces that chain from one end to the other; it is not a wire, it is a simplification of where the backbone runs. Drag to rotate, scroll to zoom.
Domains are the stretches of chain that do a particular job, and they are what the coloured blocks above the picture mark out. Click one and the view flies to it. In this project the important question about any domain is usually whether the fusion still has it — the labels say lost where it does not.
Anything in yellow sticks is a bound small molecule, which is usually the drug.
What the founding paper said, and why it matters more than its patient count
Taki and colleagues analysed a complex translocation involving chromosomes 7, 11, 19 and 22 in infant acute monocytic leukaemia and identified MLL on 11q23 fused to MYO1F on 19p13.2-13.3. Their structural description is unusually direct about what is absent: the fusion protein contains almost the entire MYO1F, however, C-terminal MYO1F has neither the transactivation domain nor the dimerisation domain found in various MLL fusion partners 2.
That is the same shape of problem as MLLT11: a partner that is clinically real and mechanistically unaccounted for. The difference is that MLLT11 is 90 residues with nothing in it, while MYO1F contributes roughly 1000 residues of motor, neck and tail — including an SH3 domain and a TH1 membrane-binding tail. So the two counterexamples are not redundant. MLLT11 asks whether a KMT2A fusion needs any partner module at all; MYO1F asks whether a large, structured, actin-motor module can substitute for one, and if so, how.
The authors' own conclusion was a request rather than an answer: "Further analysis of this novel type of MLL fusion protein would provide new insights into leukemogenesis" 2. Twenty-one years later, the searches for this beat found no such analysis — two PubMed records intersect (MLL OR KMT2A) with MYO1F and both are case or cohort reports 6.
Two nomenclature traps in one gene
This gene is a compact illustration of both failure modes the project has documented.
Under-counting by typo, not by retirement. The 2023 recombinome's list of 48 other KMT2A rearrangements in AML prints "5x MYOF1" where its own Table 1 row reads MYO1F with a total of 5 4. MYOF1 is not an HGNC symbol; MYOF is a different gene (myoferlin). A reader searching the paper's prose for MYO1F finds the Table 1 row and misses the prose count, and a text-mining pipeline keyed to gene symbols would score the prose mention as myoferlin or as nothing. The inherited enumeration for this beat recorded MYO1F as "1 PubMed record, nothing to read"; the correct count on the current PubMed index is 2 records, both directly about this fusion, one of them establishing recurrence 6.
Cytogenetic impersonation. MYO1F is the fourth KMT2A partner on 19p13, and the founding report states plainly that at the cytogenetic level it may be difficult to distinguish MLL-ENL, MLL-ELL, MLL-EEN and MLL-MYO1F fusions created by t(11;19)(q23;p13), so cases of t(11;19) lacking a known fusion gene may turn out to be this one 2. That is the fourth documented impersonation route in this project, after MLL-SEPTIN6 on the der(X), KMT2A::SEPTIN11 producing an ordinary-looking t(4;11), ABI1 being indistinguishable from MLLT10 at 10p11-12, and t(11;17)(q23;q21) covering both MLLT6 and LASP1.
Expression, localization, cell type, cell state, epigenetics
Tissue expression. Approximately 4 kb MYO1F transcripts appeared at constant level in most tissues examined by Northern blot, but MYO1F was expressed in only 3 of 22 leukaemic cell lines 2. That combination — broadly expressed in normal tissue, largely silent in leukaemia lines — is unusual and is the opposite of what a partner whose own function mattered to the disease would look like.
Subcellular localization. Not reported for the fusion protein, and not reported for human MYO1F in any source read here. Class-I myosins in general are membrane- and actin-associated through a TH1 tail; the UniProt record for MYO1F annotates the TH1 tail at aa 728-917 and an actin-binding region at aa 579-589 but no experimentally determined human localisation 1. This entry states the absence rather than importing class-level expectations.
Subnuclear localization. Not reported, for either the wild-type protein or the fusion.
Cell type and cell state. Patient-level only: infant acute monocytic leukaemia in the index case 2 and infant AML in the recurrent series 3; all 5 recombinome patients are AML, 4 of them infants 4. No experimental model, no transformation assay, no cell-of-origin work exists.
Epigenetics. Nothing. No H3K79, H3K4, HOXA ChIP or transcriptome for MLL-MYO1F or for any MYO1F-rearranged patient sample.
Essentiality: among the least KMT2A-selective genes in the genome
MYO1F ranks 16,967 of 17,728 in the project's genome-wide KMT2A-selectivity scan — the 95.7th percentile, meaning it is one of the genes most clearly less required in KMT2A-rearranged lines than in other leukaemia lines. Chronos mean is -0.130 across 1208 lines; KMT2A-rearranged mean -0.058 against -0.162 in other leukaemia lines, a positive difference of +0.104 5.
Split by lineage, as the myeloid-confound analysis requires: +1.18 SD in the 8-line myeloid half and -0.00 SD in the 5-line B-lineage half 5. Neither half shows a dependency; the myeloid half shows a mild anti-dependency on a small absolute effect, which is noise around a gene that leukaemia lines do not need.
The ceiling on all of this is the same as for every partner here: no MYO1F-rearranged line exists in DepMap, and the founding report's finding that MYO1F is expressed in only 3 of 22 leukaemic lines means most of the panel may not express the gene at all — in which case a CRISPR knockout of it measures nothing.
| uniprot | O00160 |
|---|---|
| uniprot verified | O00160 = MYO1F_HUMAN, PRIMARY gene name MYO1F, no synonyms recorded, 1098 aa, Homo sapiens. Fetched from the UniProt REST API on 2026-09-19 and the primary gene name and length read back from the record. The length agrees independently with the 1098-residue prediction in the founding cloning report (PMID 15897884). |
| length AA | 1098 |
| ensembl | ENSG00000142347 |
| ensembl verified | ENSG00000142347 returned as the gene-level xref for symbol MYO1F by Ensembl REST xrefs/symbol/homo_sapiens on 2026-09-19, and independently confirmed to resolve to approvedSymbol MYO1F by the Open Targets target query in the lineage-split run. |
| PDB note | No structure of MYO1F, of any MYO1F domain, or of the fusion. |
| alphafold ID | AF-O00160-F1 |
| domains note | No coiled coil, leucine zipper or other oligomerisation module is annotated, and no chromatin-binding or transactivation domain. The founding report additionally describes a TH2 domain in the tail which the current UniProt feature set does not annotate; it is omitted here rather than given guessed coordinates. |
| localization | NOT ESTABLISHED for human MYO1F in any source read here. The TH1 tail (aa 728-917) and actin-binding region (aa 579-589) are class-I myosin membrane- and actin-association modules by annotation, but no experimental human localisation is cited. |
| subnuclear localization | NOT REPORTED, for either the wild-type protein or the fusion. |
| epigenetic readout | NONE PUBLISHED for MLL-MYO1F or for any MYO1F-rearranged patient sample. |
| normal expression | ~4 kb transcripts at apparently constant level in most tissues examined by Northern blot; expressed in only 3 of 22 leukaemic cell lines |
| family | class I unconventional myosin |
| nomenclature note | MYO1F has no retired HGNC symbol, so the retired-alias failure mode does not apply. It has a different one: the 2023 recombinome's AML prose prints the count as '5x MYOF1', a typographic variant that is not an HGNC symbol and that collides with the real, unrelated gene MYOF (myoferlin). Symbol-keyed text mining over that sentence would score the mention wrongly. |
PDB IDS
[]
domains — the numbers behind the chart
[
{
"name": "Myosin motor domain",
"start": 17,
"end": 690
},
{
"name": "Actin-binding region",
"start": 579,
"end": 589
},
{
"name": "IQ motif",
"start": 693,
"end": 722
},
{
"name": "TH1 domain",
"start": 728,
"end": 917
},
{
"name": "SH3 domain",
"start": 1041,
"end": 1098
}
]essentiality
{
"depmap_common_essential": false,
"chronos_mean": -0.1301,
"chronos_n_lines": 1208,
"kmt2a_selectivity_rank": 16967,
"kmt2a_selectivity_of": 17728,
"kmt2a_r_mean": -0.0576,
"other_leukaemia_mean": -0.1618,
"lineage_split_z_kmt2a_myeloid": 1.18,
"lineage_split_z_kmt2a_b_lineage": 0,
"lineage_split_verdict": "no dependency in either half; if anything a mild anti-dependency in the myeloid half, on a small absolute effect",
"expression_caveat": "MYO1F was expressed in only 3 of 22 leukaemic cell lines in the founding report, so most DepMap leukaemia lines may not express it and the CRISPR null may be an expression artefact rather than a biological one",
"mouse_ko": "No Myo1f mouse phenotype was sought or found in any source read for this entry."
}recombinome
{
"dcal_recombinome_n": 5,
"dcal_recombinome_rank": 22,
"dcal_recombinome_total_patients": 3401,
"infant_aml": 4,
"paediatric_aml": 1,
"all_cases": 0,
"adult_cases": 0,
"spliced_fusion_cases": "1 of 5",
"historical_table_entry": "t(11;19)(q23.3;p13.2), 19p13.2, MYO1F, Lo Nigro et al. (2002), AML"
}depmap kmt2a selectivity
{
"chronos_mean_kmt2a_rearranged": -0.0576,
"chronos_mean_other_leukaemia": -0.1618,
"difference": 0.1042,
"pan_essentiality_all_lines": -0.1301,
"selectivity_rank": 16967,
"rank_denominator": 17728,
"provenance": "Joined 2026-09-21 from analyses/depmap-kmt2a-selective-scan-full (generated 2026-09-19T10:12:06Z), the project's own genome-wide scan over 11 KMT2A-rearranged and 43 other leukaemia DepMap lines. NOT re-fetched and NOT independent of that scan: every caveat it carries applies here, including that the KMT2A-rearranged arm is a small number of lines drawn from nine individuals, that the rank denominator is 17,728 genes, and that a difference is not a dependency. Added because only 26 of 308 gene entries carried this number, which made any KMT2A-selectivity question answerable for 8% of the corpus.",
"how_to_read": "A negative difference means MORE required in KMT2A-rearranged lines than in other leukaemia lines. A rank near 17,728 is ANTI-selective. Neither is evidence of a fusion-driven dependency on its own: check pan_essentiality_all_lines before calling anything selective, and note that a paralogue-redundant gene can score near zero for reasons that have nothing to do with this disease - see ana-the-paralogue-redundancy-excuse-is-not-tested-and-my-proxy-decided-the-answer."
}- MYO1F is expressed in most normal tissues, and class-I myosins are central to immune-cell motility. Any agent directed at the wild-type protein rather than the fusion would risk impairing leukocyte migration in patients who are already immunocompromised — and the DepMap null offers no reassurance, because most panel lines may not express the gene.
- If this fusion transforms without a dimerisation module, then oligomerisation is not necessary for KMT2A leukaemia and an inhibitor designed against an oligomerisation interface would have no effect in these patients, who would be indistinguishable from MLL-ENL patients on karyotype.
- The cytogenetic ambiguity at t(11;19)(q23;p13) means MYO1F patients can be enrolled and analysed as ENL or ELL patients, so a partner-specific failure of a targeted agent could be absorbed invisibly into a larger subgroup's results.
- Does MLL-MYO1F transform? Like KMT2A::MLLT11, this fusion has never been expressed in a cell. It is the second partner whose discoverers stated it has neither a transactivation domain nor a dimerisation domain, and testing either one would put both competing models at risk for the first time.
- Does the MYO1F tail oligomerise? The SH3 domain and the TH1 membrane-binding tail are both interaction modules, and the 2005 'no dimerisation domain' conclusion was drawn from motif recognition, not measurement. A sedimentation-velocity run on the retained MYO1F portion would settle it.
- Does the retained myosin motor domain make the fusion a mechanoenzyme? No KMT2A fusion is known to carry an ATP-driven motor, and nobody has asked whether the motor domain is required for transformation or is simply along for the ride.
- How many karyotypic t(11;19)(q23;p13) cases assigned to ENL, ELL or EEN are actually MYO1F? The founding report flags the ambiguity explicitly and nobody has re-typed a cohort.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
Human MYO1F is UniProt O00160 (MYO1F_HUMAN), primary gene name MYO1F with no synonyms recorded, 1098 aa, Homo sapiens, with a myosin motor domain at aa 17-690, an actin-binding region at aa 579-589, an IQ motif at aa 693-722, a TH1 domain at aa 728-917, disordered regions at aa 913-1009 and 1021-1044, a basic compositional bias at aa 924-937, and an SH3 domain at aa 1041-1098; no PDB structure is cross-referenced.
UniProtKB O00160 (MYO1F_HUMAN) Unconventional myosin-If, Homo sapiens, 1098 aa · UniProt Consortium · UniProtKB 2026What this evidence cannot show
Canonical isoform. UniProt annotates no coiled coil and no dimerisation module, which is consistent with the founding fusion report's explicit statement, but an absent annotation is a curation fact and not a demonstration. Resolved by fetching the accession and reading back the PRIMARY gene name (MYO1F) and length (1098 aa); the length agrees independently with the 1098-residue prediction in the founding cloning report, which is the cross-check the project's accession rule asks for.
Evidence that argues against this (1)
- The founding report describes a tail with TH1, TH2 AND SH3 domains, whereas the current UniProt feature set lists TH1 and SH3 but no TH2 — so the domain inventory for this protein is not stable across sources, and any figure drawn from these coordinates omits a region the discovering laboratory considered a domain. Taki T et al. 2005
- E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In an infant acute monocytic leukaemia with a complex translocation involving chromosomes 7, 11, 19 and 22, MLL on 11q23 was fused to MYO1F on 19p13.2-13.3; MYO1F consists of at least 28 exons and was predicted to encode a 1098-amino-acid protein with an N-terminal head domain containing ATP-binding and actin-binding sequences, a neck with a single IQ motif and a tail with TH1, TH2 and SH3 domains; approximately 4 kb transcripts were expressed at apparently constant level in most tissues examined but MYO1F was expressed in only 3 of 22 leukaemic cell lines; the MLL-MYO1F fusion protein contains almost the entire MYO1F, however C-terminal MYO1F has NEITHER the transactivation domain NOR the dimerisation domain found in various MLL fusion partners; MYO1F is the fourth partner gene of MLL on 19p13, and at the cytogenetic level it may be difficult to distinguish MLL-ENL, MLL-ELL, MLL-EEN and MLL-MYO1F fusions created by t(11;19)(q23;p13).
The MYO1F, unconventional myosin type 1F, gene is fused to MLL in infant acute monocytic leukemia with a complex translocation involving chromosomes 7, 11, 19 and 22. · Taki T, Akiyama M, Saito S, Ono R, Taniwaki M, Kato Y, et al. · Oncogene 2005What this evidence cannot show
A single patient, and a complex four-chromosome rearrangement that is atypical even for this partner. The domain inventory is a 2005 sequence prediction. The 'neither transactivation nor dimerisation domain' statement is an argument from absence of recognised motifs in 2005 annotation, not an experimental result — no transformation assay, no self-association measurement and no reporter assay was performed. Read from the abstract.
Evidence that argues against this (1)
- Absence of a RECOGNISED dimerisation motif is a weak negative in exactly this partner class: GAS7's transforming module was called a coiled coil in 2003 and later structural work showed its self-assembly determinant is an F-BAR sheet-forming surface, a module type that 2005 annotation would not have flagged as dimerising. MYO1F's TH1 tail and SH3 domain are both interaction modules whose oligomeric behaviour in the fusion has never been measured. Hanawa-Suetsugu K et al. 2019
- E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper
The t(11;19)(q23;p13) fusing MLL with MYO1F is recurrent in infant acute myeloid leukaemias.
The t(11;19)(q23;p13) fusing MLL with MYO1F is recurrent in infant acute myeloid leukemias. · Duhoux FP, Ameye G, Libouton J, Bahloula K, Iossifidis S, Chantrain CF, et al. · Leuk Res 2011What this evidence cannot show
PubMed carries NO abstract for this record, so the claim asserted here is exactly the content of the title and nothing more: recurrence, infant AML, t(11;19)(q23;p13). The number of cases, the breakpoints and the clinical course are not asserted. Confidence is moderate for that reason, not because the finding is doubtful.
Evidence that argues against this (1)
- 'Recurrent' rests on small absolute numbers: the 3401-patient recombinome records only 5 MYO1F patients in total, all AML, so recurrence here means a handful of cases and not an established frequency, and at that count a single referral pattern determines the apparent age distribution. Meyer C et al. 2023
- E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
In Table 1 of the 3401-patient 2023 KMT2A recombinome, MYO1F is the 22nd-ranked direct translocation partner with 5 patients — 4 infant AML and 1 paediatric AML, with no ALL and no adult cases — the paper's AML prose list prints the same count as '5x MYOF1', and MYO1F is separately recorded as contributing 1 of its 5 cases by the spliced-fusion mechanism in which a truncated 5'-KMT2A is recombined upstream of an intact partner gene; the historical partner table places MYO1F at t(11;19)(q23.3;p13.2), 19p13.2, attributed to Lo Nigro et al. (2002) with disease annotation AML.
The KMT2A recombinome of acute leukemias in 2023. · Meyer C, Larghero P, Almeida Lopes B, Burmeister T, Gröger D, Sutton R, et al. · Leukemia 2023What this evidence cannot show
Read directly from the Europe PMC full-text XML of PMC10169636 with the Table 1 column structure verified from the header. Five patients is a count at which nothing about distribution is stable. The prose spelling 'MYOF1' is a typographic variant of MYO1F, not a different gene, and is recorded because it defeats symbol-keyed text mining; MYOF (myoferlin) is a real and unrelated gene.
Evidence that argues against this (1)
- This partner is specifically prone to being missed or misassigned, by its discoverers' own account: MYO1F is the fourth KMT2A partner at 19p13 and MLL-ENL, MLL-ELL, MLL-EEN and MLL-MYO1F are difficult to distinguish cytogenetically at t(11;19)(q23;p13), so a count derived substantially from karyotype-directed referral will under-represent it. Taki T et al. 2005
- E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset
MYO1F is not a dependency in any leukaemia context measured and is among the least KMT2A-selective genes in the genome: DepMap Chronos mean gene effect -0.1301 across 1208 lines, KMT2A-rearranged mean -0.0576 versus -0.1618 in other leukaemia lines (difference +0.1042, i.e. LESS required in KMT2A-rearranged lines), rank 16,967 of 17,728 (95.7th percentile) for KMT2A selectivity; splitting the index arm by lineage gives +1.18 SD in the 8-line myeloid half and -0.00 SD in the 5-line B-lineage half.
DepMap CRISPR (Chronos) gene effect via the Open Targets Platform GraphQL API: genome-wide KMT2A-selectivity scan (analyses/depmap-kmt2a-selective-scan-full/results.json) plus the lineage-split re-analysis of the deferred KMT2A partners run for this beat (analyses/depmap-deferred-partners-lineage-split/) · Broad Institute DepMap; Open Targets Platform; lineage split run by this project · Database and derived analysis 2026What this evidence cannot show
Decisive limitations, two of them: no MLL-MYO1F cell line exists in DepMap, and the founding report found MYO1F expressed in only 3 of 22 leukaemic cell lines, so in most of the panel a CRISPR knockout may be cutting a gene that is not expressed — which produces a null by construction rather than by biology. The lineage split rests on 8 and 5 lines. DepMap release version not stated by the API.
Evidence that argues against this (1)
- A wild-type-partner CRISPR null is weak evidence about a fusion regardless of direction: Sept6-null mice develop normally and MLL-SEPT6 disease is unchanged on that background, while the SEPT6 coiled coil is individually required for transformation within the fusion. Partner dispensability and fusion-module necessity are independent. Ono R et al. 2005
- E6strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
The complete searchable literature on this fusion is two records: a PubMed [Title/Abstract] search for (MLL OR KMT2A) AND MYO1F, run 2026-09-19, returns exactly 2 records — the 2005 founding case report and the 2011 recurrence report — with both titles read; no functional, structural, localisation or epigenetic study of MLL-MYO1F exists in the indexed literature.
NCBI E-utilities esearch over PubMed, query (MLL[Title/Abstract] OR KMT2A[Title/Abstract]) AND MYO1F[Title/Abstract], run 2026-09-19, 2 records, both titles read; the gene-only query MYO1F[Title/Abstract] returns 70 records · US National Library of Medicine, PubMed · PubMed (E-utilities) 2026What this evidence cannot show
A counted-absence claim. MYO1F has no retired HGNC symbol, so the under-counting failure mode that dominates the septins does not apply; but the recombinome's prose prints the symbol as 'MYOF1', which shows that typographic variants can hide mentions from a symbol-keyed search, and no search for that variant or for 'myosin 1F' spelled out was run. This count also supersedes the inherited enumeration's figure of 1 record.
Evidence that argues against this (1)
- Title/abstract counting and patient compendia disagree systematically for this partner class: MYO1F returns 2 records but 5 patients, while SEPTIN3 returns 0 records and still appears as a patient in the same compendium, and GPHN returns 3 records while appearing in no frequency count at all. Neither method alone bounds what is known. Meyer C et al. 2023
almost the entire MYO1F: motor domain, IQ motif, TH1 tail and SH3
- ← fuses withKMT2A / MLL1 — the gene that breaks
t(11;19)(q23;p13.2); 5 patients, all AML, 4 infant; recurrent in infant AML
deferred by that enumeration at '1 PubMed record, nothing to read'; the correct count is 2 records, one establishing recurrence in infant AML