Oncofusion

start / graph / gene-hcfc1

Proteinstrong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.Q5d-interactome3

HCFC1 (HCF-1) — a founding member of the wild-type KMT2A complex that the fusion cannot bind

also called HCFC1 · HCF-1 · Host cell factor 1 · HCF1

HCF-1 is one of the proteins that sticks to the normal, unbroken KMT2A protein. The fusion version of KMT2A loses the exact four-amino-acid landing pad HCF-1 needs, so HCF-1 belongs to the healthy machine and not to the cancer machine.

The same thing, for a scientist

HCFC1 (P51610, 2,035 aa) was identified alongside menin in the 2004 biochemical purification that first defined the 1-MDa MLL/KMT2A SET1-like complex. The interaction is motif-based and mapped to KMT2A residues 1799-1802, a (D/E)HxY HCF-binding motif of the class HSV VP16 uses. That motif lies C-terminal to the KMT2A breakpoint cluster region (exons 8-13, roughly aa 1100-1400), so it is deleted from every leukemia-associated KMT2A fusion including KMT2A::AFF1. The same paper showed directly that menin coprecipitates with fusion KMT2A while HCF-1, HCF-2, ASH2L, WDR5 and RBBP5 do not. HCFC1 is DepMap common-essential (mean gene effect -1.67 across 1,208 lines, 100% of lines depleted) with no KMT2A-selectivity (difference -0.073 between 11 KMT2A-rearranged and 43 other leukemia lines, against -0.866 for MEN1 computed identically).

What it looks like
loading AF-P51610-F1…
AlphaFold AF-P51610-F1 ↗
colourshape

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.

very high (90+)confident (70–90)low (50–70)very low (<50)
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.

Go deeper
Why this entry exists at all, given that it is a negative

The founding paper of the menin-KMT2A field 1 is usually cited for menin. What it actually reports is a discrimination: of the components it purified, only menin bound both wild-type and fusion KMT2A. HCF-1 and HCF-2 are the proteins on the other side of that line, and naming them makes the discrimination legible instead of leaving it as an unattributed claim about "other components".

The structural reason is clean. Menin binding maps to the first 1,406 residues of KMT2A-N, which is described as consistently retained in all leukemia-associated MLL fusion proteins 1. HCF binding requires KMT2A aa 1799-1802 1, which is past every clinical breakpoint. One contact is inside the retained fragment and one is outside it, and that single fact is the strongest single-paper argument in this graph for why menin is the druggable handle and most of the wild-type interactome is not.

How the interaction was shown, and for what

Method: immunoaffinity purification of KMT2A with mass-spectrometric identification of co-purifying proteins, then domain mapping and point-mutation analysis of the HCF-binding motif, then co-immunoprecipitation/Western 1. This is biochemistry on the endogenous complex plus mapping on transfected fragments — stronger than a single co-IP, weaker than a structure.

Shown for: wild-type KMT2A. The fusion experiments in the same paper are the negative: endogenous menin coprecipitated with a KMT2A::EP300 fusion in transfected 293 cells and with endogenous fusion KMT2A in the HB (KMT2A::MLLT1) leukemia line, while HCF-1 and HCF-2 did not 1.

Which part of the fusion: none. The contact is on KMT2A aa 1799-1802, on the discarded C-terminal fragment, not on the retained N-terminus and not on AFF1. HCFC1 is not a member of the KMT2A::AFF1 interactome.

The dependency data agree, for an independent reason

HCFC1 is one of the most broadly required genes in DepMap: mean gene effect -1.67 across 1,208 lines with 100% of lines depleted 2. Pan-essentiality is the classic way to be fooled into calling something a leukemia target. Its KMT2A-selectivity is -0.073 (mean -1.864 in 11 KMT2A-rearranged lines against -1.791 in 43 other leukemia lines), i.e. nothing, set against MEN1's -0.866 computed in exactly the same lines by exactly the same method 2.

So two independent readouts converge: no fusion contact, no fusion-selective requirement. The paralogue HCFC2 (Q9Y5Z7, 792 aa) sits at the opposite extreme — mean gene effect +0.126 with 0% of lines depleted, a gene whose loss no cell line notices 2.

Facts and figures
uniprotP51610
gene symbolHCFC1
length AA2035
protein nameHost cell factor 1
alphafold IDAF-P51610-F1
accession resolutionUniProt query gene_exact:HCFC1 AND organism_id:9606 AND reviewed:true; primary gene name read back as HCFC1 and length 2035 confirmed, not accepted on a synonym match
complexMLL/KMT2A SET1-like histone methyltransferase complex (wild-type only)
interaction evidence classaffinity purification + mass spectrometry, domain mapping, point mutation, co-IP/Western
shown for fusionfalse
druggabilityNot a target for this disease. Pan-essential, no fusion contact, no fusion-selective requirement.
subnuclear localizationNOT MEASURED. No subnuclear compartment has been published for HCFC1 in any system. What exists is CHROMATIN ASSOCIATION BY FUNCTION - HCFC1 is a chromatin-associated transcriptional regulator recruited to promoters by sequence-specific partners [E_q71_ct1] - which is locus occupancy rather than a compartment. [EXPLICIT NEGATIVE] SAY IT PLAINLY: UNPUBLISHED, not not-applicable. HCFC1 is a 2,035-residue nuclear protein and a subnuclear pool is entirely plausible; nobody has imaged one. And nothing has been measured in a KMT2A-rearranged cell - which is pointed, because this entry's whole argument is that HCFC1 is a founding member of the WILD-TYPE KMT2A complex that the fusion CANNOT bind. A compartment comparison between wild-type and fusion cells is the measurement this entry most needs and it does not exist.
epigenetic readoutH3K4 METHYLATION, BY TETHERING THE WRITER - AND IT TETHERS AN ERASER IN THE SAME BREATH, which is the part that makes this the interesting epigenetics entry of the tranche. HCFC1/HCF-1 (P51610, 2,035 aa) has no catalytic chromatin domain of its own; its readout is which enzymes it brings to a promoter. (1) IT TETHERS OPPOSITES. Separate regions of HCF-1 critical for cell proliferation associate with the Sin3 histone DEACETYLASE complex and with a trithorax-related Set1/Ash2 histone METHYLTRANSFERASE that methylates H3 at lysine 4 - and HCF-1 tethers the two together despite their opposite transcriptional outcomes; the tethering is CONTEXT-DEPENDENT, because the activator VP16 selectively binds the HCF-1 pool carrying the Set1/Ash2 HMT in the ABSENCE of the Sin3 HDAC [E_q97_epi1]. There is also a substrate-order constraint in the same paper: the Set1/Ash2 HMT methylates H3K4 BUT NOT if the neighbouring K9 residue is already methylated. (2) AND IT BRINGS KMT2A ITSELF, CELL-CYCLE-SELECTIVELY. During the G1-to-S transition HCF-1 RECRUITS THE MIXED-LINEAGE LEUKAEMIA (MLL) AND SET1 H3K4 METHYLTRANSFERASES to E2F-responsive promoters and induces histone methylation and transcriptional activation, acting as a coactivator when bound to E2F1 and a corepressor when bound to E2F4 [E_q97_epi2]. So the founding member of the wild-type KMT2A complex is also the factor that puts KMT2A on S-phase promoters, and it does so through a motif this entry already establishes is DELETED FROM EVERY FUSION: the (D/E)HxY HCF-binding motif at KMT2A aa 1799-1802 lies C-terminal to the breakpoint cluster region. THE CONSEQUENCE THAT FOLLOWS, AND IT IS A PREDICTION THIS GRAPH CAN TEST: if HCF-1 is how wild-type MLL1 reaches E2F-responsive S-phase promoters, then a KMT2A fusion is CONSTITUTIVELY BLIND to that address, and the H3K4me3 at those promoters in a fusion-positive blast must come from the retained wild-type allele or from SETD1A/SETD1B. Nobody has measured H3K4me3 at E2F-target promoters in a KMT2A-rearranged cell, and this graph's gene-crebbp entry independently reports the CDK4/RB/E2F axis being downregulated MLL-r-specifically under a p300/CBP HAT inhibitor (PMID 38693103) - two routes to the same axis from two different chromatin enzymes. IDENTIFIER TRAP, MEASURED: the approved symbol is the LOW-RECALL operand for this biology. HCFC1[tiab] AND H3K4[tiab] returns 3; "HCF-1"[tiab] AND H3K4[tiab] returns 10, and BOTH papers cited here are reachable only from the protein-name form. Same direction as the SEPT/SEPTIN row in data/symbol-collisions.json: using the correct current symbol is what causes the miss.
epigenetic identifier trapMEASURED 2026-09-21, and it runs in the UNUSUAL direction - the approved symbol is the low-recall operand. HCFC1[tiab] AND H3K4[tiab] -> 3 records (querytranslation "HCFC1"[Title/Abstract] AND "H3K4"[Title/Abstract]). "HCF-1"[tiab] AND H3K4[tiab] -> 10 records (querytranslation "HCF-1"[Title/Abstract] AND "H3K4"[Title/Abstract]). BOTH papers this entry's epigenetic_readout rests on - PMID 12670868 and PMID 17612494 - appear only in the 10 and NOT in the 3, so a beat querying the HGNC symbol would have recorded an absence. Same shape as the SEPT/SEPTIN row already in data/symbol-collisions.json: the literature keeps the hyphenated protein name. The unhyphenated 'HCF1' and the older 'VP16-accessory protein' / 'CFF' names were NOT tried and would widen it further.
contact on kmt2a
{
  "kmt2a_residues": [
    1799,
    1802
  ],
  "motif": "(D/E)HxY HCF-binding motif (HBM), the class used by HSV VP16",
  "retained_in_fusion": false,
  "reason": "C-terminal to the KMT2A breakpoint cluster region (exons 8-13, approx. aa 1100-1400), therefore deleted from every leukemia-associated KMT2A fusion including KMT2A::AFF1"
}
essentiality
{
  "depmap_common_essential": true,
  "chronos_mean": -1.6655,
  "n_lines": 1208,
  "fraction_lines_depleted": 1,
  "chronos_mean_kmt2a_rearranged": -1.8641,
  "chronos_mean_other_leukemia": -1.791,
  "kmt2a_selectivity_difference": -0.0731,
  "n_kmt2a_rearranged_lines": 11,
  "n_other_leukemia_lines": 43,
  "benchmark_men1_difference": -0.8656
}
paralogue
{
  "gene_symbol": "HCFC2",
  "uniprot": "Q9Y5Z7",
  "length_aa": 792,
  "chronos_mean": 0.126,
  "fraction_lines_depleted": 0,
  "kmt2a_selectivity_difference": 0.053
}
localization
{
  "subcellular": "NUCLEUS AND CYTOPLASM, both at ECO:0000269 in UniProt P51610 (fetched 2026-09-21). HCF-1 is described in the primary literature as a ubiquitous NUCLEAR protein required for progression through G1, and the cytoplasmic annotation is experimental alongside it.",
  "subnuclear": "NOT MEASURED. No subnuclear compartment has been published for HCFC1 in any system. What exists is CHROMATIN ASSOCIATION BY FUNCTION - HCFC1 is a chromatin-associated transcriptional regulator recruited to promoters by sequence-specific partners [E_q71_ct1] - which is locus occupancy rather than a compartment.",
  "explicit_no": "SAY IT PLAINLY: UNPUBLISHED, not not-applicable. HCFC1 is a 2,035-residue nuclear protein and a subnuclear pool is entirely plausible; nobody has imaged one. And nothing has been measured in a KMT2A-rearranged cell - which is pointed, because this entry's whole argument is that HCFC1 is a founding member of the WILD-TYPE KMT2A complex that the fusion CANNOT bind. A compartment comparison between wild-type and fusion cells is the measurement this entry most needs and it does not exist.",
  "how_shown": "UniProt P51610 SUBCELLULAR LOCATION comment, both entries at ECO:0000269, fetched 2026-09-21; chromatin-association from the regulatory-mutation study (PMID 23000143)",
  "systems": "human and murine neural systems, plus herpesvirus-derived cell biology; no KMT2A-rearranged system",
  "shown_for_fusion_localisation": "no. NOTE THE FIELD NAME: this entry already carries data.shown_for_fusion as a BOOLEAN false, which is the INTERACTOME sense - no demonstrated contact with a KMT2A fusion. That value is untouched. The localisation sense is recorded here under a separate key, per the polysemy Q65 documented and check-pi-dimensions.py encodes.",
  "node": "localization-chromatin-bound-versus-soluble-fraction-kmt2a"
}
cell type
{
  "evidence_class_key": "concept-expressed-required-maintains-three-evidence-classes",
  "identifier_note": "HCFC1 (P51610, 2035 aa, 'Host cell factor 1', synonyms HCF1 and HFC1) is NOT HCFC2. HCF-2 is a separate, smaller family member sharing the beta-propeller domain, expressed in many tissues and especially testis, with a more dynamic subcellular localisation and lower VP16-complex efficiency [counter-evidence on E_q71_ct1]. The literature also calls this gene HCF-1 and HCF, with hyphen variants, so an absence claim keyed to HCFC1 alone will under-count - both papers cited here are reachable only through different spellings.",
  "expressed_in": {
    "class": 1,
    "summary": "ubiquitous; HCF-1 is described as a ubiquitous nuclear protein required for G1 progression, and is depleted in 100% of DepMap lines."
  },
  "required_in": {
    "class": 2,
    "claim": "The only documented cell-type-specific consequence of altering HCFC1 in an organism is NEURAL: in murine embryonic neural stem cells and hippocampal neurons, HCFC1 overexpression increases ASTROCYTE production and reduces neurite growth, and the human phenotype is X-linked intellectual disability [E_q71_ct1]. NO HAEMATOPOIETIC CELL TYPE HAS BEEN MEASURED FOR HCFC1.",
    "pmids": [
      "23000143"
    ],
    "system_named_in_the_claim": "human families with X-linked intellectual disability; murine embryonic neural stem cells and hippocampal neurons - no fusion, no blood",
    "nodes": [
      "cell-type-neuron"
    ]
  },
  "FLOOR_COMPRESSION": {
    "verdict": "THE RANK IS NEAR-UNINTERPRETABLE AND THIS GENE IS THIS GRAPH'S NAMED WORKED EXAMPLE OF WHY. Chronos -1.8641 in the KMT2A-rearranged arm against -1.7910 in other leukaemias, difference -0.0731, rank 2,067 of 17,728, pan-essentiality -1.6655. The standing brief cites HCFC1 (100% of lines depleted, difference -0.073) as one of the two worked examples of floor compression, beside KAT8.",
    "the_honest_nuance": "By the STRICT threshold in the brief - both arms past -2 - HCFC1 does not formally qualify: the arms sit at -1.86 and -1.79. By the underlying criterion - a gene depleted in essentially every line has almost no dynamic range in which selectivity could appear - it plainly does. The two tests disagree for this gene, and the resolution is to report the arm means rather than the rank. A -0.073 difference on top of a -1.67 shared requirement is not a selectivity finding whatever its rank.",
    "consequence_for_this_entry": "Rank 2,067 is in the selective tenth of the scan and means nothing here. This entry's argument does not need it: the point is that HCFC1 binds wild-type KMT2A and not the fusion, which is a binding claim, not a dependency claim."
  },
  "context": "neural systems for the cell type; this project's DepMap scan for the (floor-compressed) rank"
}
cell state
{
  "evidence_class_key": "concept-expressed-required-maintains-three-evidence-classes",
  "maintains_state": {
    "class": 3,
    "states": [
      "cell-state-cell-cycle-g1-cdk6"
    ],
    "claim": "HCF-1 is described as a ubiquitous nuclear protein REQUIRED FOR PROGRESSION THROUGH THE G1 PHASE of the cell cycle [counter-evidence citation on E_q71_ct1, PMID 10196288, which states this as established background]. That is a general cell-biological property of every dividing cell, not a leukaemic or lineage state, and saying so plainly is the result.",
    "pmids": [
      "10196288"
    ],
    "system_named_in_the_claim": "general mammalian cell biology, stated as background in a herpesvirus study",
    "node_caveat": "cell-state-cell-cycle-g1-cdk6 is a CDK6-framed G1 node and is not a perfect match for a general G1 requirement. The edge is drawn weakly and labelled with the mismatch rather than written as an equivalence."
  },
  "confidence": "moderate",
  "caveats": "The G1 requirement is background prose in a virology paper rather than a measurement made there, which is why confidence is moderate. Nothing about HCFC1 has been measured in a haematopoietic cell, let alone a KMT2A-rearranged one. The only organism-level cell-type data are neural and come from OVEREXPRESSION.",
  "counter_evidence": {
    "note": "A general G1 requirement and a 100%-of-lines DepMap depletion are the same fact seen twice, and both point away from HCFC1 being a disease-specific node. The pattern across this tranche is the same one Q69 reported: genes whose requirement is universal or lineage-developmental rather than fusion-created. HCFC1's only organism-level phenotype is neural, in humans with no leukaemia.",
    "citation": {
      "title": "A noncoding, regulatory mutation implicates HCFC1 in nonsyndromic intellectual disability.",
      "authors": "Huang L, Jolly LA, Willis-Owen S, Gardner A, Kumar R, Douglas E, Shoubridge C, Wieczorek D, Tzschach A, Cohen M, Hackett A, Field M, Froyen G, Hu H, Haas SA, Ropers HH, Kalscheuer VM, Corbett MA, Gecz J",
      "year": 2012,
      "venue": "Am J Hum Genet",
      "pmid": "23000143",
      "doi": "10.1016/j.ajhg.2012.08.011",
      "type": "paper"
    }
  }
}
depmap kmt2a selectivity
{
  "chronos_mean_kmt2a_rearranged": -1.8641,
  "chronos_mean_other_leukaemia": -1.791,
  "difference": -0.0731,
  "pan_essentiality_all_lines": -1.6655,
  "selectivity_rank": 2067,
  "rank_denominator": 17728,
  "n_all_lines": 1208,
  "open_targets_dataversion": "26.06",
  "category": "FLOOR-COMPRESSED. Do not quote the rank. Both arms are near -1.8 with pan-essentiality -1.67; this graph's standing brief names HCFC1 as one of its two worked examples of floor compression.",
  "provenance": "Joined 2026-09-21 by Q71 from analyses/depmap-kmt2a-selective-scan-full (generated 2026-09-19T10:12:06Z), over 11 KMT2A-rearranged and 43 other leukaemia DepMap lines, 17,728 genes scanned. Open Targets dataVersion fetched live 2026-09-21 as year 26 month 06, per brief rule 83. NOT independent of that scan: every caveat it carries applies here, including that the 11-line index arm is myeloid-enriched and contains no infant-derived line. This entry carried NO depmap block before - the Q71 dispatch brief asserted every gene in its set carried one, and three did not.",
  "how_to_read": "A negative difference means MORE required in KMT2A-rearranged lines. A rank near 17,728 is ANTI-selective. Check pan_essentiality_all_lines and BOTH arm means first: if both arms sit past -2 the rank is floor-compressed and uninterpretable, and if both arms are POSITIVE the gene is not a dependency at all and the scan's rank does not apply to it (the CDKN2A category error Q69 named in cycle 39, and the ATM and SIRT1 cases Q71 added in cycle 40)."
}
subnuclear localization CORRECTED 2026 09 22
{
  "was": "NOT MEASURED. No subnuclear compartment has been published for HCFC1 in any system. What exists is CHROMATIN ASSOCIATION BY FUNCTION - HCFC1 is a chromatin-associated transcriptional regulator recruited to promoters by sequence-specific partners [E_q71_ct1] - which is locus occupancy rather than a c",
  "correction": "'No subnuclear compartment has been published for HCFC1 in any system' is wrong. A subnuclear statement about HCF-1 IS published, and it is NEGATIVE. PMID 31049581 (Nucleic Acids Research, doi 10.1093/nar/gkz307), abstract verbatim, fetched and read by the king 2026-09-22: 'Unlike HCF-1, which is excluded from nucleoli, HCF-2 is nucleolar.'",
  "the_right_framing": "The only published subnuclear fact about HCF-1 is where it is NOT: excluded from nucleoli. That is a measurement, not an absence, and it is the kind this project's absence taxonomy exists to distinguish -- a negative finding is not 'not measured'.",
  "caveat": "Abstract read; full text not read. The statement is made in passing as the contrast that motivates the paper's HCF-2 finding, so the underlying HCF-1 experiment may be cited from earlier work rather than performed there.",
  "found_by": "Q93 while writing gene-hcfc2; verified and written by the king"
}
How this could bite us back
  • HCFC1 is depleted in 100% of DepMap lines. Any agent touching it would carry essential-function toxicity in every tissue, and would do so without engaging the fusion at all.
  • If a future proteomic study does find HCF-1 in a KMT2A::AFF1 pulldown, the correct reading is a second binding surface, not a reason to reopen HCF-1 as a target — the pan-essentiality argument is independent of the binding argument.
  • HCFC1 is depleted in essentially every DepMap line, with a pan-essentiality of -1.6655 - deeper than any other gene in this tranche by a factor of two. There is no therapeutic window for removing it, and the rank of 2,067 is floor compression rather than selectivity.
  • Altering HCFC1 DOSE in humans causes X-linked intellectual disability (PMID 23000143). Whatever is done to this protein in an infant has a documented neurodevelopmental consequence, in a patient population already exposed to CNS-directed therapy.
  • HCFC1 is essential for cell proliferation in every system in which it has been tested, and both of its chromatin arms - a deacetylase and a methyltransferase - are delivered by the same protein. Any agent that disrupted the HCF-binding interface would remove a proliferation factor from every dividing cell while moving two chromatin marks in opposite directions, which is the worst combination of poor selectivity and unpredictable pharmacodynamics.
  • The HCF-binding motif at KMT2A aa 1799-1802 is deleted from every fusion, so this interface cannot be a fusion-selective target by construction: hitting it acts only on the wild-type allele and on the other MLL-family enzymes, i.e. on the normal cell.
Still open
  1. Has anyone tested HCF-1 against KMT2A::AFF1 directly with modern proteomics? The exclusion here rests on a 2004 purification and on the position of aa 1799-1802, and no direct test exists.
  2. The wild-type KMT2A allele is retained in KMT2A-rearranged leukemia and still carries the HBM. Does HCF-1 occupancy at shared target loci on the wild-type allele modify what the fusion does there?
  3. Does loss of the HBM contribute anything to transformation, or is it merely collateral? A fusion engineered to carry aa 1799-1802 in cis would answer it.
  4. Is HCFC1's chromatin-bound fraction different in a KMT2A-rearranged cell than in a wild-type one? This entry's central claim is that the fusion cannot bind HCFC1; if true, the fusion cell's HCFC1 should be redistributed, and a soluble-versus-chromatin fractionation would test it in one experiment.
  5. Does HCFC2 compensate for HCFC1 anywhere in blood? HCF-2 shares the beta-propeller, is broadly expressed, and localises differently (PMID 10196288) - and this graph has no HCFC2 entry, so the redundancy question has never been posed.
  6. Is H3K4me3 at E2F-responsive S-phase promoters lower in a KMT2A-rearranged blast than in a fusion-negative one? HCF-1 recruits the MLL-family H3K4 methyltransferases there (PMID 17612494) through a motif at KMT2A aa 1799-1802 that no fusion retains, so the fusion should be blind to that address - and the measurement has never been made.
  7. Which enzyme does HCF-1 actually deliver to E2F promoters in a haematopoietic cell - KMT2A, KMT2B, SETD1A or SETD1B? The 2007 abstract says the MLL family, and the answer decides whether the fusion-blindness prediction above is about KMT2A at all.
  8. Does the Sin3-HDAC pool of HCF-1 or the Set1/ASH2-HMT pool dominate at KMT2A's own target genes? HCF-1 tethers both (PMID 12670868) and this graph has never asked which one the wild-type KMT2A complex sits next to.
Every source behind this entry (5)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    HCF-1 and HCF-2 are components of the 1-MDa MLL/KMT2A SET1-like complex and bind a conserved motif requiring KMT2A amino acids 1799-1802; menin binding maps within the first 1,406 residues of KMT2A-N, which is retained in all leukemia-associated MLL fusions; and menin, but not HCF-1, HCF-2, ASH2L, WDR5 or RBBP5, coprecipitates with fusion MLL proteins.

    Leukemia proto-oncoprotein MLL forms a SET1-like histone methyltransferase complex with menin to regulate Hox gene expression. · Yokoyama A, Wang Z, Wysocka J, Sanyal M, Aufiero DJ, Kitabayashi I, Herr W, Cleary ML · Molecular and Cellular Biology 2004
    PMID 15199122DOI
    What this evidence cannot show

    Full text read via PMC (PMC480881) for the residue numbers and the fusion-binding experiments; the abstract was read as well. The fusion constructs are KMT2A::EP300 transfected into 293 cells and endogenous fusion KMT2A in the HB line — neither is KMT2A::AFF1, so for KMT2A::AFF1 the negative is inferred from breakpoint position rather than demonstrated. The 2004 mass-spectrometry depth is shallow by 2026 standards, so absence of a protein from this purification is weak evidence of absence.

    Evidence that argues against this (1)
    • A strict reading limits this paper's negative to the components it purified and the two fusions it tested. It does not exclude a later-discovered fusion contact with HCF-1 through a second, motif-independent surface, and no study has tested HCF-1 against KMT2A::AFF1 directly. The inference that HCF-1 is excluded from KMT2A::AFF1 rests on the position of aa 1799-1802 relative to the breakpoint cluster region, which is a strong structural argument and not an experiment. Oncofusion Knowledge Graph project et al. 2026
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.dataset

    HCFC1 has DepMap mean gene effect -1.6655 across 1,208 lines with 100% of lines depleted and a KMT2A-rearranged minus other-leukemia difference of -0.073 (11 vs 43 lines); HCFC2 has mean +0.126 with 0% depleted; MEN1 computed in the same lines has a difference of -0.866.

    What this evidence cannot show

    DepMap release version is not stated by the API. Eleven KMT2A-rearranged lines is a small group and the difference statistic is noisy at that n; it is used here only to say HCFC1 shows nothing, which is the easy direction to call. Proliferation screens cannot see phenotypes that are not proliferation.

    Evidence that argues against this (1)
    • Pan-essentiality is the finding here and it argues against the target rather than for it. A gene depleted in more than 90% of all cell lines is required by the patient's normal marrow as much as by the leukaemia, so its appearance in a KMT2A complex is a statement about the complex and not a therapeutic opening. 100% of lines depleted with a KMT2A-selectivity difference of −0.073 is the floor-compression problem: a gene that every line needs has almost no dynamic range left in which selectivity could be detected, so the small difference is uninterpretable rather than small. Oncofusion knowledge graph (internal analysis) 2026
  3. E_q71_ct1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    In HUMAN FAMILIES WITH X-LINKED INTELLECTUAL DISABILITY and in murine neural cells, a regulatory mutation in a YY1 binding site at Xq28 causes OVEREXPRESSION of the chromatin-associated transcriptional regulator HCFC1, and when tested on embryonic murine NEURAL STEM CELLS and embryonic HIPPOCAMPAL NEURONS, HCFC1 overexpression significantly increased the production of ASTROCYTES and considerably reduced neurite growth; two further nonsynonymous potentially deleterious changes were found by X-exome sequencing in individuals with intellectual disability.

    A noncoding, regulatory mutation implicates HCFC1 in nonsyndromic intellectual disability. · Huang L, Jolly LA, Willis-Owen S, Gardner A, Kumar R, Douglas E, Shoubridge C, Wieczorek D, Tzschach A, Cohen M, Hackett A, Field M, Froyen G, Hu H, Haas SA, Ropers HH, Kalscheuer VM, Corbett MA, Gecz J · Am J Hum Genet 2012
    PMID 23000143DOI
    What this evidence cannot show

    Abstract read in full; full text not read. HUMAN families plus MURINE embryonic neural stem cells and hippocampal neurons - NOT haematopoietic, not leukaemia, not KMT2A-rearranged (brief rule 73). This is cited precisely because it is the best-documented CELL-TYPE consequence of altering HCFC1 dose in a whole organism, and it is neural. The direction is OVEREXPRESSION, not loss, so it does not predict what a degrader would do. The human genetics and the cell-biology are in different systems and the paper does not claim the astrocyte phenotype explains the intellectual disability.

    Evidence that argues against this (1)
    • Whether HCFC1's compartment behaviour can be generalised is doubtful even within its own small family. HCF-2, the second HCF-like protein, shares the beta-propeller domain required for VP16 association and is expressed in many tissues, but shows a MORE DYNAMIC PATTERN OF SUBCELLULAR LOCALIZATION THAN HCF-1 and is significantly less efficient at supporting VP16-Oct-1 complex assembly. So HCF-1 and HCF-2 differ in both compartment behaviour and partner affinity, and a claim written from one cannot be transferred to the other - which matters because HCFC2 is a separate gene this graph does not carry. Johnson KM et al. 1999
  4. E_q97_epi1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    Separate regions of HCF-1 critical for cell proliferation associate with the Sin3 histone deacetylase complex and with a trithorax-related Set1/Ash2 histone methyltransferase that methylates histone H3 at lysine 4 but not if the neighbouring K9 is already methylated; HCF-1 tethers the Sin3 and Set1/Ash2 complexes together despite their opposite transcriptional outcomes, and the tethering is context-dependent because VP16 selectively binds HCF-1 associated with the Set1/Ash2 HMT in the absence of the Sin3 HDAC.

    PMID 12670868
    What this evidence cannot show

    Abstract fetched and read in full 2026-09-21; full text not read. HeLa-scale human biochemistry with an HSV VP16 readout - not a haematopoietic system and not a leukaemia (rule 73). 'Set1/Ash2' in 2003 nomenclature is the SET1/COMPASS family, i.e. SETD1A/SETD1B plus ASH2L, and is NOT KMT2A; the KMT2A recruitment comes from a separate paper [E_q97_epi2]. The H3K4 methylation state is not specified in the abstract, so me1/me2/me3 is not claimed here.

    Evidence that argues against this (1)
    • The tethering result cuts against using HCF-1 as a proxy for H3K4 methylation at all: the same protein delivers a deacetylase and a methyltransferase, so its net chromatin output at a given promoter depends on which pool is recruited and cannot be predicted from HCF-1 occupancy. An entry that wrote 'HCFC1 -> H3K4me3' as a simple arrow would be discarding half of what the founding paper reports. Mol Cell 2007
  5. E_q97_epi2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    HCF-1 associates with both activator (E2F1, E2F3a) and repressor (E2F4) E2F proteins in a cell-cycle-selective way, acting as a coactivator with E2F1 and a corepressor with E2F4; during the G1-to-S transition HCF-1 recruits the mixed-lineage leukaemia (MLL) and Set1 histone H3 lysine 4 methyltransferases to E2F-responsive promoters and induces histone methylation and transcriptional activation.

    PMID 17612494
    What this evidence cannot show

    Abstract fetched and read in full 2026-09-21; full text not read. Human cell lines plus insect-cell conservation work; NOT leukaemia and NOT a KMT2A-fusion system (rule 73). 'MLL' here means the MLL FAMILY of H3K4 methyltransferases in the title and 'mixed-lineage leukemia (MLL)' in the abstract, which in 2007 usage is ambiguous between KMT2A specifically and the KMT2A/KMT2B/SETD1 family - the same ambiguity data/symbol-collisions.json records for MLL2/MLL4. This entry therefore states the recruitment at family level and does not assert that KMT2A alone is the recruited enzyme. The methylation readout is 'histone methylation' induced at those promoters; no residue-and-state pair is quoted.

    Evidence that argues against this (1)
    • If the recruited enzyme is SETD1A/SETD1B rather than KMT2A, the prediction this entry draws - that a KMT2A fusion is blind to E2F-responsive S-phase promoters because the HCF-binding motif is deleted - loses its force, because the H3K4 methylation at those promoters was never KMT2A's to begin with. Resolving that requires an isoform- or paralogue-resolved ChIP that the 2007 abstract does not contain, and this graph holds gene-setd1a and gene-setd1b precisely because the COMPASS family is not interchangeable. Genes Dev 2003
Connected to