Oncofusion

start / graph / gene-rel

Proteinweak confidence 2 of 5: Preliminary, small, indirect, or not yet peer reviewed.Q38-nfkb-signalling

REL (c-Rel) — the third canonical subunit, measured once and found unchanged, and named in zero papers in this disease

also called REL · c-Rel · Proto-oncogene c-Rel · Q04864

c-Rel is the third member of the classical NF-kB family and the one nobody in this field has looked at. The single time it was measured in an MLL leukaemia cell its level did not change, and no paper has ever been written about it in this disease.

The same thing, for a scientist

REL (UniProt Q04864, 619 aa, primary gene name read back) encodes c-Rel, the third transactivating canonical subunit alongside RELA and RELB. It has its own entry because **it was measured and the measurement was null, which is different from never having been looked at** and carries more information. In RNA-seq of NIK-stabilised MLL-AF9 leukaemia stem cells, the non-canonical genes NfkB2 and Relb were significantly upregulated **without significant alteration of the canonical components Nfkb1, Rela or Rel** [E1]. So `Rel` transcript is unmoved by the manipulation that reverses the leukaemia. **That is the only measurement of c-Rel anywhere in the KMT2A literature.** A PubMed search over c-Rel spellings against KMT2A, MLL and 'mixed-lineage leukemia' returns **0 records** [E2]. **Why the null still matters.** c-Rel is the subunit most associated with lymphoid proliferative programmes and with B-cell biology specifically, and the index disease is infant B-ALL. The founding menin-NF-kB interaction paper tested p50, p52 and p65 and **did not test c-Rel** [E3], so the graph's whole menin-NF-kB axis is silent about the subunit most likely to matter in B lineage. That is a gap with a name, not a settled negative.

What it looks like
loading AF-Q04864-F1…
AlphaFold AF-Q04864-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
What 'unchanged' does and does not license

The Xiu measurement is transcript level in a mouse myeloid LSC under one genetic manipulation 1. It does not report c-Rel protein, nuclear localisation, dimer composition or DNA occupancy, and it does not test whether c-Rel is required. A gene can be transcriptionally invariant and functionally essential; this graph has already had to correct itself where an unchanged level was read as an unchanged role. The defensible statement is 'Rel transcript does not respond to NIK stabilisation in mouse MLL-AF9 LSCs' and nothing wider.

The B-lineage reason to care

c-Rel:p50 is the dimer most associated with lymphocyte survival and proliferation programmes. Every functional NF-kB experiment in KMT2A-rearranged leukaemia is myeloid — MLL-AF6, AF9, AF10, AF1p, ENL, and the human AML lines — and the founding menin interaction paper recovered p50, p52 and p65 but did not assay c-Rel 3. If canonical NF-kB retains the fusion on chromatin in a B cell, the subunit doing it need not be RELA.

That makes c-Rel the cheapest unexplored variable in this beat: one immunoblot panel of RELA, c-Rel, p50, p52 and RelB across a set of KMT2A::AFF1 B-ALL lines and patient blasts would tell the field which subunit it should have been studying, and no such panel exists.

Facts and figures
uniprotQ04864
length AA619
alphafold IDAF-Q04864-F1
resolution notegene_exact:REL AND organism_id:9606 AND reviewed:true returned 1 reviewed row; gene_primary read back as REL; length 619 aa.
pathway armcanonical
disease evidenceone null transcript measurement in mouse MLL-AF9 LSCs; zero title/abstract records
depmap kmt2a selectivity
{
  "chronos_mean_kmt2a_rearranged": -0.055,
  "chronos_mean_other_leukaemia": -0.0191,
  "difference": -0.0359,
  "pan_essentiality_all_lines": -0.0336,
  "selectivity_rank": 4728,
  "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."
}
How this could bite us back
  • c-Rel is required for lymphocyte activation and for regulatory T-cell development. Any selective targeting of it in a child would be trading an infection and autoimmunity risk for the granulopoietic risk of IKKb inhibition, not avoiding a risk.
  • Writing this entry on a null measurement risks the graph treating a hole as a finding. The evidence base here is one unchanged transcript in one mouse myeloid model, and the entry should be read as an instruction to measure rather than as knowledge.
Still open
  1. Which NF-kB dimer occupies MLL target promoters in a B-lymphoid KMT2A-rearranged cell? Every occupancy experiment in this literature is RELA in myeloid cells, and c-Rel is the subunit most associated with B-cell programmes.
  2. Does menin bind c-Rel? The founding interaction paper tested p50, p52 and p65 and did not report c-Rel, so the graph's menin-NF-kB axis has a hole exactly where the index lineage lives.
  3. Is c-Rel protein or nuclear localisation altered in KMT2A-rearranged blasts? Transcript invariance in one mouse model says nothing about this, as the same paper's own RelA data demonstrate.
  4. Would a c-Rel-selective perturbation spare the innate-immune functions that make pan-NF-kB inhibition intolerable? c-Rel has a narrower expression pattern than RELA, which is the only reason to think the answer might be yes.
Every source behind this entry (4)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    RNA sequencing of sorted leukaemia stem cells showed that stabilisation of NIK significantly upregulated the NF-kB non-canonical genes NfkB2 and Relb without significantly altering the canonical pathway components Nfkb1, Rela and Rel, indicating that NIK mainly activated non-canonical signalling in LSCs.

    Stabilization of NF-κB-Inducing Kinase Suppresses MLL-AF9-Induced Acute Myeloid Leukemia. · Xiu Y, Dong Q, Li Q, Li F, Borcherding N, Zhang W, Boyce B, Xue HH, Zhao C · Cell Reports 2018
    PMID 29320732DOI
    What this evidence cannot show

    Transcript level only, in mouse MLL-AF9 LSCs, under one genetic manipulation. No c-Rel protein, localisation, dimer composition, occupancy or loss-of-function data. Note that RelA protein and nuclear RelA fell markedly in the same cells while Rela transcript did not change significantly - direct evidence within this paper that transcript invariance says little about protein behaviour, which applies with equal force to Rel. Full text read this session (PMC5810947).

    Evidence that argues against this (1)
    • Transcript invariance is weak evidence about a subunit's role, and this paper supplies the counter-example itself: RelA protein and nuclear RelA fell markedly in the same cells in which Rela transcript did not change significantly. The identical reasoning applied to Rel would be unsafe, so the null measurement here constrains almost nothing about c-Rel protein or occupancy. Xiu Y et al. 2018
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

    Counted absence. (KMT2A OR MLL OR "mixed-lineage leukemia") AND ("c-Rel"[tiab] OR "cRel"[tiab] OR "REL proto-oncogene"[tiab]) returns 0 PubMed records. c-Rel has never been named in the title or abstract of a KMT2A or MLL paper.

    What this evidence cannot show

    querytranslation was read and contained no [UID] artefact. The bare symbol REL was deliberately excluded: it is a three-letter string that matches 'rel', 'relative' and countless truncations, and including it would inflate the count with noise - the alias-precision failure the brief records for RD/NELFE and U19/EAF2. So this bounds the absence over c-Rel spellings, not over every possible way to write the gene.

    Evidence that argues against this (1)
    • A zero at the symbol level is not a zero at the pathway level, and on this beat that is demonstrable rather than hypothetical: PMID 29320732 studies the non-canonical arm directly in MLL-AF9 leukaemia and is unreachable from any of these subunit symbol queries, because it names the kinase and the pathway rather than the subunit. The count is honest about what was searched and should not be read as evidence that the biology is unstudied - only that it is not indexed under this symbol. Xiu Y et al. 2018
  3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    The founding menin-NF-kB interaction study recovered p50, p52 and p65 as specific, direct menin interactors and mapped menin residues 305-381 as essential; c-Rel was not among the proteins tested.

    PMID 11526476DOI
    What this evidence cannot show

    Abstract fetched and read this session; the absence of c-Rel is inferred from the abstract's enumeration of the proteins found, which is an absence-of-mention rather than a reported negative result. The paper may have tested it without reporting. Cited here to bound what the menin-NF-kB axis in this graph covers, not to assert that menin does not bind c-Rel.

    Evidence that argues against this (1)
    • That c-Rel was not among the proteins tested is precisely why this claim cannot be used either way. A 2001 panel that omitted c-Rel supports no inference about c-Rel, and this entry should not read the omission as a negative - which is the error the graph's counted-absence rules exist to prevent. Xiu Y et al. 2018
  4. E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    UniProt reviewed human record Q04864 (REL_HUMAN) has primary gene name REL and length 619 aa; gene_exact:REL AND organism_id:9606 AND reviewed:true returned exactly one row and gene_primary read back as REL.

    UniProtKB entry Q04864 (REL_HUMAN), Proto-oncogene c-Rel · UniProt Consortium · UniProt 2026
    What this evidence cannot show

    Resolved 2026-09-20 with gene_primary requested and all rows read. Unusually for a three-letter symbol this returned exactly one reviewed row with a primary match, so no disambiguation was needed - unlike TRAF2, whose identical query returns TANK first.

    Evidence that argues against this (1)
    • A reviewed database record fixes the canonical sequence and primary symbol on the date it was read, and nothing more. It does not establish that the protein is expressed in KMT2A-rearranged blasts, which isoform is present, or that the subunit acts in this disease. Accession and length are asserted so that a swap fails verify-accessions.py - a defence this graph needs because IKK-family and NF-kB subunit names collide heavily in the older literature - not because the record carries biological weight. Xiu Y et al. 2018
Connected to