Oncofusion

start / graph / gene-ythdc1

Genemoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q6-interactome-rna-layer

YTHDC1 — the nuclear m6A reader in the MLL-METTL3-CTCF axis, with an inhibitor, and no dependency signal in KMT2A-rearranged lines

also called YT521-B · YT521 · YTH domain containing 1

YTHDC1 is the protein that reads the chemical tag on RNA inside the nucleus, and in MLL-rearranged myeloid leukaemia it works with METTL3 and CTCF to hold the genome's 3D folding together. There is now a drug-like inhibitor of it, but the dependency data show no extra need for it in KMT2A-rearranged cells.

The same thing, for a scientist

YTHDC1 is the principal nuclear m6A reader. In MLL-rearranged AML it forms a complex with METTL3 and CTCF at promoters and enhancers, and its loss compromises CTCF-dependent 3D genome organisation; it recognises m6A-modified architectural RNAs such as MALAT1 to enhance R-loop formation and sustain TAD activity at leukaemic loci [E1]. It also nucleates an m6A-dependent phase-separated nuclear body that suppresses transcription [E2], and a small-molecule YTHDC1 inhibitor has been reported as an AML strategy [E3]. **The dependency data are flat to positive**: Chronos mean -0.481 with 42.5% of 1,208 lines depleted, and index-arm differences of **+0.134** B-lymphoid and +0.003 myeloid, the 83rd percentile of a 466-gene null [E4]. `(MLL OR KMT2A) AND YTHDC1` returns 1 record, the same paper as [E1]; the alias YT521-B returns the same one [E5].

Go deeper
The one paper and the alias that did not help

YTHDC1's entire connection to this disease is a single 2025 multi-omics paper 1. Searched with the current symbol it returns 1 record; searched with the heavily used older alias YT521-B it returns the same 1 record 5.

That is worth stating because this project's transferable lesson from cycle 40 was the opposite: searching retired symbols recovered nine-tenths of the septin literature. The rule is that alias expansion must be run, not that it always pays. Here it returns nothing new, and the honest conclusion is that the literature is genuinely one paper rather than a symbol artefact.

A reader with an inhibitor and no selectivity signal

YTHDC1 is the most awkward entry in the beat. It has a mechanism in MLL-rearranged AML 1, a striking cell-biological role in an m6A-dependent nuclear body 2, and a published small-molecule inhibitor in AML models 3 - and a dependency row that goes the wrong way in the index arm in both lineage halves 4.

Three readings.

The window is in the modification, not the reader. If YTHDC1's requirement is proportionate to bulk nuclear m6A, then KMT2A-rearranged cells have no special need for it even if m6A matters to them.

42.5% of lines depleted means YTHDC1 is substantially required across cancer generally. A broad requirement with no index-arm enrichment is the profile of a general dependency, and inhibiting it would be expected to have a general effect.

Four lymphoid lines. +0.134 is within the null and should not be read as protection.

The entry does not choose, but it does refuse to list YTHDC1 as a KMT2A-selective target, because nothing in the data supports that.

Facts and figures
uniprotQ96MU7
length AA727
gene symbolYTHDC1
uniprot resolutiongene_exact:YTHDC1 AND organism_id:9606 AND reviewed:true; one hit; primary gene name read back as YTHDC1; length 727 aa checked
rna functionprincipal nuclear m6A reader; YTH-domain recognition of m6A on mRNA and architectural RNAs
chemical mattersmall-molecule YTHDC1 inhibitor reported in AML mouse models (PMID 41637525)
pubmed counts
(MLL OR KMT2A) AND YTHDC11.000(MLL OR KMT2A) AND YTHDC1: 1(MLL OR KMT2A) AND YT521-B1.000(MLL OR KMT2A) AND YT521-B: 1YTHDC1 AND leukemia22.0YTHDC1 AND leukemia: 22
essentiality
{
  "depmap_common_essential": false,
  "chronos_mean": -0.4806,
  "pct_lines_depleted": 42.5,
  "n_lines": 1208,
  "kmt2a_r_lymphoid_diff": 0.1342,
  "kmt2a_r_myeloid_diff": 0.0034,
  "null_percentile": 83.05
}
pubmed counts — the numbers behind the chart
{
  "YTHDC1 AND leukemia": 22,
  "(MLL OR KMT2A) AND YTHDC1": 1,
  "(MLL OR KMT2A) AND YT521-B": 1
}
depmap kmt2a selectivity
{
  "chronos_mean_kmt2a_rearranged": -0.471,
  "chronos_mean_other_leukaemia": -0.522,
  "difference": 0.051,
  "pan_essentiality_all_lines": -0.4806,
  "selectivity_rank": 14821,
  "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
  • Depleted in 42.5% of 1,208 DepMap lines with no index-arm enrichment: an inhibitor would be expected to act broadly rather than selectively.
  • The nuclear m6A reader sits downstream of the modification, so any resistance that restores target output by another route bypasses it.
  • The phase-separation framing, if it is wrong, would misdirect a design programme toward disrupting a compartment that is really clustered binding.
Still open
  1. Does the METTL3-YTHDC1-CTCF TAD mechanism exist in KMT2A::AFF1 B-lymphoid cells?
  2. Is the YTHDC1 nuclear body a phase-separated compartment or clustered binding, which the published assays cannot distinguish?
  3. Does the YTHDC1 inhibitor have any effect in a t(4;11) line - an experiment the compound makes cheap and nobody has reported?
Every source behind this entry (6)
  1. E1moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    In MLL-rearranged AML, METTL3 forms a complex with YTHDC1 and CTCF that co-localises at promoters and enhancers; YTHDC1 loss compromises CTCF-dependent 3D genome organisation, and YTHDC1 recognises m6A-modified architectural RNAs such as MALAT1 to enhance R-loop formation and sustain CTCF-mediated TAD activity at leukaemic loci.

    PMID 41430607DOI
    What this evidence cannot show

    Abstract read; full text not read. MLLr+ AML, no B-lymphoid arm. R-loop and Hi-C measurements are method-sensitive and the causal chain runs through several inference steps.

    Evidence that argues against this (1)
    • This project's scan gives YTHDC1 a POSITIVE index-arm difference in both lineage halves (+0.134 B-lymphoid, +0.003 myeloid) at the 83rd percentile of a random-gene null. The KMT2A-rearranged cells in DepMap show no extra requirement for the reader the mechanism is built on. Oncofusion knowledge graph (internal analysis) 2026
  2. E2moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    m6A on mRNA facilitates a phase-separated nuclear body, organised through YTHDC1, that suppresses transcription.

    PMID 34048709DOI
    What this evidence cannot show

    Abstract-level. A phase-separation claim, and this graph's standing caution applies: 1,6-hexanediol sensitivity and classical FRAP recovery cannot distinguish liquid-liquid phase separation from binding to spatially clustered sites, and only calibrated half-bleaching discriminates them. The abstract does not state which assay establishes the phase-separated character.

    Evidence that argues against this (1)
    • Binding to spatially clustered m6A sites is a substantive rival model and in a chromatin context the more testable one, because clustered binding predicts sequence-specific targeting while phase separation predicts bulk enrichment. Without a model-free discriminating assay the nuclear-body interpretation is unsettled. Fu R et al. 2025
  3. E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.paper

    Small-molecule inhibition of YTHDC1 has been reported as a strategy against acute myeloid leukaemia in mouse models.

    PMID 41637525DOI
    What this evidence cannot show

    Abstract-level, 2026, not independently reproduced. Mouse models of AML. Establishes that the YTH reader domain is ligandable; says nothing about a therapeutic window in B-lymphoid infant disease.

    Evidence that argues against this (1)
    • A ligandable pocket and efficacy in a mouse AML model do not transfer across the lineage boundary, and this beat's dependency data give YTHDC1 no KMT2A-rearranged enrichment in either lineage half. The compound is a reason to ask the question in B-lymphoid cells, not an answer. Oncofusion knowledge graph (internal analysis) 2026
  4. E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.analysis

    DepMap CRISPR (Chronos) gives YTHDC1 a mean gene effect of -0.481 across 1,208 lines with 42.5% depleted below -0.5, and index-arm differences of +0.134 (four B-lymphoid) and +0.003 (seven myeloid) against 43 other leukaemia lines, at the 83.1st percentile of a 466-gene null.

    What this evidence cannot show

    Four lymphoid index lines; a positive difference within the null is an absence of signal, not evidence of protection.

    Evidence that argues against this (1)
    • A two-to-three-week dropout in established lines cannot detect a requirement acting on genome architecture over longer timescales or on leukaemia initiation, and the published mechanism is about 3D genome maintenance rather than acute proliferation. Fu R et al. 2025
  5. E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database

    Counted: `(MLL OR KMT2A) AND YTHDC1` returns 1 PubMed record and the alias query `(MLL OR KMT2A) AND YT521-B` returns the same single record; `YTHDC1 AND leukemia` returns 22 and `YTHDC1 AND acute myeloid leukemia` returns 11.

    PubMed E-utilities esearch counts, alias-expanded and title-checked · NCBI PubMed (queries run 2026-09-20) 2026
    What this evidence cannot show

    Alias expansion was run and returned nothing new here, which is the point: the rule is that it must be run, not that it always recovers literature.

    Evidence that argues against this (1)
    • A single-record literature makes every statement about YTHDC1 in this disease provisional, including the negative ones. Nobody has looked at B-lymphoid KMT2A-rearranged cells at all. Fu R et al. 2025
  6. E6established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    UniProtKB Q96MU7 (YTDC1_HUMAN) is human YTHDC1, 727 amino acids; resolved by gene_exact query with the primary gene name read back as YTHDC1 and the length checked.

    What this evidence cannot show

    Single reviewed human hit whose primary gene name equals the query symbol. This project has logged six cases of gene_exact returning a protein that merely lists the queried symbol as a synonym, so the read-back is the point, not the query.

    Evidence that argues against this (1)
    • An accession establishes identity and length. It says nothing about whether this protein matters in KMT2A-rearranged leukaemia, which is what the rest of this entry is for. UniProt Consortium 2026
Connected to