Oncofusion

start / graph / ana-index-fusion-h3k79-evidence-that-exists

Analysismoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q-partner-provenance

The imports that were never necessary: H3K79 methylation in KMT2A::AFF1 has its own literature, and this graph barely cites it

also called index fusion H3K79 evidence · unnecessary imports · Krivtsov 2008 undercited

Not every borrowed result needed borrowing. There is a body of work done directly in MLL-AF4 leukaemia showing the H3K79 tag matters there, and this project has been leaning on mouse MLL-AF9 papers instead of citing it.

The same thing, for a scientist

The third bucket the audit was asked for - imports that turn out to have independent index-fusion evidence nobody cited - is real and has a clear centre of mass. Four papers establish H3K79 biology **in the index fusion or the index disease**: | PMID | what it establishes in MLL-AF4 / MLL-r ALL | entries citing it in this graph | |---|---|---:| | 18977325 | H3K79 profiles define murine *Mll-AF4* and human MLL-rearranged ALL; DOT1L suppression silences MLL-AF4 targets | **1** | | 28076791 | MLL-AF4 spreading co-occurs with menin spreading and H3K79me2/3, and governs DOT1L-inhibitor sensitivity | 9 | | 31243293 | DOT1L inhibition defines a distinct subset of H3K79-dependent enhancers | 5 | | 26711339 | MLL-rearranged **ALL** activates BCL-2 through H3K79 methylation and is ABT-199-sensitive | 2 | **Thirteen entries in total cite at least one** [E5], against 46 citing the murine MLL-AF9 keystone set. The most index-specific of the four - Krivtsov 2008, which built the *Mll-AF4* model and profiled H3K79 in it - is cited **once** [E1]. **What this changes.** It splits the DOT1L story cleanly in two. The **phenotype** (H3K79 methylation is present, elevated at fusion targets, and functionally required in KMT2A::AFF1 disease) is index-established and does not need importing. The **mechanism** (how the fusion reaches DOT1L) is imported, and cannot be, because the interface is absent (`mechanism-kmt2a-aff1-dot1l-recruitment-route`). An audit that reported '54% imported' without this split would have overstated the damage.

Go deeper
Why the index literature is under-cited, mechanically

Three reasons, all checkable and all fixable.

  1. It is older. Krivtsov 2008 predates the DOT1L-inhibitor era, so it is not in the citation neighbourhood of the papers a search for 'DOT1L KMT2A' returns first.
  2. It uses the retired-style symbol. MLL-AF4[tiab] AND H3K79[tiab] returns 7 records; KMT2A-AFF1[tiab] AND H3K79[tiab] returns 2, neither of which is Krivtsov 5. A beat keyed to current HGNC nomenclature misses the index-fusion literature almost entirely - the same failure mode the septin beat found in cycle 40, here operating on the project's own central fusion.
  3. It reports a profile, not a mechanism. Entries are usually written around a mechanistic claim, and a correlative profiling paper does not supply one, so it loses to a paper that does even when it is closer to the disease.
The search that should be standard before importing anything

Before importing a mechanism into KMT2A::AFF1, run the index-fusion query in both nomenclatures and check the count:

MLL-AF4[tiab] AND <topic>[tiab]         # 7 for H3K79
KMT2A-AFF1[tiab] AND <topic>[tiab]      # 2 for H3K79
t(4;11) is unusable as a PubMed term - parentheses and semicolons parse as syntax

t(4;11)[tiab] AND DOT1L[tiab] returns 3 records, and the karyotype string is exactly the trap the NUP98 beat documented: a parenthesised karyotype is parsed as syntax, so any absence claim built on one is worthless 5. Use MLL-AF4, AFF1, AF4 and KMT2A-AFF1 as separate queries.

Facts and figures
scriptad hoc PMID string match plus NCBI esearch counts, 2026-09-19
statusrun
result summaryThe H3K79 phenotype is index-established (4 papers, 13 citing entries) while the recruitment mechanism is imported; Krivtsov 2008, the most index-specific of the four, is cited once in 848 entries.
citing entries
189773251.00018977325: 1267113392.00026711339: 2312432935.00031243293: 5280767919.00028076791: 9union13.0union: 13
pubmed counts
KMT2A-AFF1 AND H3K792.000KMT2A-AFF1 AND H3K79: 2t(4;11) AND DOT1L3.000t(4;11) AND DOT1L: 3MLL-AF4 AND H3K797.000MLL-AF4 AND H3K79: 7
index fusion pmids
[
  "18977325",
  "28076791",
  "31243293",
  "26711339"
]
citing entries — the numbers behind the chart
{
  "18977325": 1,
  "26711339": 2,
  "28076791": 9,
  "31243293": 5,
  "union": 13
}
pubmed counts — the numbers behind the chart
{
  "MLL-AF4 AND H3K79": 7,
  "KMT2A-AFF1 AND H3K79": 2,
  "t(4;11) AND DOT1L": 3
}
How this could bite us back
  • Concluding that the H3K79 phenotype is index-established could licence a DOT1L-inhibitor programme in infant disease on evidence that is, at bottom, one 2008 profiling paper and one laboratory's cell-line work.
  • If the mark is present in KMT2A::AFF1 without a fusion-DOT1L interface, then removing DOT1L removes an activity the fusion did not place there, and the therapeutic window is set by how much H3K79 methylation normal lymphopoiesis needs rather than by fusion selectivity.
Still open
  1. Which other imported mechanisms in this graph have an index-fusion paper nobody cited? The H3K79 case was found by searching one topic in two nomenclatures; the same search has not been run for menin, PAF1c, SEC or PRC1.
  2. Is the murine Mll-AF4 model of Krivtsov 2008 still available and used by anyone? If it is, the CRISPR-ChIP partition could be tested in the index fusion in a system that already exists.
Every source behind this entry (5)
  1. E1strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    A conditional murine Mll-AF4 model produces B-precursor ALL or AML; ChIP-chip H3K79 methylation profiles correlate with Mll-AF4-associated expression in murine ALL and in human MLL-rearranged leukaemias; human MLL-rearranged ALLs are distinguishable from other ALLs by their H3K79 profiles; and DOT1L suppression inhibits expression of critical MLL-AF4 target genes.

    H3K79 methylation profiles define murine and human MLL-AF4 leukemias. · Krivtsov AV, Feng Z, Lemieux ME, Faber J, Vempati S, Sinha AU, Xia X, Jesneck J, Bracken AP, Silverman LB, Kutok JL, Kung AL, Armstrong SA · Cancer Cell 2008
    PMID 18977325DOI
    What this evidence cannot show

    Abstract fetched from PubMed and read on 2026-09-19; full text not read, every identifier copied from that record. ChIP-chip rather than ChIP-seq and uncalibrated, so this is profile correlation and not quantitative occupancy. The mouse model is a chimeric knock-in Mll-AF4, not the human t(4;11) product; the human arm is partner-mixed MLL-rearranged ALL. 'Suppression of DOT1L' in 2008 means RNAi.

    Evidence that argues against this (1)
    • Profile correlation plus knockdown sensitivity establishes that the mark matters in this fusion; it does not establish that the fusion recruits the writer, so this paper closes the phenotype gap and leaves the mechanism gap exactly where it was. Biswas D et al. 2011
  2. E2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    MLL-AF4 binding spreads into the gene body at a subset of targets with spreading of menin binding, increased transcription, increased H3K79me2/3, disrupted H3K36me3 and unmethylated CpG regions, and spreading-gene expression is preferentially downregulated by DOT1L inhibition.

    MLL-AF4 Spreading Identifies Binding Sites that Are Distinct from Super-Enhancers and that Govern Sensitivity to DOT1L Inhibition in Leukemia. · Kerry J, Godfrey L, Repapi E, Tapia M, Blackledge NP, Ma H, Ballabio E, O'Byrne S, Ponthan F, Heidenreich O, Roy A, Roberts I, Konopleva M, Klose RJ, Geng H, Milne TA · Cell Rep 2017
    PMID 28076791DOI
    What this evidence cannot show

    Abstract fetched from PubMed and read on 2026-09-19; full text not read, every identifier copied from that record. Human MLL-AF4 ChIP-seq; the abstract does not name the cell lines, so the breadth of the t(4;11) evidence is not bounded by this citation alone. 'Spreading' is the authors' own peak-calling convention. **Resolved, and it removes a hedge this item did not need.** GSE83671 is **30 of 59 samples SEM** - annotated `Paediatric pro B-cell line derived from ALL with t(4;11)`, i.e. the index fusion - alongside ML-2 t(6;11), THP-1 t(9;11), KOPN-8 t(11;19), MV4-11 t(4;11), CCRF-CEM, RCH-ACV, a presentation-ALL patient-derived xenograft, **cord blood and second-trimester fetal bone marrow**, and an EPZ-5676 dose series. The breadth IS bounded by this citation, and favourably. Resolved 2026-09-19 by the Q18b unstated-systems sweep from the study's deposition, not from its text; bounded as a statement about the deposited data rather than about the whole study.

    Evidence that argues against this (2)
    • Menin spreading and H3K79me2/3 tracking MLL-AF4 spreading together is equally consistent with both following elevated transcription; the DOT1L-inhibitor contrast compares marked genes with each other rather than testing how the mark arrived. Godfrey L et al. 2019
    • The correction runs the other way for once, but the concentration remains: 30 of the 59 samples are a single cell line, so the index-fusion evidence here is SEM-heavy rather than broad across KMT2A::AFF1 material, and an SEM-specific artefact would be hard to distinguish from a fusion-general result. NCBI Gene Expression Omnibus 2026
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    DOT1L inhibition reveals a distinct subset of enhancers dependent on H3K79 methylation.

    DOT1L inhibition reveals a distinct subset of enhancers dependent on H3K79 methylation. · Godfrey L, Crump NT, Thorne R, Lau IJ, Repapi E, Dimou D, Smith AL, Harman JR, Telenius JM, Oudelaar AM, Downes DJ, Vyas P, Hughes JR, Milne TA · Nat Commun 2019
    PMID 31243293DOI
    What this evidence cannot show

    Identifier, title, journal, year and DOI fetched from PubMed on 2026-09-19 and copied from that record; abstract not re-read this cycle, so the claim is asserted at the level of the title's own statement. Same laboratory and largely the same human MLL-AF4 system as E2, so E2 and E3 are not independent replications.

    Evidence that argues against this (1)
    • An enhancer subset defined by drug response is defined by the drug's kinetics as much as by chromatin: DOT1L inhibitors act slowly through dilution of an existing mark, so 'dependent' may mean 'fastest to lose the mark'. Krivtsov AV et al. 2008
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    MLL-rearranged acute lymphoblastic leukaemias activate BCL-2 through H3K79 methylation and are sensitive to the BCL-2-specific antagonist ABT-199.

    MLL-Rearranged Acute Lymphoblastic Leukemias Activate BCL-2 through H3K79 Methylation and Are Sensitive to the BCL-2-Specific Antagonist ABT-199. · Benito JM, Godfrey L, Kojima K, Hogdal L, Wunderlich M, Geng H, Marzo I, Harutyunyan KG, Golfman L, North P, Kerry J, Ballabio E, Chonghaile TN, Gonzalo O, Qiu Y, Jeremias I, Debose L, O'Brien E, Ma H, Zhou P, Jacamo R, Park E, Coombes KR, Zhang N, Thomas DA, O'Brien S, Kantarjian HM, Leverson JD, Kornblau SM, Andreeff M, Müschen M, Zweidler-McKay PA, Mulloy JC, Letai A, Milne TA, Konopleva M · Cell Rep 2015
    PMID 26711339DOI
    What this evidence cannot show

    Abstract fetched from PubMed and read on 2026-09-19; full text not read, every identifier copied from that record. 'MLL-rearranged ALL' is partner-mixed rather than t(4;11)-specific. A 36-author multicentre study - robust in aggregate, hard to attribute any single mechanistic step to.

    Evidence that argues against this (1)
    • BCL-2 dependence is broadly lineage-associated in leukaemia, so an ALL-versus-AML difference in ABT-199 sensitivity need not be caused by the fusion's chromatin programme; the H3K79 link is the part of this claim that would need separating out. Pei S et al. 2020
  5. E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

    In this graph, 13 of 848 entries cite at least one of the four index-fusion H3K79 papers and 46 cite at least one of the five murine MLL-AF9 keystone papers; PubMed returns 7 records for MLL-AF4[tiab] AND H3K79[tiab], 2 for KMT2A-AFF1[tiab] AND H3K79[tiab], and 3 for t(4;11)[tiab] AND DOT1L[tiab].

    What this evidence cannot show

    Graph counts are our own, by literal PMID string match over node JSON on 2026-09-19, and count an entry once whether the citation is load-bearing or a counter-evidence note. The PubMed counts are esearch `count` values retrieved on 2026-09-19 and were not individually title-checked; the t(4;11) query in particular is unreliable because PubMed parses the parentheses and semicolon as syntax rather than text, which is why it is quoted as a demonstration of the trap rather than as a finding.

    Evidence that argues against this (1)
    • A 13-versus-46 citation ratio partly reflects that the MLL-AF9 papers answer more different questions - dependency, allele dosage, catalysis, elongation - while the index-fusion papers all answer one. Under-citation and narrower scope look identical in a count. Oncofusion knowledge graph et al. 2026
Connected to