What maps a DNA break, and the fact that none of the modern instruments has ever been pointed at KMT2A in a blood progenitor
also called END-seq HTGTS sBLISS at KMT2A · DSB mapping assay landscape · why nobody has a break map of the KMT2A BCR
Three good methods now exist for finding every DNA break in a genome at single-base resolution. None of them has been published on the KMT2A gene in a blood stem cell, which is why the field still argues about where the break is and what makes it.
The same thing, for a scientist
The KMT2A breakpoint literature's structural claims - TOP2 consensus sites, DNase hypersensitivity, scaffold attachment - come from gel-based assays of the 1990s and 2000s. Three modern genome-wide DSB-mapping instruments have since been built. **END-seq** monitors DNA ends and resection at base-pair resolution in vivo and detects at least one DSB per cell among 10,000 cells without one [E1]. **HTGTS**, in its linear-amplification-mediated form, detects DSBs across the human genome by their translocation to a bait break, and is bait-anchored, which makes it the only one of the three that returns the PARTNER of each break [E2]. **sBLISS** labels DSB ends in situ in any cell that can be brought into suspension, including patient-derived cells and organoids, and returns a genome-wide endogenous break landscape [E3]. Searches run 2026-09-20 return **zero** records for KMT2A with END-seq and **zero** for KMT2A with HTGTS; the single record returned for the BLISS family is an unrelated carfilzomib paper [E4]. The break map that would settle where the KMT2A BCR is cut, in which cell type, and under what condition does not exist in the published record, and the instruments to make it have been available for a decade.
What each instrument can and cannot tell you1 deeper
| assay | measures | resolution | returns the partner? | key limit |
|---|---|---|---|---|
| END-seq 1 | free DNA ends and resection, genome-wide | base pair | no | population snapshot; a rapidly religated cleavage complex is undercounted |
| HTGTS (LAM-HTGTS) 2 | breaks detected via translocation to a chosen bait DSB | base pair at the junction | yes | requires a bait break; measures joined events, so it convolves break rate with joining |
| sBLISS 3 | endogenous and induced DSB ends labelled in situ | base pair | no | needs cells in suspension; label-and-amplify chemistry is sensitive to end structure |
For this beat the distinction that matters is the third column. Break maps tell you where KMT2A is cut. Only HTGTS tells you what it joins to, which is the partner-choice question mechanism-3d-genome-proximity-partner-choice and ana-kmt2a-partner-rank-vs-contact-frequency are about. This graph's mechanism-class-switch-recombination-programmed-breaks already lists as an open question that translocation-capture sequencing has never been applied with the KMT2A breakpoint cluster region as bait in a fetal-like progenitor. That experiment is specified in exp-htgts-kmt2a-bcr-bait-fetal-progenitor.
A fourth thing these assays change
All three measure endogenous breaks, not drug-induced ones. Every KMT2A break map that exists was made by adding etoposide, which is the condition relevant to therapy-related disease and not to the de novo infant case. An sBLISS profile of an untreated human fetal liver CD34+ cell would be the first measurement of whether the KMT2A BCR is a spontaneous break hotspot at all, as opposed to a drug-inducible one - and this graph's exp-human-cd34-kmt2a-breakage-assay records that KMT2A rearrangement bands appear in untreated cytokine-stimulated progenitors, which predicts the answer is yes.
The bounded absence, with the queries
Run 2026-09-20; querytranslation inspected for every zero and none contained a [UID] term or a mis-mapped MeSH phrase.
| query | count | note |
|---|---|---|
(MLL OR KMT2A) AND END-seq | 0 | clean translation, "END-seq"[All Fields] |
(MLL OR KMT2A) AND HTGTS | 0 | clean translation, "HTGTS"[All Fields] |
(MLL OR KMT2A) AND (BLISS OR sBLISS OR DSBCapture OR BLESS) | 1 | the single hit is PMID 33937032, a carfilzomib preclinical study, matched on an unrelated token - bounded rejection recorded |
The absence is bounded to title/abstract indexing. A break map covering KMT2A inside a genome-wide study whose abstract does not name the gene would be invisible to these queries, and that is the likeliest way this absence is wrong: the loop-anchor papers map TOP2 breaks genome-wide and the data would contain the KMT2A locus without the abstract saying so. The honest claim is that no published study reports a DSB map AT KMT2A in a haematopoietic progenitor, not that no deposited data covers the locus. Re-analysis of deposited END-seq and sBLISS data for the KMT2A interval is therefore the cheapest version of this whole question and should be tried before any new sequencing.
| script | analyses/kmt2a-dsb-map-reanalysis/run.py |
|---|---|
| status | planned |
| cheapest next step | re-analyse deposited END-seq and sBLISS datasets over the GRCh38 KMT2A interval chr11:118436456-118526832 before commissioning any new sequencing |
| why it matters | every existing KMT2A break measurement is drug-induced; none of these three assays has been used to ask whether the BCR is a SPONTANEOUS break hotspot in an untreated haematopoietic progenitor |
| cost tier | low |
| result summary | not run |
assays
[
{
"name": "END-seq",
"measures": "free ends and resection",
"returns_partner": false,
"ref": "E1"
},
{
"name": "LAM-HTGTS",
"measures": "breaks via translocation to a bait DSB",
"returns_partner": true,
"ref": "E2"
},
{
"name": "sBLISS",
"measures": "endogenous and induced DSB ends labelled in situ",
"returns_partner": false,
"ref": "E3"
}
]counted absences 2026 09 20 — the numbers behind the chart
[
{
"q": "(MLL OR KMT2A) AND END-seq",
"count": 0
},
{
"q": "(MLL OR KMT2A) AND HTGTS",
"count": 0
},
{
"q": "(MLL OR KMT2A) AND (BLISS OR sBLISS OR DSBCapture OR BLESS)",
"count": 1,
"note": "single hit PMID 33937032 is an unrelated carfilzomib study - bounded rejection"
}
]- A break map is descriptive. Knowing exactly where KMT2A is cut in a fetal progenitor changes no treatment and, given that every candidate node upstream of the break is an essential gene, is unlikely to yield a target.
- A negative - the BCR shows no spontaneous break enrichment - would undercut the entire endogenous-fragility framing this graph carries in several entries, and should be actively sought rather than avoided.
- Do the deposited break maps from the loop-anchor studies contain coverage of the KMT2A breakpoint cluster region, and what do they show there? This is a download and an interval query, and it is the single cheapest open question in this beat.
- Is the KMT2A BCR a spontaneous DSB hotspot in an untreated human haematopoietic progenitor, or only a drug-inducible one? No assay capable of answering this has been applied.
- Can END-seq or sBLISS be run on human fetal liver CD34+ cells at achievable input? sBLISS's suspension format suggests yes; END-seq's agarose-plug format is the harder one.
- Would a TOP2-adduct-aware protocol be needed at this locus? sBLISS labels processed ends, so a trapped cleavage complex is invisible until TDP2 or the proteasome has acted, which means the assay measures the second step of a two-step process.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
END-seq monitors DNA end resection and double-strand breaks genome-wide at base-pair resolution in vivo and can detect at least one DSB per cell among 10,000 cells not harbouring DSBs; chromatin features beyond sequence preference dictate the repertoire of enzyme-induced breaks.
DNA Breaks and End Resection Measured Genome-wide by End Sequencing · Canela A, Sridharan S, Sciascia N, et al. · Molecular cell 2016What this evidence cannot show
Sensitivity figures are for engineered synchronous nuclease cuts. Requires cells embedded in agarose plugs, which constrains input material and complicates primary fetal tissue. Measures ends present at one moment in a population.
Evidence that argues against this (1)
- The same paper reports breaks 'distributed over extended regions' during class switch recombination, so a base-pair-resolution instrument does not convert a diffuse lesion into a point lesion. A hotspot called from END-seq inherits the width of the underlying biology, which matters at the KMT2A BCR where the candidate mechanisms predict cleavage anywhere across a 0.4-8.3 kb interval rather than at a defined position. Canela A et al. 2016
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Linear amplification-mediated high-throughput genome-wide translocation sequencing robustly detects DNA double-strand breaks across the human genome by their translocation to other endogenous or ectopic DSBs, revealing off-target hotspots and translocations between targets on homologous chromosomes.
Genome-wide detection of DNA double-stranded breaks induced by engineered nucleases · Frock RL, Hu J, Meyers RM, Ho YJ, Kii E, Alt FW · Nature biotechnology 2015What this evidence cannot show
The method requires a bait DSB, usually nuclease-induced, and reports breaks only insofar as they JOIN to that bait. It therefore measures the product of break frequency and joining propensity, and a locus that breaks often but rejoins locally will be under-represented.
Evidence that argues against this (1)
- A nuclease-made bait break is chemically unlike a TOP2-made break, which carries a protein adduct and a 4-nt 5'-cohesive overhang, and the two are joined by pathways with opposite effects on translocation: nuclease-induced translocations in human cells are GENERATED by classical NHEJ, whereas at TOP2 breaks the equivalent TDP2-dependent route SUPPRESSES translocation. An HTGTS partner spectrum obtained with a Cas9 bait at KMT2A is therefore not guaranteed to reproduce the spectrum a TOP2-made break would give. Gomez-Herreros F et al. 2017
- E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
sBLISS labels double-strand break ends in situ and sequences them to give genome-wide profiles of endogenous and induced DSBs in any cell type that can be brought into suspension, including patient-derived cells, organoids and isolated nuclei.
Genome-wide detection of DNA double-strand breaks by in-suspension BLISS · Bouwman BAM, Agostini F, Garnerone S, et al. · Nature protocols 2020What this evidence cannot show
A protocol paper. In-situ end labelling depends on the chemistry of the break end, so a protein-adducted TOP2 end is not labelled until the adduct is removed - which means sBLISS reports post-processing ends, not trapped cleavage complexes.
Evidence that argues against this (1)
- What an endogenous break map shows is strongly shaped by the transcriptional context rather than by the locus alone: sBLISS maps in cancer cells place transcription-stress-associated DSBs under the control of the superenhancer regulatory landscape, with gamma-H2AX marking and repair turnover differing sharply between DSB-enriched genes. A KMT2A break map would therefore be a statement about that cell's transcriptional state as much as about the locus, and could not be generalised across cell types. Hidmi O et al. 2026
- E4moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database
PubMed searches run 2026-09-20 return zero records linking KMT2A or MLL to END-seq and zero to HTGTS, with clean querytranslations, and the single record returned for the BLISS family is an unrelated preclinical carfilzomib study.
PubMed E-utilities counted-absence searches for genome-wide DSB-mapping assays at KMT2A, run 2026-09-20; querytranslation inspected for each zero · Q32-break-machinery · NCBI E-utilities 2026What this evidence cannot show
Title/abstract indexing only, and method acronyms are exactly the kind of term a paper may use only in its methods section. The claim is that no published study REPORTS a DSB map at KMT2A in a haematopoietic progenitor, not that no deposited dataset covers the locus.
Evidence that argues against this (1)
- Genome-wide TOP2 break mapping HAS been performed in studies that specifically analyse MLL translocations, so deposited data covering the KMT2A locus almost certainly exists even though no abstract reports a locus-level map. The right first move is re-analysis of those deposited break maps over the KMT2A interval, not new sequencing, and an absence claim that does not say this would misdirect the work. Gothe HJ et al. 2019
re-analysis of deposited break maps over the KMT2A interval is the cheapest test of the loop-anchor claim at this locus
no published DSB map covers the KMT2A BCR in a haematopoietic progenitor
names HTGTS as the only one of the three assays that returns the partner
the existing assay measures rearrangement; these three measure the break itself and have never been applied here
executes the gap that analysis names: no modern break-mapping method has been published on KMT2A in a blood progenitor
converts that entry's assay survey into measured per-assay signal at the element