Is fusion intended in biology sometimes? Yes, and the answer has a strategic edge

Comparative structural biology; the evolutionary account has a published alternative (PMID 34394111) and should not be presented as settled.

• The direct measurement has overtaken the analogy. This entry uses RAG domestication to argue that programmed genome breakage is an engineered-risk system and, by extension, that a fusion-junction-directed nuclease would be borrowing a dangerous tool. Leibowitz et al. now show the danger directly for the tool actually on offer: CRISPR-Cas9 editing generates micronuclei and chromosome bridges that initiate chromothripsis, demonstrated by single-cell whole-genome sequencing in model cells and by editing at a clinically relevant locus in clinically relevant cells. The 'borrowing' half of this entry's argument should be re-grounded on that result rather than on an evolutionary analogy, which makes the conclusion firmer and the reasoning shorter. Read: abstract only. Leibowitz ML et al. 2021

Mouse genetics; the separability is partial and the 'not required' calls have limited power for partial effects.

• A second laboratory measured the same separability and reached the opposite conclusion about the AID half of it. Translocations between the Igh switch region and Myc were measured in AID-deficient mice, and AID made no measurable contribution to the generation of the INITIAL translocations - only to the outgrowth of translocation-positive cells. If AID is not required for the breakage-and-joining event, then 'requires AID and UNG but not H2AX, 53BP1 or Ku80' is not a clean genetic dissection of a programmed pathway's oncogenic output; it may be a dissection of what lets a translocated cell expand. Unniraman S et al. 2004

Yeast; the chemical analogy to TOP2 is real but SPO11 and TOP2 differ in regulation, partners and adduct-resolution pathway.

• 'The same chemistry' holds at the level of the covalent intermediate and breaks down at the level of the enzyme. Meiotic break formation requires a TopoVIB-Like partner subunit, identified in mouse and plants as an essential conserved family: the meiotic machine is a two-subunit type IIB topoisomerase-like complex, whereas TOP2 is a homodimeric type IIA enzyme with no equivalent partner requirement. The 1997 assignment, made on sequence homology in yeast, could not see this. So the analogy that licenses reading meiotic break biology across to KMT2A cleavage is narrower than it looks - a shared 5'-phosphotyrosyl intermediate, not a shared enzyme or a shared regulation. Robert T et al. 2016

10 cases; caller- and reference-dependent.

• The defensive conclusion drawn here is right but its mechanism attribution is partly wrong. This entry says healthy cells make chimeric transcripts by trans-splicing and read-through, so an RT-PCR fusion signal is weak evidence of a rearranged genome. Yan et al. measured, using a cross-species RNA spike-in, that approximately 1% of RNA-seq reads are artifactually chimeric, produced by random template switching during reverse transcription rather than by any cellular process, and that real events are distinguished by reproducing the same chimeric fragment across independent biological isolates. A substantial share of the chimeric signal the field attributes to healthy-cell biology may be library chemistry. The practical instruction is unchanged and in fact simplified: confirm at the genome, and require cross-isolate reproducibility of any RNA-level chimera claim. Read: abstract only. Yan B et al. 2021

Applies to ETV6::RUNX1 and explicitly not to KMT2A.

• Searched and empty, recorded as a fact. The generalisation from E5 -- RAG off-target activity drives oncogenic rearrangement in ETV6::RUNX1 ALL, therefore programmed machinery gets hijacked -- is not established for KMT2A. This pass found no primary report that RAG or AID activity generates KMT2A breaks; the mechanism attributed to KMT2A breakage in this graph is topoisomerase-II-mediated cleavage, which is chemistry shared with SPO11 (E3) but is not a lymphoid programmed-recombination pathway. The entry's strategic claim therefore rests on an analogy across fusion families and should say so. No citation is attached because nothing was found to cite.