Theoretical predictions on the equality of radiation-produced dicentrics and translocations detected by chromosome painting.

Existing models of chromosome aberrations produced by ionizing-radiation predict equal numbers of dicentrics and translocations if the dose is so low that complex aberrations can be ignored. We show that, for a specific subset of aberrations detected by FISH, dicentric/translocation equality is predicted even at higher doses. Assuming one-colour whole-chromosome painting (with unpainted chromosomes counterstained and centromeres recognizable) the relevant restriction is that the final metaphase pattern be, in the terminology of Simpson and Savage, 'apparently simple'. This means that the painted pattern is required to have the colour/centromere appearance corresponding to a single complete reciprocal exchange but its actual formation, as reflected for example in lengths, is allowed to be more complicated. The restriction to apparent simplicity is significantly less limiting than ignoring all complex aberrations. Our analysis of predicted dicentric/translocation equality in this case uses examples, a combinatorial counting method, Monte Carlo computer programs, and a duality proof. However, we argue that for 'visibly complex' dicentrics or translocations, no similar equality is expected in general. Corresponding experimental results are briefly surveyed. Checking dicentric/translocation equality experimentally can provide a significant test of current chromosome aberration models.