I’m not a researcher. I think we discovered the easily-comparable variation used in forensics quite a long time ago before discovering (both healthy and pathoge

I’m not a researcher. I think we discovered the easily-comparable variation used in forensics quite a long time ago before discovering (both healthy and pathogenic) mosaicism. I _think_ the former relies on optical comparison while the latter relies on a full sequence. So, a court would prefer the long legacy of statistics about polymorphism before more-complicated sequencing. But that leaves out your specific edge case. For example, some cancers alter regions of chromosomes or jettison the Y chromosome altogether. An interesting counterexample may be monozygotic twins. I imagine an optical comparison could confirm either or both twins match, but later-developed tests might rule one out. Those tests might be sequencing, or microbiome, etc. If the sample shows evidence of cancer, and only one twin has (some) cancer, I think we should not presume that the DNA evidence is strong enough to prove anything.