Re: Once upon a time detector vans existed
Back in the lab in the '90s we once did a back of the envelope calculation of the number of identical patterns from independent injection events you needed to conclude a transgene pattern was real and not influenced by where in the genome it had integrated*. The answer turned out to be 3. This also includes no expression where you had mutated a binding site in your transgene enhancer which abolished expression. I was doing a bit of that but nevertheless acquired at least 5 before accepting the result.
it works because the genome is so large.
*Which assumes integration is essentially random. Not entirely true and sequence dependent as we know from the gene therapy on people with SCIID immune deficiency where the transgene in their bone marrow gave them leukaemia. But for all practical reasons it is not unreasonable.