I don't buy this argument. If you look at how GoF research is performed, serial passage[1] is the usual method and in that case you just use what works, you don

I don't buy this argument. If you look at how GoF research is performed, serial passage[1] is the usual method and in that case you just use what works, you don't 'choose' specific mutations. [1] https://en.m.wikipedia.org/wiki/Serial_passage https://en.m.wikipedia.org/wiki/Serial_passage