I don't understand something: I was taught that mutations are random, not even predictable (the latter comes from Stuart A. Kauffman book that I've read recently). How did they construct the set of "all possible mutations"? Do we even know enough on chemistry / physics level to make such claims?
Human DNA has ~3 billion base pairs, and each base pair could be switched to one of the other 3, so there are 9 billion possible single-change mutations. With 8 billion people, most having multiple mutations, most mutations exist.
So it's not the set of all possible mutations, just single base pair ones.
>With 8 billion people, most having multiple mutations, most mutations exist.
Not saying that this isn't the case, but the conclusion doesn't derive from those premises alone though.
The fact that number of people > possible mutations, or even that people have more than one each, is not enough. There could be a trillion people and still the total sum of mutations seen be a tiny subset of the 9 billion possible mutations.
The missing element is about the distribution of the mutations, whether all are equally likely or at least or are possible to arrive at, and so on. E.g. the premises could very well hold, but some compounding factor could push towards a subset of them appearing, etc.
