There are two relevant posts on Uncommon Descent: Branko Kozulic Responds and Branko Kozulic responds to Professor Moran, Part II.
I'll respond to the second one since it is more specific. He begins with ...
The idea of 100 mutations being fixed in the human population in each generation over a period of 185,000 generations, or 5,000,000 years, has always appeared intuitively unrealistic to me, possibly because I am primarily a practical scientist.Many things in science seem counter-intuitive. That's why you have to make an effort to understand the science. In this case, you've been arguing against evolution for many years so you've had plenty of opportunity to get beyond intuition.
My first question was: How many mutations are fixed now, in my generation? I am pretty sure that the answer is zero. Fixation means, we agree, that all individuals in the human population acquire the same mutation in a generation, or, equivalently, that the other allele is completely lost. If so, then the question logically follows: When did this change from 100 to zero fixed mutations per generation take place? Alternatively, is there perhaps something wrong with my initial assumptions?Your definition of fixation leaves a lot to be desired. I do not agree with it.
I'll use the definition on Wikipedia [Fixation (population genetics)].
It wasn't very hard to find this definition. In future, I suggest you check Wikipedia when you are tempted to make up definitions. It will save time.In population genetics, fixation is the change in a gene pool from a situation where there exists at least two variants of a particular gene (allele) to a situation where only one of the alleles remains. The term can refer to a gene in general or particular nucleotide position in the DNA chain (locus).
In the process of substitution, a previously non-existent allele arises by mutation and undergoes fixation by spreading through the population by random genetic drift and/or positive selection. Once the frequency of the allele is at 100%, i.e. being the only gene variant present in any member, it is said to be "fixed" in the population.
The situation with humans gets a bit complicated because the population has expanded so much in the past 100,000 years. If we think about the simple situation in the past when the effective population size was about 10,000, there were about 100 alleles fixed (and lost) per generation. In each generation there would have been a very small number of individuals who carried an ancient allele and a very large number (99.9%) who carried an allele that first appeared one million years earlier. By chance, the individuals carrying the ancient allele failed to pass it on to their offspring. The result is that the other allele becomes fixed in the population (100%). This happens, on average, 100 times in each generation.
After wandering around, a solution entered my mind while reading the following sentence in Professor Moran’s reply: "In my calculation, the values of Ne are irrelevant, so any value will work equally well, as long as you realize that you are starting with an ancestral population containing existing variation." Here, the key phrase is "containing existing variation." That is the essential assumption. What it means in reality is this: throughout the whole period leading up to the fixation – i.e. in the case we are considering, a population of 10,000 individuals over 40,000 (=4N) generations (which, when multiplied by 20 years/generation, equals 800,000 years) – a group of individuals may split off from the population, but no newcomers are allowed to enter it – that is, the population must remain closed in terms of mating. Here, a newcomer is defined as any individual from a group that had lived separately, say on an adjacent island or across a mountain, for some generations. Whenever a newcomer joins the population, the fixation process is broken, because migrants bring with them new variation that did not exist initially. Migration is a powerful force acting against genetic divergence among sub-populations (Hartl & Clark, Principles of Population Genetics, pp. 295-309).In my example we were dealing with the entire species and not subpopulations.
There has to be some gene flow between subpopulations otherwise they become equivalent to species (in the strict sense of the biological species concept). So, you are correct. There will be migrants and subpopulations. What this means is that the actual path to fixation (or elimination) gets complicated as it becomes fixed in one subpopulation then spreads to another. However, that doesn't affect the overall averages. There will still be about 100 "new" alleles fixed in the species per generation.
Herein lies the weak point in Professor Moran’s calculations. Based on our knowledge of human history and behavior, we know that groups of people grow in numbers, split in two and merge, repeatedly. This commonly known fact has to be ruled out, in order to allow for the fixation of 100 mutations per generation. And now we can see that the value of Ne will have a major impact: a large Ne makes it more likely that a newcomer will join the population, and also that a group will split away from it and then re-enter it (in part) later on.Like I said earlier, the math gets more complicated with subpopulations but I think you can see that the sizes of the individual subpopulations don't have a significant effect on the big picture. They cancel out for the subpopulations just as they do for the species as a whole.
In view of the above, I believe I am entitled to continue rejecting the feasibility of 22,000,000 mutations being fixed in the human population over a period of 5,000,000 years, or 100 mutations each generation.You are entitled to reject all of population genetics for all I care. What you are NOT entitled to do is pretend that you have refuted all of the textbooks by just thinking about subpopulations and migrants. Do you honestly think that no population geneticist has ever thought about these things—and dealt with them— in the past 100 years?
Here it is appropriate to mention a reservation relating to the exact number of fixed differences between the human and chimp genomes that has been expressed by Professor Felsenstein: the actual number might be lower than 22,000,000.We don't know the exact number but 22 million seems like a pretty good approximation in order to illustrate the point.
I don't think Joe Felsenstein really questions this value for the species as a whole. What he points out is that you don't really need to emphasize fixed alleles in order to show that the population genetics equations work for humans and chimpanzees. The equations, and our understanding of evolution, work even if you just compare the differences between any randomly chosen human and a randomly chosen chimpanzee.
The entire field of molecular evolution is based on the idea that a lot of evolution occurs by fixation of nearly neutral alleles by random genetic drift. It's the reason why there's an approximate molecular clock and it's the reason why we can construct phylogenetic trees based on sequence comparisons. There's a massive amount of data supporting the basic concepts of population genetics. This is not just something that scientists made up to confuse creationists with counter-intuitive ideas.
Another example is the very great similarity of the DNA sequences in the human and chimpanzee genomes. In fact, extensive comparisons of long sections of the DNA of man and chimpanzee show that the differences are extraordinarily trivial. Human and chimpanzee sequences differ on average at only one base per hundred. As far as we can tell, not only are the DNA sequences virtually identical, but every gene identified in the human genome has its counterpart in the chimpanzee genome. So all the morphological differences between man and chimpanzee, involving the form and relative shape of the limbs, the genital organs, sperm morphology etc., and all the mental differences are generated from DNA sequences which are virtually identical. The distance between man and chimp which seems so significant at a gross morphological level is trivial in DNA sequence space. In fact, the differences between the DNA of man and chimp can be accounted for by simple well-known mutational processes which are occurring all the time in nature at present.