But John Sanford's models don't bear close examination.
He based his model a certain distribution of the "value" of mutations, i.e. how many are beneficial, and how many are deleterious. He also defined an "effectively neutral zone" where mutations of small effect will not be selected for or against. Sanford took these models from a researcher called Kimura. but, Kimura's paper actually gives several models for the distribution of mutation "worth" or "value" versus frequency. Some of these models led to massively too fast evolution, and some led to devolution. Sanford chose one of Kimura's distributions that led to devolution, without adequately explaining why. He also decided that all beneficial mutations would be too small to be beneficial, to fall in the "effectively neutral zone". There's a heading in the review you link to, "good and bad mutations inseparable". But laboratory experiments show that beneficial mutations do appear in populations. This shows that Sanford's parameters do not match reality, and hence his conclusions are meaningless.
Secondly, Sanford assumes that mutations are independent. So that if a mutation is only slightly deleterious, that it won't be selected against. This ignores the fact that deleterious mutations can be cumulative.
Imagine that we have an organism with one tiny deleterious mutation. The difference this single mutation makes will be too small for there to be a noticeable selection pressure. But, if an organism has 50,000 such slightly deleterious mutations, the cumulative weight of those mutations can severely cripple the organism, and it will be flushed out of the gene pool. Sanford's claim that natural selection is ineffective is based on this inappropriate independence assumption.
Kimura does good simulation/model research. He considered the uncertainty we have concerning how many mutations are beneficial, and tried experiments with different distributions of mutation worth versus frequency. And found that these different distributions led to different results. We don't know what the distribution of deleterious versus beneficial mutations are in humans, and we don't know how these mutations combine to cumulatively affect the viability/competitive fitness of the individual. Kimura uses his model to help understand what we do and do not know. Sanford doesn't.
This article convincingly deconstructs Sanford's techniques and conclusions: STAN 4 | Letters to Creationists It is very interesting reading, and IMHO is a very useful article to read to understand the limitations of computer simulation.
Upvote
0