The Talk.Origins Archive: Exploring the Creation/Evolution Controversy

Plagiarized Errors and Molecular Genetics

A Response to Carl Wieland

[Last Update: March 26, 2002]

Dear Carl Wieland,

I have just come across your article on the Web entitled 'Junk-making' viruses neutralize an evolutionary argument, in which you claim that the evolutionary argument based on shared pseudogenes is invalid. As you may be aware, I have written an extensive article (posted at the Talk.Origins Archive) in which I outline in detail why retroposons/pseudogenes shared between species argue strongly for common ancestry and therefore support evolution. I am writing to you now to explain why I believe the conclusions of your article to be erroneous.

1. You write "there is no consistent pattern of pseudo-genes (sic) in humans, chimps and gorillas from which it could be argued that humans are closer to chimps than they are to gorillas. Some pseudogenes are shared by humans and chimps, not by gorillas, while others are shared by humans and gorillas, but not chimps."

According to the current evolutionary view, humans, chimps and gorillas shared common ancestors for hundreds of millions of years from the origin of life until about 6 million years ago, when a separate gorilla lineage diverged. Therefore, the evolution model predicts that most human retroposons/pseudogenes would be shared between all three species since any retroposons/pseudogenes that arose in human ancestors before 6 million years ago (MYA) would have been passed on equally to those ancestors' human and ape descendants. This prediction is supported by current evidence that most human pseudogenes examined are shared by chimps and gorillas, and it is this evidence that argues strongly that these three species shared a common ancestor. If the data you cite in your second sentence above actually exist, they bear only on the question of whether chimp or gorilla ancestors first diverged from the human lineage, and not on the fundamental validity of evolution.

Current evidence suggests that the chimp lineage diverged from human ancestors about 5 MYA. Therefore, the evolution model would predict that human DNA should harbor some retroposons/pseudogenes not shared with chimps or gorillas, namely those that arose less than 5 MYA. This prediction is also supported by current evidence, in that several retroposons/pseudogenes whose sequence characteristics suggest that they arose recently are in fact not found in chimps and gorillas. Similarly, there are recent chimp retroposons/pseudogenes not found in humans or gorillas.

The interval of time between the divergence of gorillas from humans and the divergence of chimps from humans is believed to be about 1 million years, which is a tiny portion of the evolutionary history of these three species. Accordingly, evolution predicts that this period would have been too brief to have accumulated many retroposons/pseudogenes that we might now find shared by chimps and humans but not gorillas. Consistent with this prediction, I have not found any documented examples of such pseudogenes, although there are many examples of other more frequent types of genetic alteration (deletions, nucleotide substitutions, small insertions) which are shared by humans and chimps but not gorillas. [note added 4-26-00: An example of a sequence present in human and chimpanzee but not gorilla or other primate species is described in Keller et al., Mol Biol Evol 16:1019, 1999.]

What about a hypothetical pseudogene/retroposon shared by humans and gorillas but not chimps? As I argue in section 5.8 of my Web posting, although such a retroposon/pseudogene is not predicted by evolution, a few such examples can easily be accommodated with an evolutionary explanation: a pseudogene or retroposon that arose in the common ancestor of humans, chimps and gorillas may get deleted in chimps. Such deletions are not common, but are known to occur.

But suppose as DNA sequence data accumulates in the future we discover not a few, but many pseudogenes shared by humans and gorillas, and far fewer ones shared by humans and chimps; would this threaten the foundations of evolutionary theory as you imply? Absolutely not; it would, however, cause us to rethink the notion that gorilla ancestors diverged from the human lineage before chimp ancestors. The current view of an earlier gorilla divergence (and the estimated dates given above) derive from statistical calculations based on the very slightly greater similarity of DNA sequences between human and chimp than between human and gorilla; but only a small fraction of the DNA of these species has been sequenced, so these similarity estimates carry some uncertainty. Because most scientists feel that shared retroposons represent better clues to species relatedness than routine sequence similarity comparisons, a greater number of shared gorilla-human retroposons would probably compel the conclusion that the gorilla lineage diverged from the human more recently than the chimp lineage. The slightly greater chimp-human sequence similarity than gorilla-human would, if it persisted as more sequence data accumulate, suggest a greater mutation rate in the gorilla lineage than in human or chimp.

Alternatively, if future investigations discover just as many chimp-and-human-but-not-gorilla shared retroposons/pseudogenes as gorilla-and-human-but-not-chimp, this could be explained if chimp and gorilla ancestors diverged simultaneously (or very closely in time) from the human lineage. In this case the incompletely shared retroposons/pseudogenes could have inserted into the common ancestor such a short time before the lineage split that they had not spread throughout the population before the lineage split occurred; some retroposons/pseudogenes might become fixed in gorillas and humans but not chimps, and others might become fixed in chimps and humans but not gorillas.

In any case, I am surprised that you were able to find enough examples of gorilla-human-but-not-chimp pseudogenes to make the generalization in your second sentence quoted above. I would appreciate it if you would send me the literature references to all these examples. I can't help wondering if you are generalizing from the single example that creationist John Woodmorappe has cited, that of an epsilon immunoglobulin pseudogene supposedly shared by gorilla and human but not chimp. If you read my Web posting (the boxed sidebar under section 5.8), you will see that this example is erroneous.

2. You cite a creationist model of pseudogene formation (shown in your Fig. 1) in which genes A, B and C present in early populations of humans, chimps and gorillas lead to pseudogenes in modern populations of these species. If you think that this model accurately reflects the current observations of pseudogenes, you seriously misunderstand the details of the pseudogene data. Your model suggests that the pseudogene A' in humans arose independently from the pseudogene A' in chimps and in gorillas. The problem with your formulation is that many examples exist where not only do humans and chimpanzees both have a pseudogene derived from the same source gene, but they both have the SAME pseudogene; i.e. the pseudogene has the SAME defects and is located in the SAME location in the DNA of the two species, inserted into the SAME sequence at the target DNA site. There is no way that the random origin of retroposons/pseudogenes independently in two species, as described in your model, can explain identical defects and identical locations of the same pseudogene shared between species. On the contrary, your model predicts that random pseudogenes arising independently in chimps and humans would carry different mutations and, at least for processed pseudogenes, be located in different locations; both of these predictions are contradicted by the evidence.

3. Finally, you claim that "at least some of the 'pseudogenes may not properly be such at all, and may turn out to have a function like so many of their 'junkyard' cousins have." These arguments are also dealt with in my Web article (sections 5.1, 5.2 and 5.4). I suggest that you (and any other readers of this letter) consult my article. To briefly summarize these sections:

(a) It is true that a very few examples exist of retroposed genes that have no mutational defects and are functional; these are properly called "processed genes." They can easily be distinguished from processed pseudogenes, and do not suggest any function for those pseudogenes.

(b) A very few examples are known of retroposed sequences that have regulatory function. These represent examples of very rare favorable mutations, a class that creationists often claim cannot exist. We know that sequences are randomly being inserted into our DNA by retroposition even today, as genetic accidents without apparent divine intervention, including insertions that can cause inherited diseases in subsequent generations or induce cancer in the individual in which they arise. If these insertions have no beneficial function, there is no reason to suspect any function for the similar insertions that are shared between species.

(c) As scientists, we do not declare dogmatically that no function will ever be found for most pseudogenes. However, we base our current conclusions on currently available data. And those data indicate that shared retroposons/pseudogenes are functionless genetic accidents and can best be explained as originating in a common ancestor. Indeed, no other credible scientific explanation has been proposed either in the scientific literature or in creationist or mainstream scientific responses to my Web article.

Because my Web article already links to sites with opposing points of view (and includes my responses to those sites) I plan to ask the Webmaster of Talk.Origins to include a link to your site, and a link to a copy of this response to it. I would appreciate any comments you might have to this response, and am sure that Talk.Origins would be willing to include a link to your comments. If you have any interest in providing the readers of your site with an opposing point of view so that they can make up their own minds based on a comprehensive review of the data, you might want to consider linking to my site.

Sincerely,

Edward E. Max

[Note added later:
Shortly after receiving this letter, Carl Wieland responded cordially, agreeing that his page on Answers in Genesis might have been misleading. He stated that he would remove his page, and it promptly disappeared from the Answers in Genesis site. ]

[Return to Plagiarized Errors and Molecular Genetics]