Plagiarized Errors and Molecular Genetics

A Response to "AFDave"

by Edward E. Max, M.D., Ph.D.

Copyright © 1998-2007

[Last Update: December 14, 2007]

"Misleading" is how AFDave has characterized my TalkOrigins essay in his comments on the Richard Dawkins Forum and on a May 28, 2007 post on his own Website [Dr. Edward Max on Vitamin C and Plagiarism]. In addition he makes two observations about my essay that, he claims, reflect a "fundamental error in logic." He marshals very little evidence in support of his claims, but relies on extensive quotations from a Daniel Criswell, who has a PhD in molecular biology and is a professor at the Institute for Creation Research graduate school.

I'd like to respond briefly to AFDave.

My TalkOrigins essay discusses how similarities in non-functional DNA sequences between different modern species (e.g. human and chimpanzee) strongly suggest that these sequences were copied from a common ancestor of the two modern species. Creationists argue that the Creator designed chimpanzees and humans to look and function similarly, so similarities would be expected in the DNA that controls body structure and function. But when we find human/chimp similarities in DNA features that reflect non-functional molecular accidents, it is hard to invoke an "intelligent designer" explanation, so inheritance from a common ancestor seems more likely.

As one example, my essay discusses the DNA sequence encoding the last enzyme in the biochemical pathway leading to vitamin C (ascorbic acid) synthesis. Most mammals make this enzyme (abbreviated GLO or GULO) and can therefore synthesize their own ascorbic acid, so they can live without any vitamin C in their diet. Most primates do not make GLO and therefore need to eat vitamin C. When scientists have examined the position of the human genome corresponding to the location of the functional GLO gene of rodents, they find that the human genome contains a recognizable GLO sequence that has been crippled by several mutations. Such crippled sequences that cannot encode a functional protein are called pseudogenes. Since creationists have no explanation why an intelligent Creator would insert a non-functional pseudogene into human DNA just at the position where a functional gene lies in other species, this pseudogene is most likely a relic of an earlier function GLO gene that suffered various mutations that inactivated it. Moreover, pseudogenes lying at the same position In chimpanzee, macaque and orangutan DNA all share a common crippling mutation with the human sequence; this is consistent with the idea that this mutation occurred in a common primate ancestor of humans and these three species.

In alleging my "fundamental error in logic," one point made by AFDave (also supported by Criswell) is that we have no proof that pseudogenes have no function; pseudogenes may have a function that we have not discovered yet. If so, they could have been designed independently in all four species to serve this unknown function. Readers of AFDave's post may have concluded that my essay overlooked this possibility, or that I intentionally omitted discussing it so as to "mislead" my readers. In fact, I discuss at length the possibility of function in pseudogenes (and other putatively non-functional quirks in DNA) in section 5 of my essay, which I invite readers to examine. I explain several reasons why scientists believe that most of these sequences can be assumed to be non-functional, including the fact that they can arise by random events in laboratory experiments without any obvious supernatural explanation and without conferring any apparent function, and that they can be deleted from the genome without causing apparent harm to individuals lacking them. And in my essay I bend over backwards to be fair by mentioning a few examples where some function has been suggested for certain unusual pseudogenes. AFDave ignores all the arguments I presented for non-function of most pseudogenes, so he avoids having to defend against them.

Criswell suggests that "maybe we did lose [the function of the human GLO gene], but as a result of the 'Fall'." The same argument has been made by other creationists, and I addressed it in section 5.7 of my essay:

This interpretation seems plausible, and--if we ignore the "Fall" part--not very different from the evolutionary idea that pseudogenes arise by random genetic accidents. However, this interpretation completely ignores the fact that many pseudogenes are shared between apes and humans, located in the same positions and sharing the same genetic defects, apparently the result of the same genetic accident or "degenerative change" in a common ancestor. (If these shared pseudogenes arose after the "Fall" as suggested by [Criswell], did the "Fall" perhaps occur before man diverged from the apes?)

Criswell mentions various organs that were previously incorrectly judged to be functionless evolutionary relics until later studies discovered that they do have functions. Criswell writes of pseudogenes: "Just because scientists do not currently know the function of a portion of DNA does not mean that it does not have any function and therefore it is an evolutionary leftover." I also addressed this point in my essay (section 5.2):

Imagine a defendant at a murder trial defending himself--against overwhelming incriminating evidence--with the parallel argument: that since some convicted criminals have later been exonerated, he (the current defendant) should therefore be acquitted now, because someday in the future, evidence might be found to clear him! This defense would be as ridiculous as Dr. [Criswell's] argument is. Scientists (and juries) must draw their conclusions based on the best evidence available at the time. It is true that later evidence may exonerate a convicted criminal or overturn a scientific theory. This possibility should foster humility, and caution us against dogmatic conclusions (and perhaps against the death penalty); but it should not dissuade us from drawing the most reasonable conclusions from the data at hand.

Given the evidence cited in my essay, the most reasonable conclusion at present is that most pseudogenes have no function.

Criswell goes on to say "It has been reported that pseudogenes play a regulatory role in yeast for the functional genes that they share sequence homology with (Hirotsune et al., 2003)." This is one of the few citations of evidence provided by AFDave and Criswell, and readers might be interested to examine it to judge the level of creationist scholarship and whether creationist treatment of evidence is valid or misleading. Check the paper cited by Criswell (Hirotsune et al. Nature 423:91, 2003):

1. This paper does not report about function for "genes" (plural); it reports function for a single pseudogene, Mkrn1-p1, namely that the RNA expressed from this sequence regulates expression of the functional Mkrn1 gene.
2. The pseudogene discussed is in mice not yeast.
3. The conclusions of the Hirotsune paper have been completely refuted by a more recent publication: "The putatively functional Mkrn1-p1 pseudogene is neither expressed nor imprinted, nor does it regulate its source gene in trans" Gray et al. PNAS 103:12039, 2006. I did not refer to this paper in my essay because the paper was not published in 2003 when I last revised my essay (I'm hoping for time to update it this year). But given that AFDave and Criswell were writing in 2007, one has to wonder whether they omitted mentioning this 2006 paper out of ignorance or to mislead their readers. Either way, the omission does not reflect well on creationist scholarship. The Hirotsune paper was cited by many creationists challenging the idea that shared pseudogenes imply a common ancestor, and the refutation of this paper undercuts the already meager support for their view.

Apart from AFDave's invocation of possible function for pseudogenes, he makes a second argument about my essay, alleging another "fundamental error in logic." My essay suggests that similar non-functional DNA "errors" (e.g. pseudogenes) shared between living species are analogous to textbook errors shared between different textbooks. Identical errors imply copying of some sort (plagiarism in the case of the textbook) as distinguished from independent origin. AFDave criticizes this analogy, saying:

THE "ERROR TEXT" IN THE GLO GENE IS NOT IDENTICAL, WHEREAS IT IS IDENTICAL IN THE CASE OF THE TEXTBOOK.

If I understand AFDave's point here, he is suggesting that though the GLO pseudogenes of man and chimp contain some of the same "errors" (mutations), the fact that other parts of the pseudogenes are not precisely identical invalidates my analogy to plagiarized textbook errors. Does he think that evidence of plagiarism would be invalidated if the plagiarized paragraphs contain a few typos not present in the original version? AFDave never explicitly says this, and never articulates an alternative explanation for the shared pseudogenes that he thinks would be favored (over copying from a common ancestral pseudogene) by the existence of slight differences between pseudogene sequences of chimp and human. It's not clear whether he thinks that these slight differences contradict the predictions of evolution or favor some alternative creationist model. In the absence of a clear explanation of AFDave's view, we can only speculate about the basis for his criticism.

It seems to me that "shared errors" (e.g. pseudogenes shared between human and ape) can be explained in three ways:

1. The pseudogenes are not really errors but were inserted independently for a purpose we just don't understand. (This possibility was discussed above.)
2. The pseudogenes arose independently and accidentally but are similar by chance.
3. The pseudogenes reflect copying from a common source.

AFDave's criticism could perhaps indicate that he favors the second explanation, as though his alternative explanation for nearly identical GLO pseudogenes is that highly similar but completely functionless DNA sequences stretching over hundreds of basepairs could have arisen by chance independently in the corresponding region of human and chimpanzee DNA. The probability of this occurring purely by chance depends on the length of the DNA sequence and the percent of matching sequence between the two species; but for almost all reasonable examples, the probability is vanishingly small, so AFDave has a major task to explain how this could happen. In contrast, the existence of minor differences between pseudogenes shared between species is exactly what would be predicted by the evolutionary model of copying from a common ancestral source. If the GLO gene was inactivated in a common ancestor of modern primates, then after the lineages diverged on the way to evolving into chimpanzees and humans, the GLO pseudogene would be expected to undergo further independent mutations in each lineage. These additional mutations would not be weeded out by natural selection because there is no selective pressure to preserve a DNA sequence that is already non-functional. The resulting GLO pseudogenes now in chimp and man would then contain some DNA changes that are unique to each individual species, while retaining most mutations that were present in the common ancestor of chimp and man. This picture of shared sequence plus additional unique mutations is exactly what we find in the GLO pseudogenes and most of the millions of other shared pseudogenes and retroposons I discuss in my essay.

Thus the criticisms of AFDave and Criswell, relying on hoped-for future evidence or on no clear logic at all, fail to weaken the conclusions of my essay.

[Return to Plagiarized Errors and Molecular Genetics]