Genetic Recombination Misused
A Critique of Chris Ashcraft’s ‘A New Look at Genetic Recombination’
This is a critique of an article by Chris Ashcraft, published on the Creation Science Resource web site.
The text of the article to which this is a response can be found here:
A New Look At Genetic Recombination
I am providing this link because Ashcraft has revised his article on the Creation Science Resource website (he does not acknowledge that he has done so). The revised article contains almost as many technical and logical errors as the version to which this is a response. Certainly, the general thrust of the article remains the same in both versions and is equally unconvincing. I expect Ashcraft will continue to edit the article in an attempt to improve it. I might consider producing an updated version of this article, if at some time in the future, Ashcraft provides a substantive revision.
Ashcraft's revised article can be found here:
The five basic ideas of evolution according to Darwin (1) are:
These ideas are all accepted by the overwhelming bulk of scientists, with the possible exception of gradualism, which in its purest manifestation has been challenged by Gould’s and Eldrege’s concept of punctuated equilibrium. The other ideas now form a set that can justifiably be called the fact of evolution.
The generally accepted theory of the process by which evolution proceeds is based on the Modern Synthesis of Darwinism and genetics of the 1920s and 1930s (2), and today is called neo-Darwinism. This theory states that the fundamental mechanism for evolution is the creation of variability in the genome by mutational change and the selection via phenotypes of genotypes. When people refer to the Theory of Evolution, they usually mean neo-Darwinism.
Ashcraft’s article attempts to challenge this situation and is entitled ‘A New Look at Genetic Recombination’. The author claims that our understanding of homologous recombination has recently been revolutionised. He claims that recombination (as opposed to mutation) is the principal source of genetic variability, that the process is ‘intentionally’ controlled by the organism and therefore that the basis of Darwinian evolution is refuted. Is there any merit in this proposition? Has Ashcraft built an insightful argument using evidence from recent research and reached conclusions that have escaped the scientific community? He seems to think so, as he says: "The purpose of genetic recombination is to edit the genome in a largely uncharacterized manner, and therefore any alterations found to exist are probably the result of these reactions. This simple logic has escaped scientists who instead require mutations to be ultimately responsible for evolutionary adaptations." But the fact of the matter is that Ashcraft fails to produce any compelling new ideas and fails to convince us that his basic thesis has any merit.
At first sight, Ashcraft’s article seems to be a serious study of the subject, complete with multiple references that give the impression of scholarship. It is a pity that the content falls so far below the form. As we shall see, it fails by some significant margin to achieve the standards to which it aspires. On closer investigation, we find that it contains logical and technical errors. His call to the ‘Creation Science’ community to undertake research in the area that he has identified as important is unlikely to bear fruit, because there is no research being done by this community. The extent of their ‘science’ is to publish contentious interpretations of other people’s work in articles such as his. His list of references is very telling in this respect; the substantive references are all from what he would call ‘secular scientists’; the content of the references from the only creationist (Lane Lester) is simple error-strewn apologetics with no scientific merit whatsoever.
So what is wrong with Ashcraft’s article? There are basically three reasons why it fails to convince:
The basic proposition (that the existence of homologous recombination means that mutation is an unimportant mechanism in evolution) is flawed:
Ashcraft begins the article with this statement: ‘When Charles Darwin formulated his theories regarding evolution, he proposed that random mutations were responsible for varieties such as the finches on the Galapagos Islands before population geneticists determined that genetic recombination alone is responsible for this quantity of change’. Thus Ashcraft sets the stage for the demolition of the importance of mutations and the promotion of recombination as the principal source of genetic variation, with a fundamental historical error and an unsupported claim. It is hard to imagine a more unfortunate start. First, Darwin proposed nothing regarding random mutations for the simple reason that genetic mutation was unknown in Darwin’s time (3). Darwin clearly saw that there was variation from generation to generation but failed to find a mechanism for this variation. Second, Ashcraft provides no evidence for the notion that genetic recombination alone is responsible for the variations in Galapagos finches (4). Although hybridisation is found to be a stronger factor than mutation in the short term creation of genetic variation (5) in Galapagos finches, the rearrangement of alleles in new combinations must follow the generation of those alleles in the first place – a process for which genetic mutation is required. Indeed molecular data has been used to trace the ancestry and the evolutionary tree of the Galapagos finches.(6), (7).
Ashcraft’s basic thesis is this: ‘Homologous recombination is the source of the variation on which Natural Selection acts; mutation plays no role in evolution; since mutation is a cornerstone of Darwinism, Darwinism fails’. There are fatal flaws in this argument.
To begin with, the fact that homologous recombination occurs and has significance for evolution, does not in itself mean that mutations neither occur nor have significance. To refute Ashcraft’s fundamental argument, it is enough to show that mutations occur and create new alleles (in fact are the major source of new alleles). For example, it is known that starving bacteria evolve to enable them to metabolise substrates that are not available to wild-type strains. A considerable amount of work has been carried out on this subject. There is more than one known mutational mechanism that enables different strains of E coli to metabolise lactose and lactulose. These include a –1 base frame shift (a deletion) and the insertion of an Insertion Sequence (IS) (8) (9). And functional mutations clearly occur in mammals. The very recent publication of the mouse genome allows comparative genomics between mouse and man with fascinating results (10). The genomic history of mouse and man can be clearly traced in synteny, in chromosomal rearrangements, in repetitive sequences and in orthologous genes. All of these data clearly indicate the fundamental importance of mutation in changing the mouse and human genomes from their common ancestor to their current state.
At no point in his article does Ashcraft offer any evidence in support of his assertion that mutations cannot provide material for evolution. He makes statements such as: ‘The ability to induce homologous recombination during times of stress would also provide obvious survival advantages through the introduction of new genetic information. Mutations would be clearly disruptive in their affect [sic]’ and ‘Environmental adaptations are indeed the result of natural selection upon individuals with genetic differences, but it is extremely unlikely that random changes to a complex genetic code could produce any significant usefulness or evolutionary assistance to an organism’. This is the crux of his argument but it turns on unsupported assertions and is actually the old Creationist error that mutations cannot increase information or produce beneficial effects. It is in direct contradiction to the actual scientific evidence (8), (9), (10) .
The next fundamental error in Ashcraft’s logic is to set up the straw man of the randomness of mutations. There is a great deal of misunderstanding abroad about the meaning of random mutations. Ashcraft repeatedly claims that the fact of recombination denies the existence of mutations as a force in evolution (wrong as demonstrated above) and that this then leads to a conclusion that Darwinism is falsified. But Darwinism (or the fact of evolution) does not rely on mutation as the force for the creation of new phenotypes. Phenotypic variation in neo-Darwinism is based on the mechanism of random mutations where the ‘randomness’ is strictly limited to mean that mutations are not biased in a particular evolutionary direction. Since the existence of genetic change as a consequence of a wide range of mutations which occur at well understood rates can hardly be denied, since changes in genotype lead to changes in phenotype, and since Natural Selection acts inexorably on phenotypes, then mutation must be a source of genetic change that fuels evolution.
In addition to these major logical flaws, Ashcraft’s argument contains many other logical flaws, and some examples are listed below. Ashcraft says:
The following fallacies are also present: fallacy of exclusion (overwhelming evidence for the occurrence and importance of mutations is ignored); affirming the consequent (of the form: ‘if intelligent design causes cells to intentionally modify their DNA sequence, then we will observe modification in sequences over time - we do observe these modifications and therefore intelligent design did cause cells to intentionally modify their DNA sequence’); illicit minor (of the form: all recombination causes change in DNA sequences, and all recombination is regulated by cellular mechanisms, therefore all changes in DNA sequence are regulated by cellular mechanisms).
So, Ashcraft’s attempt to deny the importance of mutational change in evolution, being based entirely on his promotion of homologous recombination as the principal mechanism for change and in the absence of any direct data or evidence in support of it, fails on logical grounds.
The propositions that recombination is ‘intentional’ and entirely non-random and is the principal source of genetic change are not demonstrated:
The statement that recombination is ‘intentional’ is repeated several times (and he bases his conclusion of Intelligent Design on this). The term is never defined, but one is left with the bizarre impression that he considers the cell to be in some way conscious and directing its own modification. This is, of course, personification of a natural process, as gross as personifying the weather or proposing a homunculus in the head as the seat of consciousness. Now here is a slippery slope leading to myth and superstition. Perhaps Ashcraft means that recombination is a process which is directed to a particular end, or which is regulated in a complex way, or which has entirely predictable or deterministic outcomes. It is hard to tell. He certainly contrasts ‘intentional’ homologous recombination with ‘random’ mutations. Let us look at the evidence he puts forward for suggesting that the cell is carrying out ‘intentional’ modifications of its DNA code through the mechanism of homologous recombination.
First he makes the point that the mechanism for homologous recombination through all species including all eukaryotes and prokaryotes is conserved (11) and that that mechanism is complex (12). So it is; see also (13), (14), (15), (16) . But there is no suggestion that the mechanism for DNA maintenance and repair and the very precise copying and exchange of DNA in homologous recombination is anything other than complex. Complexity does not ‘intention’ make. We are back in the presence of Paley here.
Ashcraft next points out that recombination does not occur uniformly along chromosomes. Indeed positive interference in recombination is a very well known phenomenon and meiotic homologous recombination has recently been studied as part of the Human Genome project (17) where considerable variation along the chromosomes is found. But the obvious question is ‘so what’? If the idea here is to promote the non-random nature of homologous recombination in contrast with the ‘randomness’ of mutations, one should note that mutations also occur preferentially at some sites – indeed the term mutational hotspot is well accepted. (18) . It is indeed hard to point to any non-random characteristic of recombination that is not shared by mutation. And it is also a fact that recombination also occurs randomly in the sense that the number of meiotic crossovers and their position cannot be predicted for any single event.
Ashcraft’s contention that the fact that homologous recombination relies on a complex mechanism makes its outcome somehow ‘intended’ is ridiculous. How on earth is the cell to know in what direction to ‘intentionally’ modify in order to become fitter in a particular environment? What are we to make of a hypothesis that does away with the need for selection? If Natural Selection has a role then it acts on all phenotypic change whether caused by mutation or recombination
Actually, many of the products of recombination have a deleterious effect and are regarded as mutations (homologous recombination is a process that can result in mutation – the terms are not mutually exclusive). Take, for example, mitotic crossing over, a process which is less frequent than meiotic crossing over, but which does occur. It can lead to loss of heterozygosity, which in turn can cause the daughter cells to develop as cancers or for other recessive undesirable traits to be expressed. Mitotic gene conversion can have the same effect. Recombination between pseudogenes or other repetitive sequences and coding regions can cause genomic instability (as well as being a mechanism for development of immune response). Deletions (which are absolutely regarded as mutations) can be caused by homologous recombination as can tandem duplications (19) . Turning from eukaryotes to prokaryotes, several of the homologous recombination events would result in what we would recognise as mutations. See below.
Ashcraft also makes much of gene conversion and of horizontal gene transfer. His claim is that these processes are a major source of novel genetic material. Let us look at them in a little more detail.
Gene conversion is the non-reciprocal transfer of genetic material, for example from a pseudogene to an active gene in eukaryotes in somatic cells during mitosis or as a non-reciprocal event in meiosis instead of crossing-over. It is certain that gene conversion from pseudogenes exists. However, Ashcraft fails to recognise the origin of pseudogenes, which are created by gene duplication and the subsequent silencing of copies of the gene by further mutation such as the insertion of additional stop codons. So the source of material for gene transfers (pseudogenes) is mutation. Secondly, Ashcraft fails to acknowledge the importance of the fact that change in the V-region of B-cells occurs via gene conversion in some vertebrate species (eg chickens, rabbit, cows and pigs use gene conversion) whereas somatic hypermutation is used in others (sharks, frogs, mice and humans) (20). All species seem to be competent for both processes but in some species one or the other process predominates. The key point is that although gene conversion is important in some species for mutating the V region of B cells, no-one suggests that it is a key process for producing genetic variability in general evolution (ie variation which affects the germ line).
Turning to horizontal gene transfer, Ashcraft claims that all species are competent at accepting foreign genetic material in vivo. He does not acknowledge that horizontal gene transfer is largely a process limited to Archaea and Bacteria in nature and that, although genetic material can be introduced by a gene gun, blasting DNA coated on gold particles at a cell at supersonic speeds is hardly a natural phenomenon. He fails to produce evidence that any comparable process occurs naturally in eukaryotes. Horizontal gene transfer in bacteria occurs in a number of ways:
a) Transformation, the take up of short sequences of DNA present in the environment from dead bacteria
b) Transduction – the infection of a bacterium by a bacteriophage and exchange of DNA with the bacteriophage
c) Conjugation – exchange of genetic material either through episomal conjugation, high frequency recombinant conjugation or resistance plasmid conjugation.
Of course, none of these processes leads us to conclude that mutation is not important. They are the homologue of meiotic recombination in eukaryotes, providing instant and short term genetic variation. A study of lateral transfer into higher eukaryotes has concluded that the evidence is lacking (21) for lateral transfer between bacteria and humans and there is no evidence for lateral gene transfer within eukaryotes.
So, Ashcraft fails to convince us that homologous recombination is an ‘intended’ process, that it is entirely deterministic or non-random and that it is the principal source for genetic variation.
The article contains significant technical errors and confusion in many areas
There are significant technical errors within the article. Ashcraft’s tendency to lump all homologous recombination together and to fail to distinguish between, for example, meiotic recombination and recombination in prokaryotes causes confusion. We have already seen that he begins the article with a very basic error. Where else does he go wrong?
1) He says: ‘Ultimately Darwinian evolution is an attempt to explain the unintentional development of life on earth, and requires a random mechanism for generating new genetic information’. Ashcraft displays his prejudice. Darwinian evolution is intended to explain the diversity of species, from a scientific, ie natural, viewpoint. The mechanism for creating phenotypic variation is an outcome of the theory not a prerequisite of it. Darwin struggled to find a mechanism for introducing variation. Genetic mutation was certainly not one that he considered and a ‘random’ mechanism is certainly not a prerequisite.
2) The entire introduction to the section ‘Review of Genetic Recombination’ confuses and conflates meiotic recombination with other types of homologous recombination.
3) He says: ‘The entire nuclear genome condenses during meiosis into organized units called chromosomes that interact as distinct pairs. These paired homologous chromosomes possess sequences that are very similar, and code for variations of the same characteristic like sex determination’. However sex determination is not coded in autosomes but depends on whether a particular sex-chromosome is present (in mammals, animals with a Y-chromosome contain the sex determining gene SRY and are male and those without are female) and the Y chromosomes are the exception to the rule in that they do not recombine with X chromosomes in males during meiosis.
4) He says: ‘Offspring variations are largely produced by the intentional recombination of alleles during meiosis, and it was the regulation of this process that was demonstrated by Mendel's experiments’. Ashcraft clearly fails to understand the importance of the random disjunction of chromosomes (independent assortment) during meiosis in producing offspring variation, and it is this rather than recombination which is demonstrated by Mendel’s experiments and Mendel’s laws
5) In Figure 5, Ashcraft suggests that intragenic recombination is a routine source of new alleles and genetic variation. But Mark Ridley points out that (22) intragenic recombination has the same effect on the gene as no recombination except in those cases where the recombination occurs between the site of interest and another heterozygous site within the gene, and this latter condition happens too infrequently to affect the identity of genes on the time scale of natural selection.
6) He says: ‘The mitochondria were previously thought to offer evidence that new alleles form through mutation alone because it was believed the organelle's genome was obtained exclusively from maternal contributions. However, it has been recently reported in the journal Science that recombination between parental genomes also occurs in mitochondrial DNA. Evidence of mixing of paternal with maternal DNA in the mitochondria was evident, and it was concluded there had been recombination between the parental genomes’. He fails to mention the fact that several papers have challenged this work (23) and that the accepted situation is still that change in the mitochondrial DNA occurs through mutation. Furthermore, he entirely ignores the fact that the Y-chromosome is another DNA sequence which in mammals does not recombine but which still displays mutational change over time.
7) ‘Without a doubt the immunity system provides the best example of the ability of the cell to generation [sic] new genetic information through homologous recombination’. The problem with this statement is that it has little relevance to evolution since the process is limited to somatic mutations that have little or no influence on the germ line. Furthermore, as we have seen, homologous recombination is not the only process by which the immune response proceeds.
8) ‘There are many examples of viruses such as Polio and Small Pox that have been completely eliminated from the face of the planet because it is unquestionable that functional antibodies will be assembled following an exposure to almost any foreign substance’. This is an error because although some vaccinations are near 100% efficacious against certain antigens, other vaccinations are less than 100% efficacious. Efficacy or immunogenicity as measured by the assessment of seropositive reactions is generally less than 100%. Completely efficacious vaccines are the exception rather than the rule.
9) ‘The purpose of genetic recombination is to edit the genome’ To the extent that genetic recombination has a ‘purpose’, it is to repair DSBs, to maintain mitochondrial homoplasmy and to create physical chiasmata in meiosis to avoid non-disjunction leading to aneuploidy.
10) ‘The undisputed function of these reactions during meiosis is to recombine parental DNA, and thereby produce offspring variability. Subsequent selection upon these recombinants has led to the variety of domestic breeds commonly used by humans today’. The most casual investigation with a search engine will reveal the fact that animal and plant breeders are well aware that mutations occur in organisms and that many differences between breeds rely on the existence of these mutations or sports. Many breeds or strains display a characteristic that occurred out of the blue and is clearly not present (even recessively) in the gene pool of its parents. Where these rely on a single nucleotide polymorphism, it is impossible that they could have arisen in any way other than through mutation of a single base pair.
11) ‘Although it is clear that many organisms possess and are utilizing foreign or altered DNA, the participation of the cell in these processes is unequivocal’. This is simply gobbledygook – what does it mean?
Ashcraft seeks to persuade us that recent scientific research challenges neo-Darwinism and he fails to achieve this aim. What he offers is logical fallacy and wishful thinking in an unusually (for a creationist) glossy frame. He is to be congratulated on rising well above the general run of creationist apologetics and deserves sympathy for falling so far below a serious challenge to the Theory of Evolution.
1 Mayr, What Evolution Is, p86, Weidenfeld and Nicolson, 2001
2 Gould, The Structure of Evolutionary Theory, 70 – 71, 503 – 591,
Harvard University Press, 2002
3 Darwin, On the Origin of Species
4 Unpredictable Evolution in a 30-Year Study of Darwin's Finches,
Peter R. Grant and B. Rosemary Grant Science 296, 707-711
5 Phenotypic and genetic effects of hybridization in Darwin's
finches, Grant, P.R., and Grant, B.R.. Evolution 48, 297-316.
6 Patterns of Evolution in Darwin's Finches: Microsatellites Provide
a New Perspective, Kenneth Petren, B. Rosemary Grant and Peter R.
Grant, from the Proceedings of the American Association for the
Advancement of Science, Pacific Division, Volume 18, Part 1, June 19,
7 The Beak of the Finch: A Story of Evolution in Our Time, Jonathan
Weiner, Vintage Books, 1995
8 Cairns et al, The origin of mutants. Nature 335:142–145
9 Hall, Adaptive evolution that requires multiple spontaneous
Mutations involving an insertion sequence. Genetics
10 Initial sequencing and comparative analysis of the mouse genome,
Mouse Genome Sequencing Consortium, Nature 420, 520 - 562
11 Homologous genetic recombination as an intrinsic dynamic property
of a DNA structure induced by RecA/Rad51-family proteins: a possible
advantage of DNA over RNA as genomic material. Shibata, T.,
Nishinaka, T., Mikawa, T., Aihara, H., Kurumizaka, H., Yokoyama, S. &
Ito, Y. Proc. Natl. Acad. Sci. U.S.A. 98(15) 8425-8432 (2001)
12 Hierarchic Regulation of Recombination. Kunihiro Ohta. RIKEN
Review 41:28-29 (2001)
13 Mammalian XRCC2 promotes the repair of DNA double strand breaks
homologous recombination, Johnson et al, Nature 401, 392 - 399
14 Nbs1 is essential for DNSA repair by homologous recombination in
higher vertebrate cells, Tauchi et al, Nature 420, 93 - 98
15 Binding of double strand breaks in DNA by human Rad52 protein,
Van Dyck et al, Nature 398, 728 - 731
16 The Rad50 zinc-hook is a structure joining MRE11 complexes in DNA
combination and repair, Hopfner et al, Nature 418, 562 - 566
17 Comparison of human genetic and sequence-based physical maps, Yu
et al, Nature 409, 951 - 953
18 Preferential Formation of Benzo[a]pyrene Adducts at Lung Cancer
Mutational Hotspots in P53 Mikhail F. Denissenko, Annie Pao, Moon-
shong Tang, and Gerd P. Pfeifer, Science 274, 430-432
19 Homologous recombination as a mechanism for genome
rearrangements: environmental and genetic effects, Bishop and
Schiestl, Hum Mol Gen 9, 2427 - 2434
20 Ablation of XRCC2/3 transforms immunoglobulin V gene conversion
into somatic hypermutation, Sale et al, Nature 412, 921 - 926
21 Microbial Genes in the Human Genome: Lateral Transfer or Gene
Loss? Salzberg et al, Science 292, 1903
22 Evolution, Mark Ridley, Blackwell Science, 1996
23 Kivisild and Villems; Jorde and Bamshad; Kumar et al; Parsons and
Irwin; all separate papers in Science 288, starting p 1931
[Revised 21st February 2003 to include a link to
the text of the original Ashcraft article since he has substantially revised
it on the Creation Science Resource website and to make some other minor changes]