Skip navigation

About IDEA Center

News & Events



IDEA Student Clubs


Contact Us



Convergent Genetic Evolution: "Surprising" Under Unguided Evolution, Expected Under Intelligent Design

Casey Luskin

A 2010 article in Trends in Genetics, "Causes and evolutionary significance of genetic convergence," addresses the apparently "convergent" appearance of genes or gene sequences and how unguided evolution can explain this. The paper defines convergence as the "independent appearance of the same trait in different lineages." Thus, genetic convergence is the independent appearance of the same genetic trait in different lineages. The article starts by explaining how widespread convergent evolution is: The recent wide use of genetic and/or phylogenetic approaches has uncovered diverse examples of repeated evolution of adaptive traits including the multiple appearances of eyes, echolocation in bats and dolphins, pigmentation modifications in vertebrates, mimicry in butterflies for mutualistic interactions, convergence of some flower traits in plants, and multiple independent evolution of particular protein properties.

(Pascal-Antoine Christin, Daniel M. Weinreich, and Guillaume Besnard, "Causes and evolutionary significance of genetic convergence," Trends in Genetics, Vol.26(9):400-405 (2010) (internal citations omitted).)
But what causes these similar traits to appear in widely diverse organisms? It turns out that "convergent" phenotypic similarity is often based upon "convergent" genetic similarity: "studies have traced phenotypic convergence to modifications of homologous genes; in this paper such phenomena will be further referred to as convergent recruitment" (emphasis added; internal citations omitted). Or, as the abstract states:Asccumulating studies on this topic have reported surprising cases of convergent evolution at the molecular level, ranging from gene families being recurrently recruited to identical amino acid replacements in distant lineages. (emphasis added)In other words, as the paper explains, convergent phenotypic traits occur due to convergent genetic evolution, which supposedly "results from a strongly biased potential for a given phenotypic change as a consequence of mutations in different genes."

Neo-Darwinian evolution isn't supposed to be goal-directed, but some force is causing the same sequences--at the genetic level--to appear independently over and over again. In an undesigned world, this is extremely unlikely.

Though the authors of course do not advocate any sort of purpose behind evolution, their paper's teleological language about the "potential" or "predisposition" for beneficial evolutionary change is striking (all emphases added):
  • "Convergent recruitment of the same gene lineage from multigene families affords an ideal system for studying the predisposition of particular genes for a given novel function."
  • "We predict that, compared to the other members of the gene family, a recurrently recruited gene lineage will generally have a catalytic activity and an expression pattern closer to those needed for the novel reaction compared to other members of the same gene family.
  • convergent recruitment suggest [sic] that only a few genes have the potential to create a specific phenotypic change"
  • In keeping with this teleological theme, the paper even alludes to the notion of pre-adaptation: [G]enes must meet two broad criteria to be eligible for a novel function: (i) they must have the possibility of being recruited for a new task without deleterious effect due to the loss or modification of the ancestral function, and (ii) their expression profiles and kinetics must be suitable for the new task.The paper explains that [r]egarding the second criterion, the pool of candidates for a new function is likely to be limited to genes encoding enzymes with compatible catalytic properties or to genes that can acquire them via successive substitutions without strongly deleterious transitional stages . . . Transitions to proteins with better suited kinetics but which involve transitional stages with lower fitness are less likely to take place, and evolutionary paths that lead to optimized enzymes via successively advantageous single nucleotide substitutions will be more frequently followed.If functional genetic pathways to new useful phenotypes are so fraught with peril, then the authors are faced with the reality that "[c]onvergent recruitment indicates that genes suitable for creating a given phenotype are rare." Or, as the article's abstract states: "molecular evolution is in some cases strongly constrained by a combination of limited genetic material suitable for new functions and a restricted number of substitutions that can confer specific enzymatic properties."

    The view that there are severely limited functional phenotypes is echoed by George R. McGhee in his book The Geometry of Evolution: Adaptive Landscapes and Theoretical Morphospaces: "convergent evolution means that there are a limited number of ways of making a living in nature, a limited number of ways of functioning well in any particular environment." (p. 34)

    We're now left with a striking situation. Apparently "rare" genes that are "suitable" can only undergo a "restricted number of substitutions" in order to find one of a "limited number of ways of functioning." If neo-Darwinism has difficulty finding functional gene sequences once, how is it stumbling upon the same gene sequences over and over again?

    Is there a better explanation for why similar gene sequences appear over and over again in different organisms, even when common ancestry cannot suffice as the explanation? What known cause can do this? Stephen C. Meyer suggests one:Agents can arrange matter with distant goals in mind. In their use of language, they routinely 'find' highly isolated and improbable functional sequences amid vast spaces of combinatorial possibilities." (Stephen C. Meyer, "The Cambrian Information Explosion," in Debating Design, p. 388 (William A. Dembski and Michael W. Ruse eds., Cambridge University Press, 2004).)Convergent evolution implies that these rare functional sequences are discovered not just once, but repeatedly. Perhaps a better explanation for the existence of rare functional sequences is intelligent design. Paul Nelson and Jonathan Wells explain why intelligent design is a viable explanation for the repeated appearance of similar gene sequences:An intelligent cause may reuse or redeploy the same module in different systems, without there necessarily being any material or physical connection between those systems. Even more simply, intelligent causes can generate identical patterns independently." (Paul Nelson and Jonathan Wells, "Homology in Biology," in Darwinism, Design, and Public Education, pg. 316) Might convergent genetic evolution actually be a pointer to intelligent design?

    Implications of Genetic Convergent Evolution for Common Descent
    The aforemtnioned paper in Trends in Genetics, "Causes and evolutionary significance of genetic convergence," notes that that genetic convergence is not uncommon, even though only a "restricted number of substitutions" at the genetic level can create novel phenotypic traits. This data not only shows that functional genotypes are rare, but it also poses a much deeper problem for evolutionary thinking--one that challenges the very basis for constructing phylogenetic trees.

    The main assumption behind evolutionary trees is that functional genetic similarity implies inheritance from a common ancestor. But "convergent" genetic evolution shows that there are many instances where functional similarity is not the result of inheritance from a common ancestor. So when we find functional genetic similarity, are we to assume that it represents a homologous DNA sequence, or a convergently similarity sequence? This poses great difficulties for those who wish to build evolutionary trees under the assumption of common descent.

    I am hardly the first to recognize this. Evolutionary paleoecologist Simon Conway Morris--who is not an intelligent design (ID) proponent--quite explicitly observes that convergence poses a major difficulty for the construction of phylogenetic trees:I believe the topic of convergence is important for two main reasons. One is widely acknowledged, if as often subject to procrustean procedures of accommodation. It concerns phylogeny, with the obvious circularity of two questions: do we trust our phylogeny and thereby define convergence (which everyone does), or do we trust our characters to be convergent (for whatever reason) and define our phylogeny? As phylogeny depends on characters, the two questions are inseparable. ... Even so, no phylogeny is free of its convergences, and it is often the case that a biologist believes a phylogeny because in his or her view certain convergences would be too incredible to be true. ...

    During my time in the libraries I have been particularly struck by the adjectives that accompany descriptions of evolutionary convergence. Words like, 'remarkable', 'striking', 'extraordinary', or even 'astonishing' and 'uncanny' are common place...the frequency of adjectival surprise associated with descriptions of convergence suggests there is almost a feeling of unease in these similarities. Indeed, I strongly suspect that some of these biologists sense the ghost of teleology looking over their shoulders.

    (Simon Conway Morris, Life's Solution: Inevitable Humans in a Lonely Universe, pp. 127-128 (Cambridge University Press, 2003).)
    Another good example comes from a recent treatise published by Harvard University Press which states that convergent evolution causes "difficulties" for building trees:Cladistics can run into difficulties in its application because not all character states are necessarily homologous. Certain resemblances are convergent -- that is, the result of independent evolution. We cannot always detect these convergences immediately, and their presence may contradict other similarities, "true homologies" yet to be recognized. Thus, we are obliged to assume at first that, for each character, similar states are homologous, despite knowing that there may be convergence among them.

    (Guillaume Lecointre & Hervé Le Guyader, The Tree of Life: A Phylogenetic Classification, p. 16 (Harvard University Press, 2006).)
    Likewise, a paper in Annual Review of Ecology and Systematics explains the "difficulties" facing molecular phylogenies as a result of convergence:Given the difficulties associated with alignment and with the establishing the conditions of consistency and convergence, it is clear that molecular phylogenies should not be accepted uncritically as accurate representations of the degree of relatedness between organisms.

    (Rudolph Raff, Charles R. Marshall, and James M. Turbeville, "Using DNA sequences to unravel the Cambrian radiation of the animal phyla," Annual Review of Ecology and Systematics, Vol. 25:351-375 (1994).)
    Thus, the textbook Explore Evolution says the following:Convergence is a deeply intriguing mystery, given how complex some of the structures are. Some scientists are skeptical that an undirected process like natural selection and mutation would have stumbled upon the same complex structure many different times. Advocates of neo-Darwinism, on the other hand, think convergent structures simply show that natural selection can produce functional innovations more than once. For other scientists, the phenomenon of convergence raises doubts about how significant homology really is as evidence for Common Descent. Convergence, by definition, affirms that similar structures do not necessarily point to common ancestry. Even neo-Darwinists acknowledge this. But if similar features can point to having a common ancestor--and to not having a common ancestor--how much does "homology" really tell us about the history of life? (p. 48) When building evolutionary trees, evolutionists assume that functional genetic similarity is the result of inheritance from a common ancestor. Except for when it isn't. And when the data doesn't fit their assumptions, evolutionists explain it away as the result of "convergence."

    Using this methodology, one can explain virtually any dataset. Is there a way to falsify common descent, even in the face of convergent genetic similarity? If convergent genetic evolution is common, how does one know if their tree is based upon homologous sequences or convergent ones? Critics like me see the logic underlying evolutionary trees to be methodologically inconsistent, unpersuasive, and ultimately arbitrary.