Introductory Glimpses of Evolvability for Biologists

Posted October 20, 2017 by Matthew Andres Moreno in Research, Review / 0 Comments

This is one of a series of posts on evolvability. It is based off my undergraduate thesis, which I wrote at the University of Puget Sound under advisors Dr. America Chambers and Dr. Adam Smith. The original thesis is available here.

Introductory Glimpses of Evolvability for Biologists

The idea that phenotypic outcomes of mutation are non-arbitrary can be unfamiliar, or even uncomfortable, to biologists [Kirschner and Gerhart, 2005, p 219]. It is consensus among evolutionary biologists that genetic mutation is random. The alternative — the theory of adaptive mutation — is controversial and widely discredited [Sniegowski and Lenski, 1995]. It is therefore essential to note that discussions of evolvability are not predicated on adaptive mutation. The key difference is that adaptive mutation hypothesizes that genetic mutation is non-arbitrary while discussions of evolvability center on the idea that the phenotypic outcomes of mutation are non-arbitrary.

Even with this potential misunderstanding ironed out, the idea that phenotypic outcomes of mutation are non-arbitrary can still seem abstract or counterintuitive. Biology authors Kirschner and Gerhart explain that many biologists assume "phenotypic variation arises haphazardly from it [random mutation] as random damage; the organism’s current phenotype does not matter for the variation produced, and the output of variation is nearly random" [Kirschner and Gerhart, 2005, p 219]. Developing an intuition for evolvability requires moving towards a more sophisticated notion of the phenotypic consequences of mutation. This section will present a pair of biological examples of non-arbitrary outcomes under mutation, the first illustrating non-arbitrary outcomes in relation to phenotypic viability and the second illustrating non-arbitrary outcomes in relation to phenotypic novelty.

Artificial Selection for Asymmetry in Drosophila

Artificial selection experiments performed by Tuinstra et al. and Coyne nicely illustrate the non-arbitrary nature of phenotypic variation. In these experiments, performed with fruit flies, researchers were able to successfully select for bilaterally symmetric phenotypic criteria, such as overall smaller eyes, but were unable to successfully select for bilaterally asymmetric phenotypic traits, such as different-sized eyes. That is, selective breeding for bilaterally symmetric traits yielded observable phenotypic changes of the course of a number of generations while selective breeding for bilaterally asymmetric traits did not. In contrast, other artificial selection criteria, such as overall smaller eyes, yielded observable phenotypic changes over the course of a number of generations. A cartoon summarizing these results is provided in Figure 1.

image Figure 1 A cartoon summarizing the results of artificial selection experiments on Drosophila. Bilaterally symmetric eye traits could be successfully selected for while bilaterally asymmetric eye traits could not.

The success of artificial selection for bilaterally symmetric traits on Drosophila demonstrates the presence of heritable phenotypic variation for those traits. Likewise, the failure of artificial selection for bilaterally asymmetric traits on Drosophila suggests a scarcity of heritable phenotypic variation for those traits. Tuinstra et al. hypothesize that the very nature of the developmental process constrains the phenotypic variation that can be observed in offspring, in this case curtailing the abundance of offspring that lack bilateral symmetry. Specifically, they hypothesize that a lack of bilateral symmetry-breaking information during the embryological development of Drosophila explains the negative result of artificial selection for bilaterally asymmetric phenotypic traits. As Tuinstra et al. phrase it, “the developmental system does not seem to allow this type of variation.” In this way, the distribution of phenotypic diversity in offspring is biased away from asymmetric variation.

The results from these artificial selection experiments can be cast in terms of evolvability: in Drosophila, the distribution of phenotypic outcomes in under mutation is not entirely arbitrary. More specifically, it could be claimed that phenotypic outcomes under mutation are biased toward viability. For Drosophila, it would not be unreasonable to expect bilaterally symmetric flies to generally be more fit than bilaterally asymmetric flies. The likely more fit bilaterally symmetric phenotypic outcomes are exactly what we see favored by developmental constraint in Drosophila.

Heritable Variation for Body Size in Dogs

In addition to qualities that constrain against non-viable mutational outcomes, biological organisms can possess qualities that facilitate significant heritable variation for some phenotypic trait. The regulatory action of hormonal signals such as somatotropin exemplify such a quality. This compound, also known as growth hormone, is well known for its widespread anabolic effects on tissues throughout the body. Mutations affecting the regulatory pathways that regulate somatotropin production and release, the receptors and cell signaling components that mediate cellular response to somatotropin, and the protein itself all provide avenues for significant heritable variation in body size [Devesa et al., 2016].1 Dog breeds exhibit a range of body weights spanning nearly an entire order of magnitude. Clearly, heritable variation for canine body size is accessible. Indeed, among certain groups of dogs, much of this variation can be explained by just six genes, several of which are associated with pathways somatotropin participates in [Rimbault et al., 2013]. This observation can be cast in terms of evolvability. The presence of such hormonal signaling pathways can be viewed as making a broad range of heritable phenotypic variation more readily realizable via mutation.

Figure 2 A cartoon comparison of a Shetland Sheepdog and a Collie that illustrates the heritable variation in body size among dogs.


Coyne, J. A. (1987). Lack of response to selection for direction asymmetry in Drosophila melanogaster. Journal of Heredity, 78(119).

Devesa, J., Almenglo, C., and Devesa, P. (2016). Multiple Effects of Growth Hormone in the Body: Is it Really the Hormone for Growth? Clinical medicine insights. Endocrinology and diabetes, 9:47–71.

Kirschner, M. and Gerhart, J. (2005). The plausibility of life : resolving Darwin’s dilemma. Yale University Press.

Tuinstra, E., De Jong, G., and Scharloo, W. (1990). Lack of response to family selection for direction asymmetry in Drosophila melanogaster: left and right are not distinguished during development. Proc. R. Soc. Lond. B, 241(1301):146–152.

Rimbault, M., Beale, H. C., Schoenebeck, J. J., Hoopes, B. C., Allen, J. J., KilroyGlynn, P., Wayne, R. K., Sutter, N. B., and Ostrander, E. A. (2013). Derived variants at six genes explain nearly half of size reduction in dog breeds. Genome research, 23(12):1985–95.

Sniegowski, P. D., and R. E. Lenski. "Mutation and Adaptation: The Directed Mutation Controversy in Evolutionary Perspective." Annual Review of Ecology and Systematics, vol. 26, no. 1, 1995, pp. 553-578.

  1. Recent research implicates somatotropin in a number of processes unrelated to its classical association with metabolism and growth. Although the phenotypic consequences of mutations affecting somatotropin pathways are not exclusively limited to body size, somatotropin response nonetheless provides an avenue for heritable phenotypic variation in that regard.

Matthew Andres Moreno

I am a doctoral student at the Michigan State University Department of Computer Science and Engineering. My research interest is in using digital evolution techniques to investigate scientific questions about evolution. I am currently investigating the relationship between plasticity and evolvability as well as major evolutionary transitions.

More Posts - Website - Twitter

Leave a Reply