InterJournal Complex Systems, 454
Status: Submitted
Manuscript Number: [454]
Submission Date: 1126
Comment on manuscript revision number 54706
Referee report on MS # 385
Author(s): Guy Hoelzer

Subject(s): CX.34, CX.35, CX.19

Category: Brief Article

Abstract:

I found this paper to be a generally clear description of the limitations of the gene centered approach to modeling evolution. The authors have also done an admirable job of revealing some of the implications of complexity in the context of biological evolution and adaptation. I strongly recommend accepting this paper for publication by InterJournal. I have several minor criticisms that the authors should consider if the paper is to be modified before final publication. p. 2, top: "One of the difficulties with this conventional view of evolution is that many organisms reproduce sexually and the offspring of an organism are thus often as different from the parent as other organisms that it is assumed to be competing against. The gene centered view [2] was introduced to address this fundamental paradox." This is not quite accurate. In a sexual system, an individual is more closely related to (i.e. similar to) its parents or full siblings than to other individuals in the population. Hamilton (not Dawkins) formalized this using a gene centered model to calculate the coefficient of relatedness (r). He did this as part of his effort to model the evolution of altruism, which was previously considered a major challenge to Darwinian theory. As the use of reference #2 suggests, Dawkins has taken this approach to an extreme, and he is often associated with it. Actually, I think it would be more appropriate to attribute the development of gene centered models to Fisher. He is most responsible for popularizing evolutionary models based on single loci, which were derived from the original Hardy Weinberg model. Ironically, Fisher was also responsible for creating the field of quantitative genetics, which is the traditional alternative to gene centered models. Fisher and many others have dealt with heritability and similarity among relatives in the context of quantitative genetics. The authors focus here on the issue of the unit (or target) of selection. Fisher modeled selection as if it acted on single genes, but I do not think he believed that nature worked this way. In short, I do not think it is accurate to suggest that such a paradox existed, or that the gene centered view was introduced for the purpose of addressing it. p. 2, top: "Correlations between genes arise when the presence of one allele in one place in the genome a •ects the probability of another allele appearing in another place in the genome." This phenomenon is called linkage disequilibrium. I understand that the authors aimed this paper at a non-biological audience, but I think that it would be improved if they include, at least parenthetically, the terms used by evolutionary biologists throughout the paper. p. 2, middle: "Correlations occur in selection when the probability of survival favors certain combinations of alleles,rather than being determined by a product of terms given by each allele separately." This is called epistasis. p. 2, bottom: "Simulations of models that illustrate trait divergence through symmetry breaking can be found elsewhere [3 ]." The authors can also cite Hoelzer (submitted, Interjournal MS # 345) here. p. 3, top: "Offspring are assumed to be selected from the ensemble which represents all possible combinations of the genomes from reproducing organisms." This is called panmixia. p. 3, top: "If we further simplify the model by assuming that each gene controls a particular phenomic trait..." The appropriate term here is "phenotypic", not "phenomic". p. 3, top: "As the presence of one allele in the population changes..." The word "presence" should be replaced by "frequency". p. 3, top: "As the presence of one allele in the population changes in the population..." Note the redundancy. p. 5, top: "This problem with fitness assignment would not be present if each allele separately coded for an organism trait." The usual way to avoid the "problem" described here is to use fitness as the trait of interest. This is not a trick, because anything we choose to describe as a trait is arbitrary with regard to nature. A trait is any measurable quality of an entity, which can certainly include survival and reproductive success. Aside from these relatively minor concerns, I think this is an excellent paper that introduces new ideas to the field of evolutionary biology, and introduces evolutionary biology to a broader audience. The modeling and logical argumentation are all sound.

Retrieve Manuscript Abstract
Submit referee report/comment


Public Comments: