Population genetics. (Q5566955)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 3283243
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Population genetics. |
scientific article; zbMATH DE number 3283243 |
Statements
1969
0 references
Hardy-Weinberg law
0 references
stable equilibrium of gene frequency
0 references
two-allele case
0 references
heterozygotes
0 references
homozygotes
0 references
mean fitness between two generations
0 references
multiallelic case
0 references
discrete random Markovian systems of Wright type
0 references
joint effects of epistasis and linkage
0 references
Fisher's theory of evolution of dominance by natural selection of modifiers
0 references
Population genetics. (English)
0 references
The author states in the preface to this little book that he has aimed it rather more at the mathematician interested in genetics, than the geneticist. Since the reviewer feels that he belongs to this former category, the following review is written from the same viewpoint, and may, hopefully, complement reviews in genetical journals. Generally speaking, the book reflects closely the author's association with the group of mathematical geneticists founded by \textit{P. A. P. Moran} at the Australian National University. Indeed, the influence of Moran's own book on the subject [Statistical processes of evolutionary theory. Oxford: Clarendon Press (1962; Zbl 0119.35901)] is clearly evident, incorporated within the general structure of the author's own particular past and present research interests, various aspects of which are more or less quantized into chapters. The first seven chapters are concerned with single locus theory, including the multiallelic case. Random effects are introduced in Chapter 4, the first three chapters on deterministic behaviour being a standard type of introduction, and dealing respectively with the Hardy-Weinberg law, the effects of selection and mutation, and the Fundamental Theorem of Natural Selection. In the first chapter it is pleasing to find a clear and separate statement of the important principle that random mating of individuals is equivalent to random union of gametes; the effect of selection on the Hardy-Weinberg law, discussed in Section 1.6 and a jumping-off point for the two following chapters, is also written with clarity. The second chapter quickly derives the necessary and sufficient condition for the existence of a stable equilibrium of gene frequency in \((0,1)\) in the two-allele case viz. that the heterozygotes have larger fitness than both homozygotes. In Chapter 3, while that part of the Fundamental Theorem pertaining to the increase in mean fitness between two generations, even in the multiallelic case, is easily established, the following interpretation (in the two allele case) of this increase as the ''additive'' part of the total variance in fitness, retains (at least for the reviewer) the usual obscurity. Indeed, the fact that the non-de-creasing nature of mean fitness in general appears to be confined to the single locus situation may lead other treatments not to place the same emphasis on the Theorem. On the other hand, the author's discussion of the difficulties which arise in attempting to extend the notions of the Theorem to the situation where fitness depends on two loci, in Chapter 8, is conscientious and lucid. Moreover, his introduction of ''interaction'' in order to simplify the problem in this situation is an interesting one, and in particular the specification of interaction to be ''additive'' does yield results analogous to the single locus situation (as indeed the author's more recent researches confirm). Even so, one is left a little uncertain whether such simplifying assumptions are of genetic relevance, or more a mathematical convenience to lead to results which are genetically desirable. Chapter 4 considers discrete random Markovian systems of Wright type (and their variants in diverse situations), concluding on the formulation of the generalized model involving selection and mutation. Having noted that it is in general difficult to study the relevant characteristics of such discrete systems analytically, the author lays the foundations for an approach by diffusion approximations in Chapter 5, which he applies in Chapter 6. (Re modestly states earlier, in the Preface, that his treatment of diffusion is heuristic and old-fashioned; in a book on population genetics such treatment is surely a virtue.) Of particular note in these chapters is the pleasing systematic and unified treatment of their subject, the introduction and use of the transient function, and a final section comparing certain aspects of the author's development with that of Fisher. The present reviewer found Chapter 9 on the joint effects of epistasis and linkage one of the first such treatments, the most stimulating in the book. While it is always necessary to bear in mind that an analysis on vastly simplified and fairly arbitrary two-locus models (in order to achieve mathematical tractability) is at best superficial, a number of rather exciting (as well as some expected) phenomena emerge. For example: ''.... linkage can cause a 'less fit' gene to replace steadily a 'more fit' one, at least when the former is rare.'' or ''... in a large proportion of cases the 'sudden increase in mean fitness' phenomenon can be expected.'' In particular, Section 9.6 on linkage and the survival of a new mutant yields a number of startling results. This last-mentioned section applies the theory of the multi-type Galton-Watson process, which, with the single-type process, in a single locus situation, is introduced earlier, in Chapter 7, in connection with the progress of mutants. Since the application of multi-type processes to genetic problems is of recent origin, the author's use of them is to be commended; on the other hand, the general reader would have benefited by a statement, at some stage, of the simple stochastic assumptions underlying such theory, and some comment whether these may reasonably be expected to hold whenever the author ''fits'' a process of this kind onto a deterministic model. A chapter on dominance gives an account of both Fisher's theory of evolution of dominance by natural selection of modifiers, with appropriate cautionary remarks, and the more recent theory of Kimura; and together with a valuable final chapter which summarizes succinctly some of the key conclusions, completes the book. There is also a brief glossary of genetic terms. Overall, the reviewer feels that in the particular area of the subject of its title with which it deals, the book is an important and timely contribution. No book is entirely free of defects and there are a few worth mentioning in the present instance also. For example, the referencing appears to be far more subjective than objective, in that there are a number of eminent omissions, at least as regards the statistico-genetic literature, which one would have thought more relevant than some given. Other examples are the larger than usual number of misprints, and occasional very loose statements such as ''Theorem'' 4.1. In addition, it might be worth mentioning that it is not difficult to supplement at least one comment of the author, on pp. 80--81. If the offspring distribution of a simple Galton-Watson process is Poisson, with mean \(m>1\)then not only are tables of the expected time to extinction conditional on extinction occurring available without resorting to diffusion theory, but also -- denoting the quantity by ET -- it is known that \(\text{ET} \sim -2 \log (m - 1)\) as \(m\to 1+\), and similar asymptotic expressions for higher moments are known also.
0 references