As the previous chapter indicated, a gene is much more, and more subtle, than simply a linear coding sequence. That sequence works only in the context of surrounding regulatory sequences and in the presence of the appropriate regulatory environment in the nucleus. Until recently, our knowledge of genetics was too rudimentary to enable us to make much headway in understanding how complex phenotypes result from genotypes. Darwin considered the problem essentially insoluble.

Phenotypes are also more complicated than can be explained simply by the linear coding sequence of a gene. Proteins have secondary and tertiary structures upon which their physiological functions depend. Even single proteins are only partial phenotypes, because their concentrations, induction, and tissue-specific environments and expression are also relevant to their action. Variation in regulatory and many other genes affects how the protein will function, just as variation in the sequence of the protein itself does.

An understanding of the relationships between genes and phenotypes requires an understanding of the logic by which life operates and has evolved. This is important because mutations can interrupt function at any point, and the pattern of pathology of a given genetic disease is often a direct reflection of where in the system the disruption occurs.

Offensive and defensive strategies against evolving threats

Infectious and parasitic diseases have been paramount among the threats to human health and survival for most of our (and our mammalian ancestors') evolutionary history. These diseases present three particular evolutionary problems. First is the essentially open-ended variety of pathogens to which we might be exposed. Secondly, while being able to attack nearly any kind of microorganism, we must be inhibited from attacking our own, very diverse, cells. Thirdly, we reproduce and can evolve genetic adaptations only slowly, but the pathogens that infect us typically reproduce in the millions, in days or even hours. These seem like evolutionary battles we could hardly win.

One way the immune system recognizes foreign pathogens is by their molecular structure; but in contrast to molecular recognition systems such as those between hormones and their receptors, which can be specifically programmed, the immune system has evolved to use a mechanism that can recognize an unlimited diversity of molecules {Roitt et al., 1989}. That mechanism forces us to revise the standard assumption that the inherited genome is constant during the life of an individual.

The previous chapter dealt with the genetic basis of categorical phenotypes such as the presence or absence of a disease. But how do we assess the genetics of quantitative phenotypes that cannot be broken down into distinct categories? Such traits may be affected by a large number of loci acting together, as well as by environmental factors. Important diseaserelated traits such as blood pressure, obesity measures, or cholesterol and triglyceride levels are examples. For such traits, what we need to understand is the effect of the genotypes, and the environment, on the phenotype.

This chapter introduces many important concepts and models for quantitative traits {Crow and Kimura, 1970; Falconer, 1989; Hartl and Clark, 1989; Hedrick, 1985}. The material is rather dense and relentless, but the concepts are important to understand, as they enable us to look further at the genetic causal spectrum, and to decompose it into its constituent parts.

Einstein's trains: assigning phenotypic effects to genotypes

Allelic and genotypic values

Albert Einstein made himself famous by looking out of a train window and speculating as to why he could not tell whether his train was moving north or the train on the next track was moving south.

The usual idea of a gene is of a specific region of DNA that codes for a single protein or enzyme, and the position of a gene on a chromosome is known as its locus. Variants of the DNA sequence at this locus among individuals are known as alleles. This ‘one gene, one enzyme’ model has long been the basis for research by human geneticists trying to identify traits, or phenotypes, whose inheritance patterns are consistent with the action of individual genes.

However, recent advances in genetics have greatly revised our concept of what genes are and how they work, showing that the relationship between DNA sequence and phenotype is both more complex and more interesting than we had thought. Some functions of DNA do not even depend on its nucleotide sequence, and DNA sequence variation includes a variety of direct and indirect forms of feedback among various regions of the DNA within and between cells.

Human life begins with a fertilized egg that carries a set of chromosomes containing genetic instructions, plus enough basic compounds and nutrients to commence the cell cycle. This latter material includes messenger RNA (mRNA) coded for by the mother's genes, which provides enough ‘information’ to direct the development of the new individual until its own genetic mechanism can be switched on. Thereafter, the organism depends entirely on its own genes, which contain the coding sequences and all the other signals needed for controlling embryological development, and the subsequent growth, tissue renewal, and physiology of the organism during its lifetime.

I chose Vesalius' classic figure of the human circulation for the Frontispiece of this book, because although it itself is not about genetics it is an appropriate metaphor for the genetic architecture that is typical of so many traits. The circulation is a redundant, anastamotic system that provides many vascular pathways to deliver blood to the same tissue, in a hierarchically structured way that has a few major arteries grading into many smaller arterioles and capillaries.

The spectrum of causation and the gradation of phenotypes

Biologists, perhaps like all people, seem to be divided into those who think qualitatively and those who think quantitatively, but the distinction is one of degrees rather than kind. In many instances phenotypes can be divided into categories, such as the presence or absence of a disease, for which associations with different genotypes can be documented generally by case-control studies or more precisely from mendelian patterns in families. For practical purposes we can say that some genotypes ‘cause’ a particular phenotype.

From the beginning of evolutionary population biology, studies of genetic disease have had an important place in the attempt to understand the relationship between the genes we inherit and the traits, or phenotypes, that we manifest during life. Diseases attract our deep personal interest. What causes them? Can they be inherited? Will I be affected? The very existence of genes that lead to traits that are deleterious to the individual also poses an apparent challenge to the idea of adaptive evolution by natural selection: how can such genes give rise to anything other than trivial frequencies?

This is an exciting time to be working in human genetics. A host of new statistical and molecular methods have become available in recent years; they enable us systematically to answer fundamental questions that previously we could only speculate about. Major new successes appear in the popular and scientific press every week, as genes related to one disease after another are identified. This has captured public imagination and its hopes. What is this all about? What does it tell us about basic biology?

In fact, the application of new methods to problems in genetic disease has, along with other developments in biology, led to a number of new generalizations about the genetic control of biological phenotypes that were unpredicted by classical evolutionary genetic theory, and which lead us to a revised view of genetic variation, how it works, and of the relationship between genotype and phenotype.

We have seen how to infer that genes have various effects on a trait, but how does the spectrum of genetic effects arise, and how is the variation distributed over space and time? This chapter provides a brief travelog of some of the relevant concepts of population genetics, the theoretical basis of evolutionary biology {Crow, 1986; Hartl and Clark, 1989; Hedrick, 1985; Li and Graur, 1990; Nei, 1987}. These concepts are used in subsequent chapters.

Life is fundamentally stochastic: the fate of a new mutation

(Nearly) each new mutation is unique

For most of this century it was thought that only a few alleles could exist at a locus, of which new copies arose via recurrent mutation at some rate, generally estimated to be about 10-5 per locus per generation, providing a continual supply of a given allele. This view was based on phenotype data, but at the sequence level the probability that the same mutation will recur is very small. Think of a cistron (coding stretch) of, say, a thousand nucleotides. Three thousand different non-synonymous (nucleotidechanging) point mutations are possible.

The ‘Rusty Rule’

The previous chapter focused on the evolutionary determinants of the frequency of disease-related alleles, and the trail and structure left by the unique history of mutations in each population. Evolution systematically generates variation, but is the amount of variation so great as to change our traditional notions that there is ‘a’ locus, or ‘a’ mutation that is responsible for ‘a’ disease in the majority of cases?

This chapter uses examples to characterize the level of heterogeneity that led me to introduce (Chapter 2) the ‘Rusty Rule’ that whatever can go wrong will go wrong – in some family, at some time. In fact, the amount of heterogeneity associated with most traits can lead to great difficulty in applying the segregation and linkage methods of Part II, a problem we are just beginning to face.

Etiological and phenotypic heterogeneity for qualitative traits

A number of the classic genetic diseases have now been studied at the DNA level. In each case, a similar story is told, one that is consistent with evolutionary genetics. It is the story of many different alleles, or even loci, associated with the same phenotype, or of what seemed originally to be a unitary phenotype decomposing before our eyes as subtle phenotypic differences associated with identifiable allelic differences are discovered.

We can understand the basic principles of genetic epidemiology by studying the behavior of alleles at a single locus in nuclear families. We take advantage of evolution-based constraints on the distribution of genetic variation in families. The analysis of trait distributions in families is known as segregation analysis after Gregor Mendel's Law of Segregation of individual alleles at a locus. The idea is to see if the pattern of phenotypes in families is consistent with a genetic model.

Families are ascertained via one or more index individuals, or probands (propositi), who may be either randomly identified, or chosen because of their disease or other phenotype status. The nature of the sampling must be built into the genetic model. This chapter illustrates the principles of segregation analysis by the study of genotypes that are strongly associated with categorical phenotypes {Cavalli-Sforza and Bodmer, 1971; Elandt-Johnson, 1971; Levitan, 1988; Li, 1961; Morton, 1982; Thompson and Thompson, 1986; Vogel and Motulsky, 1986}.

Appendix 5.1 provides some basic probability theorems for those who need them.

Nuclear families and sibships

The distribution of traits in families

A diploid, sexually reproducing organism has two sets of genes, one inherited from each parent. Each time that individual produces his/her own gamete (sperm or egg), one of his/her inherited alleles, at each locus, will be randomly chosen and transmitted in the gamete. There is thus a probability of ½ that an offspring will inherit a specific parental allele.

It is over forty years since the publication of Hutt's classic text The Genetics of the Fowl. During that period there have been enormous advances in the understanding of genetics in general. With the advent of gene cloning these advances are likely to continue at an increased pace for at least the next decade. On a different level there have also been sweeping changes in the commercial rearing of the domestic fowl as broilers and layers. These changes have been in part attributable to the new strains that have been introduced as a result of research into breeding. An area that has changed least is the way in which the poultry fancier breeds for ‘perfection’ as he or she sees it. It is with these factors in mind that this book has been written. Whilst not aiming to produce an encyclopaedic work like that of Hutt's and the new multiauthor reference work Poultry Breeding and Genetics, edited by R.D. Crawford, it is intended that this book will provide both an introduction to, and a useful survey of, the recent developments in the genetics and evolution of the domestic fowl. There are some areas covered in Hutt's book that have changed little since 1950, particularly the ‘classical’ genetics of many of the physiological and anatomical characters.

Introduction

It is convenient to divide the evolutionary history of the present day domestic fowl into three phases: the first is the evolution of the genus Gallus, the second is the emergence of the domestic fowl from its progenitor(s) within the genus Gallus, and the third is the appearance of the large number of present day breeds and varieties. These three phases occupy very different time spans. The origin of life on earth is estimated to have occurred about 3000 million years ago, whereas the genus Gallus probably dates from about 8 million years (Helm-Bychowski & Wilson, 1986). The domestication of the fowl in the region of the Indus valley is believed to have occurred by 2000 BC (Zeuner, 1963), but more recent archaeological evidence shows that a much earlier domestication occurred in China c. 6000 bc (West & Zhou, 1989). The origin of most of the present day breeds and varieties dates from the last century, although a few are considerably older. Figure 1.1 illustrates the phases of evolution up to that of the genus Gallus, in relation to other significant events. In this chapter these three phases are considered in turn giving some of the supporting evidence for each stage in evolution.