Chapter 1 An overview of animal breeding programs

Transcription

1 Chapter 1 An overview of anial breeding progras Julius van der Werf INTRODUCTION... 1 IMPORTANT FACTORS IN BREEDING PROGRAMS... 2 BREEDING STRATEGIES... 3 STRUCTURE OF BREEDING PROGRAMS... 5 INTRODUCTION TO MARKER ASSISTED SELECTION... 5 Introduction The face of anial breeding has changed significantly over the past decades. Anial breeding used to be in the hands of a few distinguished breeders, individuals who sees to have specific arts and skills to breed good livestock. Nowadays, anial breeding is uch doinated by science and technology. In soe livestock species, anial breeding is in the hands of large copanies, and the role of individual breeders sees to have decreased. There are several reasons for this change. Firstly, the breeding industry has taken up scientific principles. Looking was replaced by easuring, and an intuition was partly replaced by calculations and scientific prediction. Other ajor developents were caused by the introduction of biotechnology. These are roughly the reproductive technologies, and the olecular genetic technology. Not all of this is new. Artificial inseination was introduced in the fifties in cattle. No doubt that the technology had a ajor ipact on rates on genetic iproveent in dairy cattle, and just as iportant, on the structure of anial breeding progras. Nowadays, technologies like ovu pick up, in vitro fertilization, ebryo transfer, cloning of individuals, cloning of genes, and selection with the use of DNA arkers are all on the ground. Soe of the technologies are already applied, others are further developed, or waiting for application. Finally, the rapid developent of coputer and inforation technology has greatly influenced data collection and genetic evaluation procedures in livestock populations, now allowing coparison of breeding values across herds, breeds or countries. The introduction of breeding ethods typically needs to find the right balance between what is possible fro a technological point of view and what is accepted by the decision akers and users within the socio-econoic context of a production syste. Ultiately it is the consuer who decides which technology is desirable or not. In ost western societies, consuers are increasingly aware of health, environental and anial welfare issues. Food safety and ethods food production are part of their buying behavior. However, price and production efficiency reain to be ajor contributors to sustainability of a livestock industry. Successful anial breeding progras need to find the right dose of technology that helps the to be copetitive. 1

2 Iportant factors in breeding progras Which decisions need to be ade? In essence, the two key questions in anial breeding are: Where to go? and How to get there? Running an anial breeding progra involves the answer to these questions, which can be worked out in a bit ore detail as: 1. What is the breeding objective: which traits need to be iproved and how iportant are different traits in relation to each other. 2. What and who do we easure? Which traits, which anials? 3. Do we need to use any reproductive technology (Artificial Inseination, Ebryo Transfer) if possible. 4. How any and which anials do we need to select as parents for the next generation 5. How to ate the selected ales and feales Figure 1. Decision issues in anial breeding Where to Go Breeding objectives How to get there Trait easureent - Milk, wool, growth, carcass, fertility - Males vs feales - Progeny test - Nucleus vs. coercial anials Reproductive technology - Artificial Inseination - Multiple Ovulation/Ebryo Transfer Estiation of breeding value - phenotypes - pedigree - BLUP - Genetic Markers Selection & culling - Genetic evaluation - Balancing rate of genetic change and inbreeding Mating The definition of the breeding objective is the first and probably ost iportant step to be taken. Iproving the wrong traits could be equivalent or even worse than no 2

3 iproveent at all! If any breeding anials ales will be considered for reasons irrelevant to the breeding objective, than the selected group will not be as good with regard to the breeding objective as was expected. It is iportant in the selection process that the selection criterion is clear, and whether the selection is efficient in relation to the breeding objective. Breeding Strategies Reproductive rate of breeding anials and uncertainty about true genetic erit of breeding anials ake up the ost iportant liiting factors in a breeding progra. How any and which anials should be selected is deterined by these factors. Investents in breeding progras are therefore often related to trait easureent and genetic evaluation, and to technology to increase reproductive rates. Measureent Effort and Genetic Evaluation The benefit of abundant and good easureent is that we ay better be able to identify the genetically superior anials. This leads to ore accurate selection and ore genetic iproveent. Phenotypic easureents are turned into Estiated Breeding value s (EBV s). Estiation of breeding value based on an anial s phenotype alone can already be quite accurate for high heritable traits. However, anials need to be copared across herds, and genetic and environental influences have to be disentangled. To achieve this, ore sophisticated statistical ethods are used, leading to Best Linear Unbiased Prediction (BLUP) of breeding values. Besides allowing across herd coparisons, BLUP also uses all available inforation about an anials breeding value, including data on related anials. Selection accuracy is strongly dependent on the degree of data recording, which requires a range of considerations related to cost and infrastructure. In data recording, individual perforances need to be related to anial identification. If BLUP is used to generate EBV s also an anial s pedigree needs to be known (in principle, for each anial only sire and da). If pedigree is not recorded, breeding value can be assessed on own perforance only, and is liited to sexes, which express the traits of interest. BLUP relies good structure of data (use of breeding anials across herds) and proper pedigree recording. If these prerequisites are in place, investent in BLUP ethodology is usually highly cost efficient. Molecular genetic technology has rapidly developed in the past 2 decades. Genes have been found coding for factorial traits (such as any diseases). Many production traits are quantitative traits and a likely genetic odel is here that genetic differences between anials are due to any genes. However, DNA technology has also provided genetic arkers. Certain genetic arkers can iprove estiation of an anial s genetic potential as they are associated with regions that account for genetic variation. Genotyping anials for arker genotypes is therefore an investent with the ai to better assess true genetic erit of anials. Reproductive technology Most of the ain factors that deterine genetic gain are directly influenced by the reproductive rate of the breeding anials. A higher reproductive rate leads to the need for 3

4 a decreased nuber of breeding anials, therefore increasing the intensity of selection of these anials. If reproductive technology is possible, for exaple AI, the benefit could be expressed in ters of increased genetic rate of iproveent, which in turn has a dollar coponent attached to it. More offspring per breeding anial allow also ore accurate estiation of breeding value. Reproductive technology allows the intensive use of superior breeding stock. An obvious consequence is possibly that the ost popular breeding anials are overused, and the population could encounter inbreeding probles. Typically, as new technologies in anial breeding allow faster genetic change, long ter issues such as inbreeding and aintenance of genetic variation becoe iportant. For that reason, selection tools in anials breeding have becoe soewhat ore sophisticated in recent years. The ipact of reproductive technologies on rates of genetic iproveent and inbreeding will be discussed. Besides a direct effect on rate of genetic iproveent, another iportant consequence fro increasing reproductive rates is to disseinate superior genetic stock quickly. The influence of a superior breeding anial would be uch higher if thousands of offspring could be born, rather than if the superiority is passed on through the production of sons via natural ating. Another exaple is that of cloning. Cloning is not extreely iportant for increasing rate of genetic progress, but it could have a large ipact by allowing any copies of the best individual to perfor in coercial herds. As reproductive rates are basically ultiplying factors in a breeding structure, any iproveent in reproduction will justify higher investent in iproveent of the best breeding stock Selection and Mating The decision about which anials should be selected as parents for the next generation is ainly based on assessent of breeding value of individual anials. Genetic evaluation is central to anial iproveent schees. Selecting anials based on estiated breeding value axiizes the response to selection that can be achieved. However, there is one other criterion that is relevant when deciding which anials should have offspring. This criteria is coon ancestry of all selected parents. The coancestry of selected parents should stay below certain liits, since it is directly related to the build up of inbreeding. Coancestry aong selected parents is deterined by the average relationship aong the selected parents as well as the nuber of parents selected. In this course we will ore explicitly discuss selection strategies that aintain low levels of inbreeding. Decisions about which anials need to be ated are often seen in relation to doinance effects. Utilizing doinance variation is often not of priary iportance for iproveent of purebreds, but it can have ore ipact if breeding anials are selected fro different breeds or lines, as heterotic effects between breeds can be utilized. When ultiple traits are involved in the breeding objective, assortative ating could be useful, atching qualities in different parents for different There is a good possibility that in the near future, planned ating will gain in iportance, when effects of specific genotypes will be better understood. One could envisage certain genotypes with high growing potential to be cobined with specific genes that have ajor effect on eat quality. Another arguent for planned atings is to 4

5 avoid inbreeding in direct offspring as well as the rate of inbreeding in the population. However, the rate of inbreeding depends ainly on population size and nuber of parents selected. Methodology to optiize selection and ating decisions related to inbreeding will be discussed. Structure of breeding progras Most of the key decision factors entioned earlier are related to the rate of genetic change that can be ade. However, this could be genetic change in a sall fraction of the national population (in nucleus or elite breeders ). Genetic superiority should be transferred as soon as possible to ost of the coercial fars. The structure of a breeding progra is therefore relevant for two aspects of an iproveent schee: 1) The genetic iproveent aspect: how do we deterine the genetically superior anials. 2) The disseination aspect: how do we anage that those superior anials disseinate their genes quickly though the whole population of production anials We often talk about the design of a breeding progra, suggesting that breeding progras can be characterized by soe kind of structure. The traditional odel here is the pyraid with a sall group of breeding anials that are actually iproved (the elite breeders in the nucleus) and underlying levels of (possibly) a ultiplier and a coercial. The later groups ay not be involved in selection, but erely, they receive genes fro the nucleus and are therefore iproved over tie. The genetic ean of lower tiers is soewhat lower than that of the nucleus, but the rate of iproveent is in principle equal. Introduction to arker assisted selection Nucleus Multipliers Coercial Farers Over the last two decades ost livestock industries have successfully developed EBV s to allow identification of the best breeding anials. EBV s are best calculated using BLUP, eaning that they are based on pedigree and perforance inforation of several traits fro the individual anial and its relatives. BLUP EBV s are the ost accurate criteria to identify genetically superior anials based on phenotypic perforance recording. Although the idea of genetic selection is to iprove the genes in our breeding anials, we actually never really observe those genes. Selection is based on the final effect of all genes working together, resulting in the perforance traits that we observe on production anials. This strategy akes sense, since we select based on what we actually want to iprove. However, anial perforance is not only affected by genes, but also by other 5

6 factors that we do not control. Selection for the best genes based on anial perforance alone, can never reach perfect 100% accuracy. A large progeny test coes close such a figure of perfect selection, but this is expensive for soe traits (e.g. for traits related to eat quality), and we have to wait several years before the benefits fro a progeny test have an effect. Efficient breeding progras are characterised by selecting anials at a young age, leading to a short generation intervals and faster genetic iproveent per year. For selecting at younger ages, knowledge about the existence of potentially very good genes could be very helpful. Quantitative genetics uses phenotypic inforation to help identify anials with good genes. Extension to use inforation fro olecular genetics techniques ai to locate and exploit gene loci which have a ajor effect on quantitative traits (hence QTL - Quantitative Trait Loci). The idea behind arker assisted selection is that there ay be genes with significant effects that ay be targeted specifically in selection. Most traits of econoic iportance are quantitative traits that ost likely are controlled by a fairly large nuber of genes. However, soe of these genes ight have a larger effect. Such genes can be called ajor genes located at QTL. In practice, we rarely know the genotype at actual QTL, as the exact gene location (utation) is often unknown. Currently there are few exaples where QTL effects can be directly deterined, but knowledge in this area is rapidly developing. Most QTL known today can only be targeted by genetic arkers. Genetic arkers are landarks at the genoe that can be chosen for their proxiity to QTL. We cannot actually observe inheritance at the QTL itself, but we observe inheritance at the arker, which is close to the QTL. When aking selection decisions based on arker genotypes, it is iportant to know what inforation can be inferred fro the arker genotypes. Figure 3 shows the principle of inheritance of a arker and a linked QTL. We can identify the arker genotype (M) but not the QTL genotype (Qq). The last is really what we want to know because of its effect on econoically iportant traits. 6

7 Let the Q allele have a positive effect, therefore being the preferred allele. In the exaple, the M arker allele is linked to the Q in the sire. Progeny that receive the M allele fro the sire, have a high chance of having also received the Q allele, and are therefore the preferred candidates in selection. Gene location Q q q q Parents: Marker location M X Bull Cows Progeny: Q q q q q q Q q M M Probabilities: 90% 10% Figure 3: Following the inheritance of a ajor genotype affecting a quantitative trait (Q locus) with a genetic arker (M locus) closely linked to the Q locus. The sire is heterozygous for either locus and the da is hoozygous. For this exaple, we can deterine for each progeny whether they received M or allele fro their sire. The recobination rate (10%) deterines how often Q alleles join M alleles. 7