Pages:     | 1 |   ...   | 2 | 3 || 5 |

CGRFA/WG-AnGR-7/12/Inf.6 E September 2012 ...

-- [ 4 ] --

CGRFA/WG-AnGR-7/12/Inf.6 Table 8. Rate of inbreeding (percentage) and effective population size (in parenthesis) predicted under different management regimes Within-family selection Number of Random Snchez males and Gowe selection Rodrguez females 3 9 5.6 (8.9) 3.5 (14.3) 2.9 (17.2) 5 25 3.0 (16.7) 2.0 (25) 1.7 (29.4) 6 18 2.8 (17.9) 1.7 (29.4) 1.4 (35.7) 10 50 1.5 (33.3) 1.0 (50) 0.8 (62.5) Note that the formula for random contributions shown above only holds in the absence of selection on a trait, which is highly unrealistic for domestic animal populations. There is almost always some mass selection, because owners keep the individuals with high performance and thus co-select relatives more often than would occur by chance. Thus, this selection should be accounted using the method proposed by Santiago and Caballero (1995) (see Section 2). This simple approach involves dividing the F arising from the formula by a factor of 0.7. However, under the above-described regular system methodology, within-family selection can be applied without increasing the rate of inbreeding (see Section 7 for further details).

Simple strategies Sufficient control of contributions from individuals can be achieved with rather simple strategies. Where AI is practised, an approximately equal number of semen doses from each sire should be collected and distributed in order to minimize the variance in the number of offspring sired by the males. Recalling that the less-represented sex has the greatest influence on F, a simple strategy might be to limit the percentage of offspring each sire contributes to the next generation. Implicitly this means that the largest number of sires possible is involved in the breeding programme. In the most extreme circumstances, each sire should leave only one son to the subsequent generation (if the population is growing the number of sons per sire should still be equal, but will be greater than one). In this way the variance in the number of sons sired by the males is reduced to zero. In highly prolific species, some attention should be also given to equalizing contributions from females, i.e. to avoid a situation in which a limited number of breeding females contribute progeny to the next generation, and in particular to ensure that the subsequent generation of sires comprises the offspring of different females.

Minimum coancestry contributions methodology using pedigrees When pedigree data are available, the minimum coancestry contributions methodology, a more sophisticated strategy, can be applied. As explained above, the coancestry coefficient (f) is a measure of the probability that animals share identical alleles by descent. Relatives have common ancestors and are thus likely to carry identical copies of alleles. Some of the genetic information in relatives is redundant, and it does not matter which relative transmits it as long as the shared alleles pass to the next generation. Consequently, the individuals effectively contributing to the future population and the number of offspring from each individual can be derived on the basis of their coancestry with the rest of the population. In this process, animals that are closely related to the general population will be penalized (and only allowed to produce a few or no offspring), whereas relatively unrelated individuals will be chosen to produce more offspring. These latter animals are assumed to carry unique genetic information that would be lost if they did not produce offspring. This strategy minimizes F, at least in the short and medium terms.

The minimum coancestry contributions methodology is robust against deviations from ideal conditions (accounts for related founders, does not need regular schemes with equal numbers of offspring per parent, can cope with mating failures). It also allows for the imposition of restrictions that correspond to the physiology of the particular species (e.g. maximum number of offspring from any individual).

However, the methodology has some practical disadvantages. First, it requires tight control of the reproductive process and thus may only be applicable in particular populations, such as nucleus herds.

CGRFA/WG-AnGR-7/12/Inf.6 Second, the calculations are computationally complex. The aim of the methodology is to find the set of contributions c i (i.e. number of offspring per individual i) that minimizes the function c i c j f ij, where f ij is the coancestry between every possible couple of individuals i and j. Even for small populations, the number of feasible solutions is huge, and finding the optimal solution requires the use of complex algorithms and the aid of computers. Therefore, expertise is required to implement the methodology. The free software METAPOP 10 (Prez-Figueroa et al., 2009) facilitates its implementation in a conservation programme without artificial selection.

The minimum coancestry contributions methodology was originally developed to work with coancestries calculated from pedigree data. It is thus strongly recommended that the pedigree of the animals be recorded in any in vivo conservation programme, so the coancestry can be calculated for the management of the population and the F can be used for monitoring the success of the programme. The benefits obtained from the recording of the pedigree (just the sire and dam of every animal) generally exceed the extra cost the procedure involves.

Minimum coancestry contributions methodology using molecular information In addition to measurement of genetic variability and prioritization of breeds (Section 3), molecular markers can be a powerful tool in the management of populations. When pedigree data are not available, molecular information can be used in the management of populations to decrease genetic drift. The possibilities for using molecular data in the management of animal genetic resource diversity include the following:

1. recovery, reconstruction or correction of partial genealogies (e.g. through paternity analysis for solving uncertain parentage Jones et al., 2010);

2. estimation of pedigree coancestry from molecular measures of similarity (Ritland, 1996);

and 3. replacement of coancestry matrices based on pedigrees with the corresponding molecular coancestry matrices (Fernndez et al., 2005).

The outcomes of these three alternatives can be used as input for the implementation of the minimum coancestry mating strategy. Several computer tools exist for the estimation of pedigree relationships from molecular markers: for example, SPAGEDI 11 (Hardy and Vekemans, 2002) and COLONY (Jones and Wang, 2009). A review of free software available for paternity analysis and for coancestry estimation can be found in Martnez and Fernndez (2008).

Technological advances are continually decreasing the costs of molecular analyses, thus increasing the feasibility of their use in population management. In particular, the development and commercialization of panels of single nucleotide polymorphisms (SNP) has greatly expanded the precision with which molecular information can be used to manage genetic diversity (see Box 40). Further developments in genome sequencing will only expand the opportunities.

http://webs.uvigo.es/anpefi/metapop/ http://ebe.ulb.ac.be/ebe/Software.html http://www.zsl.org/science/research/software/colony,1154,AR.html CGRFA/WG-AnGR-7/12/Inf.6 Box Population management using genomic information The utility of marker genotypes for the management of populations depends on the amount of information they can provide about the diversity at the rest of the loci (i.e. the unmarked loci) in the genome. The information content is connected to the degree of correlation between genotypes at marker loci and the rest of the genome (this correlation is referred to as linkage disequilibrium), which is inversely related to the physical distance between loci in the genome and to the N e.

When the number of markers is low (e.g. usual panels of microsatellites), the amount of information provided for genomic regions between markers is limited and genealogical coancestry (i.e. calculated from pedigrees) is preferred for the management of diversity (Fernndez et al., 2005). Nowadays, however, SNP chips containing a very large number of markers are available for many livestock species (up to 800 000 for Bos taurus). This high density ensures that every locus in the genome is in linkage disequilibrium with at least one SNP. Consequently, molecular coancestry is a more precise measure of genetic relationships between individuals than pedigree data, and greater diversity can be maintained when management is based on molecular genotypes (De Cara et al., 2011).

Genome-wide information also allows for the measurement and maintenance not only of neutral genetic variability, but also of selective genetic diversity important for the productivity and evolution of the population. Therefore, SNP analysis of the genomes has become the method of choice for research and population management when DNA of individual animals is available or can be obtained, because the costs of SNP analysis are decreasing to an acceptable level.

Molecular information to clarify relationships between individuals Even when a conservation programme includes pedigree recording, it is advisable to use molecular information to check the correctness of the pedigrees (e.g. to resolve paternity uncertainties) and to determine the genetic relationships among the founders of the programme (the term founders here refers to the base population of animals with which the conservation programme begins and after which pedigrees are routinely recorded). The ancestors of these animals are unrecorded and their pedigrees are thus uncertain. Traditionally, individuals at the base population are assumed non-inbreed (F = 0) and non-related (f = 0 for all pairs of individuals and f = 0.5 for self-coancestries). Most populations under conservation have been maintained with a limited number of individuals (parents) for one or more generations. Thus, assuming non-related founders is highly unrealistic and can lead to incorrect management.

A rough idea of relationship between founders can also be deduced from historical information obtained from their place of origin (e.g. the farm or geographic area from which they came). However, a more accurate approach is to construct a matrix of estimated coancestries based on molecular information on the founders (i.e. by using any of the methods and software described above). The coancestry of animals in subsequent generations is then calculated following the classical rules of pedigree analysis. Minimum coancestry contributions methodology will integrate the information about the relationships between founders, correcting for the disequilibria generated during the unmanaged generations that elapsed before the programme started.

The need to properly characterize relationships between founders is especially important for pedigrees with few recorded generations. In these situations genealogical parameters (e.g. F or F) are poorly informative in the first generations of management, and decisions made at this point can have a huge impact on the probability of long-term success in maintaining variability. When little or no information about founders is available, extra effort should be made to ensure that all animals in the population produce offspring.

CGRFA/WG-AnGR-7/12/Inf.6 Action 5. Consider the use of embryo transfer in species with low reproductive rates As noted in Section 1, reproductive biotechnologies such as AI and embryo transfer are occasionally cited as factors contributing to breed risk, as they facilitate the spread of germplasm across long distances and can contribute to shrinking N e by decreasing the number of different parents required.

However, the real reasons for breeds being at risk are factors such as their relatively low productivity and the lack of policies for their maintenance. In fact, when used strategically, reproductive biotechnologies can enhance conservation programmes.

For example, embryo transfer can increase the number of offspring per female. Increasing the number of offspring per female can have two positive effects. First, assuming that recipient females are of another breed, embryo transfer can be used to increase the census size of the population more quickly.

Second, increasing the number of offspring per female is a very efficient way to equilibrate the ratio between male and female parents, especially if each female embryo donor is mated to multiple males.

Sexed semen can provide similar (but smaller) benefits, at least in populations where (when unsexed semen is used) only a portion of the males born are needed for breeding.

Embryo transfer can also extend the generation interval, if used to obtain offspring from females that are no longer able to maintain pregnancies of their own. This benefit can be augmented further when combined with cryoconservation (see Task 3). This is a strong argument for cryoconserving embryos from populations that have a critical or endangered risk status in a gene bank.

One constraint to this approach is that embryo transfer can be technically demanding and requires considerable training to yield good results. Embryo transfer is also costly, so a financial feasibility study should be undertaken beforehand to evaluate the costs and benefits. Finally, the status of development of embryo transfer protocols is not equal across species and breeds within species. Most of the standard protocols have been developed for widely used commercial breeds of the major livestock species and some trial and error must be expected to adapt the standard protocols to less common breeds and species.

Task 2. Adopt a mating strategy to decrease inbreeding In the long term, the number of parents chosen and the number of offspring they produce are the main factors affecting genetic variability. However, after the selection of the parents, inbreeding and its detrimental effects can be further controlled by managing how the selected parents are mated with each other.

At least for one generation, the amount of genetic variability transmitted to or lost from the population is not dependent on the mating scheme but only on the number of offspring each individual contributes.

However, the level of inbreeding (mean F) is highly dependent on which animals are mated to each other. The F of an individual is simply equal to the coancestry between its parents (f). The greater the relationship between sire and dam, the higher is the F of their progeny. Therefore, mating between relatives should be actively avoided. Several approaches (see Actions that follow) can be taken to avoid mating of relatives.

Action 1. Set a limit on the level of relationship between mates The simplest way to decrease inbreeding to avoid mating between individuals that exceed a certain threshold of coancestry for example half-sibs (e.g. f = 0.125). If potential mates are already inbred on the same ancestor, then this factor should also be accounted for, when possible. When several generations of the pedigree are known, the types of relationships that can be identified are more complex. In these cases, each livestock keeper could be provide (e.g. by the breeders association) information on specific matings that should be avoided (or alternatively that are recommended).

Action 2. Establish the ideal set of matings for the entire population A mathematically optimized approach to avoid mating of relatives has been developed and can be applied across a population. This approach is called the minimum coancestry mating design, and consists of finding the set of potential mates of selected parents that has the minimum average coancestry between partners (sires and dams). The methodology delays the rise of inbreeding, although does not reduce F in the long term (Woolliams and Bijma, 2000). As is the case with the above-described optimal method of fixed contributions, the number of possible combinations is huge and the use of mathematical and computational techniques is required to solve the problem. Minimum coancestry CGRFA/WG-AnGR-7/12/Inf.6 mating design can be performed with the aid software, such as METAPOP 13 (Prez-Figueroa et al., 2009). Obviously this methodology can only be implemented in situations where mating is under central control. This rarely occurs in field conditions, but may be encountered in ex situ populations.

Action 3. Apply simple methods that do not require pedigree information Circular mating systems In the absence of genealogies, another mating strategy can be used. The idea is to arrange n families of individuals in a virtual circle. Male offspring from family 1 are always mated to females from family 2;

males from family 2 are mated to females from family 3;

and so on. Males from family n are mated to females from family 1. This strategy is known as the circular mating design (Kimura and Crow, 1963).

An example of the implementation of such a programme is presented in Box 41. The methodology is easy to implement and ensures low F in the long term, although it may increase F in the short-term, due to partial subdivision of the population. When the population is maintained in several herds, and each herd is considered as a family, the procedure converges with the so-called rotational system of breeding management. In this system some individuals are regularly (e.g. every year or every generation) exchanged between neighbouring herds and random mating is performed in the herds. This obviously implies some degree of organization and acceptance on the part of all participating livestock keepers.

Past experience has shown that such a programme can be a great success or colossal failure, depending on the level of organization and cooperation among the livestock keepers. When starting from scratch (i.e. no previous subdivision of the population) one option could be to establish homogeneous groups by using cluster analysis methodologies to separate the population into as many lines as desired based on the genetic structure. When pedigree data are available, standard statistical software, such as SAS, SPSS or R, can be used to cluster animals according to their genealogical relationships. Various software are available for clustering based on molecular data, among which STRUCTURE 14 (Pritchard et al., 2000) is one of the most commonly used.

Box Mating systems to control inbreeding in Colombia The Ministry of Agriculture of Colombia has maintained several nucleus farms for in vivo conservation of Criollo cattle breeds since 1936. During the initial decades of the breeding programmes, inbreeding was controlled by avoiding the mating of close relatives such as sire and daughter, son and dam, full and half sibs and cousins.

However, since 1991, the breeding programme of four breeds (Romosinuano, Blanco Orejinegro, Costeo con cuernos and Sanmartinero) was changed to the circular mating design. In each of the four breeds, all the animals were grouped into one of eight families according to their genetic relationships (i.e. eight families per breed). Then, the circular mating design was followed in each breed. Males of family 1 were mated to females of family 2, and so forth:

123456781. Several years later, the system was slightly modified every three years the pattern is adjusted by skipping one family ahead: 13;






71 and 82. This change was necessary because after a few breeding cycles, the females from family 2 were related to most of the males from family 1, females from family 3 to males from family 2, and so forth. To help facilitate the process and ensure proper matings, the offspring produced always receive the name or denomination of the dams family.

Provided by German Martinez Correal.

http://webs.uvigo.es/anpefi/metapop/ http://pritch.bsd.uchicago.edu/structure.html CGRFA/WG-AnGR-7/12/Inf.6 Factorial matings When females give birth multiple times during their lifetimes (which is the ideal scenario, as it increases the generation interval), factorial matings should be used. This means that each female should be given the opportunity to mate to different males. In this way, the number of possible mating combinations is larger and outcomes are better in terms of the amount of diversity maintained and the levels of inbreeding. In contract, hierarchical mating (mating each female to a single male), can result in the production of large full-sib families. When hierarchical mating is combined with selection (natural or artificial) the probability of selecting relatives is high. Moreover, if a male carries a dominant deleterious allele, the female genetic information has a high risk of being lost because all of its descendants may carry the deleterious allele. But if the female is mated to several males, its contribution will be safely transmitted through her offspring with other male partners.

Mate selection Managing a population in two steps (i.e. first select individuals and determine their contributions, and then choose the mating design) is an option, but it may lead to complicated and less practical situations. For example, it may require the mating of a female with two males, which for many species is impossible in the same oestrous period without the use of reproductive techniques like MOET (multiple ovulation and embryo transfer). Therefore, it may be advisable to undertake both processes in a single step via a procedure called mate selection. The basis of this approach is determining the number of offspring to be born from each possible set of mates instead of from each individual. In this way, all physiological or logistical restrictions (e.g. limits on the number of mates per male or female, limits on the number of offspring per couple) can be accounted for.

Task 3. Incorporate cryoconservation in the management of genetic variation in the in vivo programme Cryoconservation (for more information, see Cryoconservation of animal genetic resources FAO, 2012) is another useful tool in a conservation programme (Meuwissen, 2007). It provides a manifold benefit to the programme, as it extends the reproductive lifespan of individuals (i.e. increases the generation interval) and increases the both the real population size and the N e, as more individuals (which are less likely to be closely related) are available for mating at the same time. The storage of semen or embryos can address different objectives.

Action 1. Store genetic material from all animals at the start of the conservation programme A first objective may be to use the collected material to create a backup of the breed, i.e. to store all the genetic diversity present at the beginning of the programme (one generation in case of embryos or somatic cells;

two or more generations in case of semen). In the event of population extinction in the future, it will be possible to recover the breed using the stored material. The creation of such a bank is advisable for not-at-risk and vulnerable breeds and is strongly recommended for endangered and critical breeds that are likely to disappear in the near future. Obviously, storing material from all individual males and females would be feasible and logical only for a population at a critical level of risk.

In most situations, the germplasm is primarily stored for insurance purposes, and the probability that it will be needed is (hopefully) small. Therefore, approaches such as the storage of somatic cells that have low collection costs but high utilization costs (i.e. for cloning) may be logical options.

Action 2. Use cryoconserved material continually for management of the genetic diversity A second objective of cryoconservation is to reinforce the in vivo programme. Cryoconserved germplasm can serve various purposes in a conservation programme. It may be used in a discrete way to help the population recover from a critical state (e.g. following a catastrophe that has reduced the population size). The cryopreserved material may also be used continually as part of the normal management procedure in a critical or endangered breed (Sonnesson et al., 2002). For example, cryoconserved semen, the number of sires can be increased, thus increasing N e and decreasing costs (relative to keeping live males). In this scenario, collection of material for the gene bank should be a continual and permanent process that replenishes the doses used. Box 42 describes how the use of cryoconserved semen has helped increase genetic variability within a breed of sheep in France.

CGRFA/WG-AnGR-7/12/Inf.6 Box Cryoconservation to increase the genetic diversity of a population in vivo an example from France The Roussin de la Hague is a French sheep breed that was considered to be at-risk during the 1990s.

However, its status has improved considerably since then. The total number of ewes is now estimated to be more than 3 000. As part of the breeds recovery programme, semen from 13 rams was collected at the beginning of the 1990s and the stock was eventually moved to the French National Cryobank.

An analysis was performed in 2010 to evaluate the genetic diversity of the rams in the ex situ collection relative to the active populations (i.e. the rams and ewes that are presently used by the breeders). Results are shown in the figure below. Based on pedigrees, individual kinship () with the active rams (x-axis) and the active ewes (y-axis) was calculated for each cryoconserved ram and compared with the average kinship of the active rams with themselves and the active ewes ().

Clearly, most of the cryoconserved rams are weakly related to the active population. Most of the rams have kinships of less than 2 percent with both current males and females and all have less relationship than the average of the in vivo population. The cryoconserved rams represent genetic diversity that no longer exists in situ. Other studies (Danchin-Burge et al., 2009) have shown that the Roussin breed went through a bottleneck in the 1990s, when only three farmers were providing a large majority of the rams. Now that the breeds demography has improved, it is time to think about improving also its genetic diversity;

some farmers have decided to use the semen from the cryoconserved rams to produce their replacement animals.

Provided by Coralie Danchin-Burge.

References Caballero, A. & Toro, M.A. 2000. Interrelations between effective population size and other pedigree tools for the management of conserved populations. Genetical Research, 75: 331343.

Danchin-Burge, C. Palhire, I., Franois, D., Bib, B., Leroy, G. & Verrier, E. 2009. Pedigree analysis of seven small French sheep populations and implications for the management of rare breeds. Journal of Animal Science, 88: 505516.

de Cara M.A., Fernndez, J., Toro, M.A. & Villanueva, B. 2011. Using genome-wide information to minimize the loss of diversity in conservation programmes. Journal of Animal Breeding and Genetics, 128: 45664.

CGRFA/WG-AnGR-7/12/Inf.6 EURECA. 2010. Local cattle breeds in Europe, edited by S.J. Hiemstra, Y. De Haas, A. Mki-Tanila & G. Gandini. Wageningen, the Netherlands, Wageningen Academic Publishers (available at http://www.regionalcattlebreeds.eu/publications/documents/9789086866977cattlebreeds.pdf).

Falconer, D.S. & Mackay, T.F.C. 1996. An introduction to quantitative genetics, 4th edition. Harlow, UK, Longman.

FAO. 2012. Cryoconservation of animal genetic resources. FAO Animal Production and Health Guidelines. No. 12. Rome (available at http://www.fao.org/docrep/016/i3017e/i3017e00.htm).

Fernndez, J., Villanueva, B., Pong-Wong, R. & Toro, M.A. 2005. Efficiency of the use of pedigree and molecular marker information in conservation programs. Genetics, 170: 13131321.

Gowe, R.S., Robertson, A. & Latter, B.D.H. 1959. Environment and poultry breeding problems. 5.

The design of poultry strains. Poultry Science, 38: 462471.

Groeneveld, E., Westhuizen, B.V.D., Maiwashe, A., Voordewind, F. & Ferraz, J.B.S. 2009.

POPREP: a generic report for population management. Genetics and Molecular Research, 8:


Gutirrez, J.P. & Goyache, F. 2005. A note on ENDOG: a computer program for analysing pedigree information. Journal of Animal Breeding and Genetics, 122: 172176.

Hardy, O.J. & Vekemans, X. 2002. SPAGeDi: a versatile computer program to analyse spatial genetic structure at the individual or population levels. Molecular Ecology Notes, 2: 618620.

Jones, A.G., Small, C.M., Paczolt K.A. & Ratterman, N.L. 2010. A practical guide to methods of parentage analysis. Molecular Ecology Resources, 10: 630.

Jones, O. & Wang, J. 2009. COLONY: a program for parentage and sibship inference from multilocus genotype data. Molecular Ecology Resources, 10: 551555.

Kimura, M. & Crow, J.F. 1963. On the maximum avoidance of inbreeding. Genetical Research, 4:


Martnez, P. & Fernndez, J. 2008. Estimating parentage relationships using molecular markers in aquaculture. In S.H. Schwartz, ed. Aquaculture research trends, pp. 59112. New York, USA, Nova Science Publishers, Inc.

Meuwissen, T.H.E. 2007. Operation of conservation schemes. In K. Oldenbroek, ed. Utilization and conservation of farm animal genetic resources, pp. 167-194. Wageningen, the Netherlands, Wageningen Academic Publishers.

Prez-Figueroa A., Saura M. Fernndez J., Toro M. A. & Caballero A. 2009. METAPOP - A software for the management and analysis of subdivided populations in conservation programs.

Conservation Genetics, 10: 10971099.

Pritchard, J.K., Stephens, M. & Donnelly, P. 2000. Inference of population structure using multilocus genotype data. Genetics, 155: 945959.

Ritland, K. 1996. Estimators for pairwise relatedness and inbreeding coefficients. Genetical Research, 67: 175186.

Snchez-Rodrguez, L., Bijma, P. & Woolliams, J.A. 2003. Reducing inbreeding rates by managing genetic contributions across generations. Genetics, 164: 15891595.

Santiago, E. & Caballero, A. 1995. Effective size of populations under selection. Genetics, 139:


Sonesson, A.K., Goddard, M.E. & Meuwissen, T.H. 2002. The use of frozen semen to minimize inbreeding in small populations. Genetical Research, 80: 2730.

Woolliams, J.A. & Bijma, P. 2000. Predicting rates of inbreeding in populations undergoing selection. Genetics, 154: 18511864.

CGRFA/WG-AnGR-7/12/Inf.6 VII. OPTIONS FOR BREEDING PROGRAMMES COMBINING CONSERVATION AND SUSTAINABLE USE Breeds often face the risk of extinction because they provide inadequate economic returns to livestock keepers, who need to be compensated for the costs of inputs and labour and to sustain their livelihoods.

The availability of highly specialized international transboundary commercial breeds with greater production potential speeds up this process. Many countries have chosen to import germplasm from such breeds in an effort to quickly increase productivity. However, the commercial breeds are often not suited to local conditions and require significant financial investment to exploit their genetic advantage in production potential. In many cases, the costs of these investments are not fully compensated by the additional production obtained. An alternative option is to take advantage of local breeds advantage in adaptation to the prevailing environmental conditions and to implement breeding programmes to increase their productivity.

Importance of adaptation Mirkena et al. (2010) summarized the genetics and adaptation in farm animals:

Adaptive fitness is characterized by survival, health and reproductive related traits. The wealth of knowledge generated so far indicates that genetic variation for adaptive performance particularly disease resistance is ubiquitous both within and among breeds of livestock indicating that genetic studies on adaptation of farm animals can be determined at three genetic levels: species, breed and unique genetic variation among individual animals within a breed. In the warmer tropical areas, where pathogens and epidemic diseases are widespread, climatic conditions are stressful, and feed and water are scarce, locally adapted autochthonous breeds display far greater level of resistance and adaptation due to their evolutionary roots as compared to imported breeds. There are three pathways of genetic improvement: improvement of local breeds through pure-bred selection, breed substitution (by other local breeds or, more frequently, by exotic breeds), and systems of crossbreeding (terminal crosses, rotations, formation of synthetic lines). Whichever pathway to follow, choice of the most appropriate breed or breeds to use in a given environment or production system should be the first step when initiating a breeding programme and due attention must be given to the adaptive performance. A major limitation is that selection for less heritable traits such as fitness-related traits results in low selection response due to measurement problems and the underlying antagonistic biological relationships between productive performance and adaptive traits. The appropriate strategy for any breeding programme would therefore be to set suitable selection goals, which match the production system rather than ambitious performance objectives that cannot be reached under the prevailing environment. An area-specific approach utilizing the existing resources and taking into account the prevailing constraints appears to be the only reasonable sustainable solution. Such an approach would also enable in situ conservation of animal genetic resources, the only viable and practical conservation method in less developed countries compared to ex situ or cryoconservation approaches. Therefore, the importance of identifying the most adapted genotype capable of coping with the environmental challenges posed by any particular production system has been indicated. The production potential of breeds at risk is usually poorly documented due to the high cost of performance recording. Evidence for adaptive and fitness traits is often anecdotal. Knowledge gaps can be addressed by implementing or supporting characterization studies, but low economic returns will continue to threaten the survival of the breed(s) in the short term. Potential for long-term survival is meaningless if short-term survival is not ensured. Various measures can be taken to improve a breeds economic performance and provide livestock keepers with returns to justify maintaining the breed. Governmental support or incentive payments can help to rescue a breed in the short term, but are unlikely to be sustainable in the long term (EURECA, 2010).

CGRFA/WG-AnGR-7/12/Inf.6 Breeding for economic performance Two main breeding strategies for enhancing economic performance are:

1. to increase production through within-breed selection;

and 2. to implement cross-breeding between locally adapted breeds (with their unique adaptive and fitness traits) or between locally adapted breeds and transboundary commercial breeds (with greater genetic potential for production).

When the population size is small, it is of critical importance to optimize the selection response and the genetic variability within the population.

As in all conservation programmes (i.e. regardless of whether genetic improvement or maintenance of genetic diversity is the main goal), any proposed breeding programme must be must be thoroughly evaluated in advance, taking into account the expected benefits and costs and possible pitfalls. A wrong decision may drive the whole programme to complete disaster and the population to extinction.

It is highly recommended that anyone planning to develop a breeding programme should contact people who have been involved in past attempts to develop such programmes in the same or similar populations and environments and learn from their successes and failures. Kosgey et al. (2006) point out some of the factors influencing the probability of success in the establishment of breeding programmes in local breeds. These factors include:

1. the ability of the programme to address the needs of local livestock keepers;

2. the compatibility of proposed changes and innovations with the existing production system;

3. the availability and appropriateness of incentives (economic and other) for the livestock keepers to participate in the programme;

and 4. the extent of support services (e.g. veterinarians) available in the area.

Several promising options are outlined by Wurzinger et al. (2011).

Improvements through breeding Rationale Selection for production traits The most obvious route to enhancing economic performance is to increase the production of commodities for the mainstream market, such as meat, milk and eggs. Success is most likely where the production potential of the local breed is already high but has not been sufficiently documented and appreciated (see Box 43). In such situations, interventions such as improvements in management and marketing may be sufficient to significantly increase economic return to the livestock keepers and improve the risk status of the breed (see Section 8). However, such situations may not be common.

Almost any livestock breed will benefit directly from attention to classic animal breeding and improvement schemes. The potential of these breeding programmes to achieve greater productivity than obtained by transboundary commercial breeds will vary from place to place, and is likely to be greatest in challenging environments where breeds that are not locally adapted face major hurdles to survival and production due to problems with adaptation and fitness. However, implementation of breeding programmes may be more difficult in these areas as well.

Selecting for enhanced production in a pure-bred local breed is an attractive option. However, selection implies changing the breed, so consideration must be given to the possibility that some changes may not serve the long-term interests of the breed or its breeders. If a selection programme is planned carefully and if the breeds adaptations to local conditions are maintained, then the result can be a well-adapted, productive breed. There are many examples that illustrate the success of such an approach: Nguni cattle in South Africa, Spanish goats in the United States of America (Box 44), Mertolengo cattle in Portugal, and Colonial Spanish Horses in the Americas. Establishing a pure-bred nucleus herd where emphasis is given to recording traits and selection of breeding animals can not only enhance selection for increased production, but also enhance publicity for the productive potential of the breed (FAO, 2003).

CGRFA/WG-AnGR-7/12/Inf.6 Box The importance of locally adapted breeds in the Plurinational State of Bolivia The Ayapaya llama is a local strain with high production potential that had been overlooked by most development programmes. These llamas are kept by the Wallatani highland community in Plurinational State of Bolivia and have better fibre traits than lowland animals have. Selection of local animals has been established as a formal activity (in contrast to unorganized past efforts) and benefits the local community. Similarly, in this environment, some local Bolivian guinea pig strains are superior to imported ones for litter size, number of young weaned and total weight produced. Identifying these local animal genetic resources of high production potential is important for achieving the dual goals of improving the livelihoods of livestock keepers while sustaining national animal genetic resources.

Source: Valle Zrate (1999).

Box Optimum body weight for Spanish goats guarantees adaptation to the climate in Texas, United States of America In the 1960s, West Texas ranchers began selecting local Spanish goats for production characters.

Selection alone, with no cross-breeding, increased mature size of females from 35 kg to 70 kg.

Breeders then discovered that females over 60 kg were less well adapted to the challenging semi arid West Texas environment. Once they relaxed their perception of the ideal weight down to 60 kg, the breeders were able to have the increased production they sought, as well as the environmental adaptation they needed. Larger and non-adapted specialized breeds had little opportunity to compete. In addition, relaxing selection pressure on size and growth rate allowed for more emphasis on meat conformation. The result has been a very productive animal genetic resource that is also exquisitely adapted to its environment.

Provided by Phil Sponenberg.

Attention to the long-term effects of selection Cross-breeding with high-output international transboundary breeds has been promoted because gains in production can be often seen in a single generation. Response to within-breed selection is not as rapid. However, most locally adapted breeds have not undergone selection that specifically targets the production of commodities. In such cases, with a proper design, it is often possible to make reasonably rapid gains in production traits in the first few generations of a selection programme. Pure-bred selection programmes also usually provide more long-term security for the communities keeping the breeds than is provided by a cross-breeding scheme. However, this argument will not always be intuitively accepted, because the initial improvement in production in a pure-bred selection programme will usually lag behind the initial boost imparted by cross-breeding and the heterosis effects in the first cross. Part of the reason for the relatively large improvements that can be obtained by selecting within a locally adapted breed is the relatively high heritability of production characters, and the relatively low heritability of traits of adaptation and resistance. This means that more rapid and secure progress can be made by selecting a locally adapted breed for increased production than can be made by selecting for adaptation in a high-output international transboundary breed. Non-standardized locally adapted breeds are very likely to be more variable than high-output international transboundary breeds, and the highest performing animals can have great productive potential. Unfortunately, the mental image of a highly productive, temperate breed can sometimes overpower the long-term strategy of selection within a locally adapted breed, with breeders impatient for a quick response to high demand for livestock products.

CGRFA/WG-AnGR-7/12/Inf.6 Selection opportunities in relation to the state of the breed Breed census size affects the potential usefulness of selection. In the case of breeds with small population (i.e. critical or endangered status), it is difficult to undertake selection without creating potentially dangerous bottlenecks. Therefore, although a achieving an N e of 50 should be a short-term goal for an in vivo conservation programme, the long-term goal should be to exceed this threshold while making genetic improvement. Principles outlined in Section 6 should be carefully applied so as to maintain genetic variation in the long term. Mating decisions in more populous breeds should consider both population maintenance and selection for improved production. In developing countries, within-breed improvement programmes can contribute to improved income and livelihood of people who depend on low-input production systems. These breeding programmes must have outputs that are consistent with the producers objectives and aim to meet some market demand and thereby provide a return on producers investment in improving the breeding stock. The bottom line is that successful adoption of a technology (e.g. AI) depends on its feasibility and its compatibility with the needs of the livestock keeper and the production system. The technology has to be relatively simple, relatively cheap, and above all, involve relatively little risk. It is necessary to look at the production system holistically, and involve the livestock keepers at every stage in the planning and operation of the breeding programme, while integrating traditional behaviour and values (Van Arendonk, 2010).

Sustainable breeding Most locally adapted breeds must be selected for increased performance so that they become competitive in the production of standard commodities. The immediate economic needs of the owners demand this. Any pure-bred selection programme should also conserve the breed as a genetically, historically and culturally distinct animal genetic resource. Options for increasing performance must be carefully evaluated for long-term effects on the evolution of the breed;

the technical, financial and infrastructural requirements of implementing a breeding programme and the ability to maintain sufficient genetic diversity within the breed to ensure its sustainability. Effective ways to measure performance cheaply and accurately are important and often require creative strategies for animal identification and record keeping (see Boxes 45 and 46). However, systems that function in one production environment may not be feasible in another (see Box 47). The goal is a sustainable system that works to identify consistently those animals that are top performers in the local environment so that they can be selected for breeding and their contribution to the next generation can be ensured.

Box A simple recording system improves cattle fertility in the Bolivarian Republic of Venezuela Some large commercial beef ranches in the Bolivarian Republic of Venezuela have changed from measuring individual growth rates of calves to putting more emphasis on female fertility and longevity as greater contributors to overall herd productivity. One easy solution to monitoring fertility was to brand an X on the back of any cow failing to wean a calf in any year. No cow is allowed two X marks, as she is culled after failing a second time. With the record system being marked on each individual animal and easily readable in the field, the result has been increased fertility in commercial cow herds. Year of birth is also branded onto the animals, enabling easy evaluation of both longevity and fertility. Similar systems might include ear tags or ear notches for cattle and other species instead of branding.

Provided by Phil Sponenberg.

CGRFA/WG-AnGR-7/12/Inf.6 Box Animal evaluation by card-grading an example from the United Kingdom Animal shows are a good way to promote a breed and to generate interest among breeders. The show ring has also provided the traditional visual inspection method for simple evaluation of breeding animals, but this system has some drawbacks. First, it focuses attention on a small number of animals that are placed first in their class, and gives star status to the champion. As a result, these few fashionable animals frequently attract undue patronage by breeders, which concentrates their influence in the breed. The outcome is loss of within-breed diversity. In addition, the show ring often emphasizes traits that have questionable value for productivity and survival.

Since the 1980s the use of card-grading (see also Box 32) has been promoted in the United Kingdom by the Rare Breeds Survival Trust (RBST) for the evaluation of livestock animals. The card-grading approach is fairly simple and straightforward, but avoids concentrating attention on a few animals for breeding.

Purpose The purpose of card-grading is to classify a population into broad groups of potential genetic merit by visual inspection and thereby prevent domination by a single animal or small group of animals.

Procedure Animals are classified into four groups by the award of a coloured card: red card for above average;

blue card for average;

yellow card for below average;

and white card for disqualified animals.

Advantages Card-grading can be applied to any species of livestock and standards can be adjusted to obtain a visual scoring system that most accurately evaluates productivity and fitness. The proportions of animals likely to receive each card can be set at levels that allow loss of genetic variability to be avoided, i.e. by ensuring that several animals are likely to receive red cards. However, the standards should be set against a theoretical ideal, meaning that at some events perhaps no animals will receive red cards.

Limitations Card-grading is a visual inspection and therefore not a perfect guide to breeding ability, especially for traits that are strongly influenced by management and other environmental factors.

It is a subjective evaluation and relies on the expertise and conformity of graders.

Despite these limitations, selection based on card-grading has the potential to yield genetic improvement at a relatively low cost.

Provided by Lawrence Alderson.

Box Molecular selection not feasible for alpacas in Peru In Macusani, Puno, Peru the alpaca export market disrupted the local market for alpaca breeding stock and other products. This caused a change in breeding objectives. An attempt was made to change from traditional systems to more high-tech systems that used marker assisted selection and pedigree-based programmes. However, these approaches all failed in the local situation, because they were not sustainable in this remote region due to lack of infrastructure and lack of cultural familiarity with these techniques. Recapturing the previous traditions of visually classing males for breeding has helped to re-establish advances in the production of alpacas with high local value and appreciation.

Provided by Phil Sponenberg.

CGRFA/WG-AnGR-7/12/Inf.6 Objective: To develop a breeding programme for local breeds to increase production.


1. Assessment of the productive potential of local breeds;

and 2. Evaluation of non-market traits (e.g. adaption and longevity).


Strategies for increasing performance in pure-bred animals and, where relevant, of cross breeds;

A comparison of these strategies for their immediate effects on commodity production as well as their effects on long-term maintenance of adapted animal genetic resources for local food security;

and Analysis of the costs of the breeding programme, which should be kept as low as possible for low input low output breeds.

Task 1. Implementation of a pure-breeding programme with selection for production Action 1. Analyse the history of selection within the breed Information on the selection, exchange and use of sires, along with any records of gains from selection should be collected and analysed. Evaluate also the population structure and production potential of animals relative to the type of production system. Determine where the top-producing animals of the breed are located, and how they are being used.

Action 2. Decide on the production and other breeding goal traits that should be improved by breeding Decide which production traits will be measured. Clearly, a measurement of the yield of the product to be sold (e.g. meat, milk, fibre or eggs) is essential. In species with multiple offspring, number of offspring per pregnancy is also an important trait. However, profitability is affected both by production and by the costs of production. Functional traits that affect the cost of production, such as longevity, fertility, environmental adaptation, and ability to withstand stressors such as walking long distances to graze, may be as important as production. Defining these latter traits carefully can benefit well-adapted breeds (see Box 48). The guidelines Breeding strategies for sustainable management of animal genetic resources (FAO, 2010) provide advice on identifying important traits and determining breeding goals. Replacement and mortality rates can be used to identify superiority in adaptation, as can rebreeding intervals or litter sizes. Cost of rearing replacements is important, as is the quality and quantity of feed required and any requirements for other special management measures. Labour and veterinary costs should be included in the assessment, as should the financial return from the sale of products and offspring. Lifetime profitability is a key component. Adapted livestock are likely to have long productive lives, as well as multiple outputs, products, and services beyond the usual market commodities. Fertility and mortality are major traits. Smaller animals will frequently exceed the performance of larger ones for these traits (FAO, 2010).

Box Fleece quality as a sustainable breeding goal for sheep in Chiapas, Mexico More than 20 years ago, sheep production among the Tzotzil people in Chiapas, Mexico, changed from pure-breeding of local breeds to cross-breeding to enhance the production of meat. However, the Tzotziles do not consume sheep meat. This factor, as well as declining quality of fleeces for local textile needs, caused the incomes of the sheep producers to stagnate compared to those obtained under the traditional systems. At this point, it was recognized that performance for culturally relevant traits is a sound breeding objective. This led to the development of an effective open nucleus breeding system, based on selection for fleece quality, visual inspection and organization of ram distribution controlled by the local community. Attention to local practices ensured greater participation as well as enhanced economic return. The producers, the environment and the local culture all became beneficiaries of a well-thought-out sustainable system.

Source: Perezgrovas et al. (1997).

CGRFA/WG-AnGR-7/12/Inf.6 Action 3. Implement identification, registration and performance recording An appropriate system for identification and pedigree recording of individual animals must be developed and implemented, and important performance and functional traits should be measured. All traits must be evaluated in all populations to ensure that no important information is omitted from the final decision-making process. Deciding how to measure production is important, as different measures may yield different results and vary in efficiency. For example, measuring only first lactation records of milk production may select different animals from those selected by measuring lifetime production. Repeated measurements of the same trait will increase accuracy, but also costs.

Production measures should maximize total economic return. Factors such as longevity and cost of inputs can demonstrate the advantages of local animal genetic resources over imported ones.

Action 4. Recording and selection for production, fitness and body conformation traits should be implemented in relevant environments Improved commodity production can be antagonistic with respect to maintenance of traditional type in some breeds, especially when animals are adapted to difficult environments. Measures of productivity for such breeds should include productivity with low inputs in their natural production environments.

Traits associated with functionality, reproduction, survival and fitness should be recorded as well.

Lifetime productivity can indicate longevity and fitness, as a useful addition to measures such as growth rate or lactation yields per day. If survival in difficult environments is necessary, then adaptation to these environments must be taken into account in selection programmes. Unfortunately, recording and selection for traits associated with function and fitness is often more difficult than for production. Heritability tends to be smaller and traits are often more difficult to record. Innovative approaches may be needed for efficiency.

Action 5. Decide on the selection and breeding strategy that is most likely to succeed in improving production For conservation, the most common approach will be to apply a pure-breeding strategy. However, sometimes the productivity of the animals that have low genetic ability for production can be enhanced by crossing them with a more productive breed (see below). In this respect, it may be worthwhile to capture the value that a pure-breed has in providing hybrid vigour to cross-bred offspring that can be marketed while the pure-breed is maintained (FAO, 2010).

Optimizing selection response and genetic variability within small populations Rationale Response to selection As noted throughout this section, one of the options for increasing the probability of survival of an endangered or vulnerable breed is to improve its profitability. Increasing the productivity of a breed will usually make it more profitable and therefore increase its chances for self-sustainability.

However, improving a populations genetic ability for productivity and maintaining its genetic variability (i.e. high N e ) are antagonistic processes. Some compromise is required.

Classical theory on the response to artificial selection states that the selection response or gain (G) in the mean value of the trait per year can be calculated using the following equation:

G = i/L, where i is the selection intensity, the correlation between the estimated and the real breeding value of the individuals (also known as the accuracy of selection and equal to the square root of heritability when selection is based on phenotypes), the additive genetic standard deviation for the trait (i.e.

genetic variation) and L the generation interval. Consequently, to obtain greater responses, the values of i, and should be increased and the value of L reduced.

CGRFA/WG-AnGR-7/12/Inf.6 Maintenance of genetic variation vs. response Selection intensity is a measure of the pressure put on the population and is related to the ratio of selected animals to candidate animals in the population. Larger values are obtained by selecting fewer individuals as parents for the next generation. However, this practice will reduce N e, an outcome that is in conflict with the main objective of a conservation programme, and will lead to higher levels of inbreeding and reduced genetic diversity. More accurate estimates of breeding values (i.e. increased ) are often obtained by using information about relatives in addition to the individual phenotypes. This strategy will lead to the co-selection of relatives, especially for traits with low heritability, contributing again to the loss of diversity and an increase in inbreeding. Short generation intervals will also increase the gain but, as described above, also increase the amount of genetic variability lost per year. Recall that in Section 6 (Task 1, Action 3) increasing the generation interval was suggested as a means of increasing genetic variability, highlighting the trade-off between genetic improvement and the maintenance of variation. The presence of in the numerator of the selection-response equation gives another reason for maintaining genetic diversity (i.e. high ) in the breed, as response is greater when is larger and no response for a trait can be obtained if there is no genetic variation. In summary, all the actions that can be taken to improve the gain in response oppose the general objectives, from the genetic point of view, of a conservation programme. Consequently, some balance must be established between the various forces.

Objective: To improve the productivity of a breed while avoiding the loss of genetic variability as much as possible.


1. Knowledge of the following characteristics of the breed to be conserved:

size of the population reproductive capacity of the species characteristics of the production system 2. Awareness of countrys livestock development objectives and existing and potential markets for animal products.


Agreement among stakeholders with regard to traits to be improved and relative importance of genetic gain and maintenance of diversity;

Clearly defined selection goal in terms of the trait(s) to be improved;

and A general breeding plan that will optimize genetic improvement and will maintain genetic diversity.

Task 1. Adopt a general breeding strategy to maintain the conserved breed Action 1. Determine which trait or traits are to be improved in the conserved breed The determination of the objective of selection (i.e. the breeding goal: the trait or traits we want to improve in the population) has to be done in consultation with stakeholders. This process is described in more detail in the guidelines Breeding strategies for sustainable management of animal genetics resources (FAO, 2010). This evaluation could be done in conjunction with the studies conducted to investigate the conservation values of the breed (Section 3). If the presence of a particular characteristic has been identified as an important justification for maintaining the breed, this characteristic should obviously be included in the breeding goal, because reducing performance for that trait would diminish or remove the justification for maintaining the breed. If this characteristic is a qualitative trait, it is important to ensure that selection to improve other productive traits does not cause the characteristic to disappear from the population.

The ability to provide products for a specific niche market can make animals more valuable (see Section 8). If a niche market is to be targeted, the trait(s) that will affect the breeds competitiveness in this market should be identified. For example, if milk from a given breed is going to be used for the manufacture of a particular type of cheese, the traits selected for should include not only the amount of milk produced, but also the quality, in terms of milk protein and fat content, as well as (if possible) traits related to cheese production. To derive a breeding goal, it is necessary to determine for each trait the CGRFA/WG-AnGR-7/12/Inf.6 increase in profit that will be obtained when the trait is improved by one unit. This increase in profit indicates the relative value of each of the traits, which can be summed up to form the breeding goal.

When feasible, a selection index should be created with traits that are measurable and that correlate as highly as possible with the breeding goal. Ideally, breeding goals should be kept as simple as possible so as to ensure that the really important traits are improved. Secondary traits can initially be accounted for simply by requiring that the breeding animals meet minimum acceptable levels for each trait. These traits can be formally incorporated into the selection index at a later point when the selection programme is well established and the population has increased its census size. Culling of affected animals for control or elimination of genetic defects may be an example of selection for a secondary trait (see Box 49).

Box Selection to eliminate genetic defects Genetic defects tend to be more common in populations with low genetic variability. Populations at the start of conservation programmes may show genetic defects at frequencies of greater than 10 percent. Consequently, in addition to developing a breeding programme for other traits, explicit actions have to be taken to remove the genetic information that provokes the disease, or at least to reduce the frequency of deleterious alleles to a reasonable level.

The effectiveness of strategies to remove genetic defects will be affected by the nature of the genetic determination of the particular defects. Genetic defects are often controlled by a single gene.

In such cases, the inheritance of the disease and the detection of carriers of the deleterious alleles are simple. In many cases the deleterious allele is recessive and thus only expressed when in homozygous form (i.e. in a double copy). The manifestation of such defects is more common in small populations (particularly those with a small N e ) because homozygosity is increased when genetic variability is decreased. When the defect is recessive, many individuals (heterozygotes) will carry the allele but not show the defect. Genealogies may be used to identify individuals with a high probability of being carriers. To eliminate the defect, first animals showing the disease and then carriers should be avoided as parents of the next generation, as long as the programme is not compromised by a too large a reduction in the number of breeding individuals. If a DNA test exists for the gene responsible for the defect, individuals can be genotyped and carriers unambiguously detected and excluded from the breeding programme.

When the trait has a polygenic determination and behaves as a quantitative trait with different degrees of expression of the disease, a regular selection programme should be implemented to eradicate the defect from the population. In any case, it should be stressed that the measures taken to eliminate the defect should include restrictions on the loss of genetic diversity so that the breed avoids troubles brought about by a rise in inbreeding. For example, it may be necessary to slow down the eradication of a recessive allele (Sonesson et al., 2003).

Obviously, not all defects are genetically determined and selection, and breeding will not influence the occurrence of such defects.

Action 2: Agree upon the acceptable rate of inbreeding in the conserved population The acceptable rate of inbreeding per generation (F) will depend on the status of the population and the characteristic of the species. For highly endangered breeds, the values proposed in Section 2 can be used for pure conservation programmes, i.e. F 1 percent (assuming this is possible). As long as the population is not in the critical or endangered categories, restrictions can be relaxed and a larger F can be chosen. In commercial breeds, there is a general consensus that the maximum acceptable F is about 2 percent, but the figure may vary between species. Remember that the more emphasis given to the maintenance of diversity, the lower the response obtained on the selected trait, and vice versa. One option is to predict the expected gain for a range of F and choose the compromise solution that best meets both objectives.

Task 2. Design a breeding programme that generates genetic improvement while maintaining genetic variability Action 1. Evaluate the circumstances under which the breeding programme will be applied CGRFA/WG-AnGR-7/12/Inf.6 Various actions can be taken to achieve some selection response while maintaining genetic variability at an acceptable level. The appropriate measures will depend on the species, the production system, the ownership of animals and level of central control of breeding decisions, the level of cooperation among breeders, the availability of technical capacity and infrastructure, and various other factors, Action 2. Consider the various options for balancing genetic improvement and maintenance of genetic variability Numerous options have been proposed for increasing genetic variability when applying selection. Five options are presented below, roughly in order of increasing complexity.

Option 1. Determine the ideal number of parents when applying selection The first approach to the control of inbreeding during selection is to determine the number of males (N M ) and number of females (N F ) that would give the desired (acceptable) rate of inbreeding (F) and then select the best N M males and the best N F females, according to the selection goal. Then each of the selected animals should contribute the equal number (i.e. equal within sex) of offspring to the next generation. The desired number of animals of each sex (according to the F desired) can be obtained by using the formulae presented in previous sections, such as F = 3/(32 N M ) + 1/(32 N F ) (Gowe et al., 1959).

The process of selecting the best animals to be parents is called mass selection. This type of selection can also be called truncation selection, because it involves selecting all animals above a certain threshold or truncation point. In this case, the truncation point for males (females) is the selection criterion, such as phenotype or estimated breeding value, of the N M th highest male (N F th ranked female).

Option 2. Apply within-family selection A simple and effective way to control the F while improving the genetic potential of a breed for a productive trait is to implement within-family selection. As explained in Section 6, within-family selection consists of selecting one male from each sire family and one female from each dam family (i.e. each sire is replaced by one of his sons and each female by one of her daughters). Following this strategy, the population maintains a larger N e than it would with random contributions (see Table 9), but there is still some room for selection. Instead of choosing a son or daughter at random from each family, the best animal(s) in each family for the traits of interest is(are) chosen, thereby obtaining some gain in the traits. The selection intensity will depend on the size of the families, and will thus vary by species. However, the rate of gain will not be exceptionally large in any species, because this approach exploits only the within-family variability and ignores the genetic differences between families. Nonetheless, within-family selection is a sensible and easy way to achieve low F in selection programmes.

Option 3. Apply family selection The opposite of within-family strategy is family selection, a method in which all selected individuals are taken from the family (or group of families) with the highest average trait value. This method provides greater response than within-family selection, but also leads to greater losses of diversity and higher F, as all the selected animals are close relative.

In reality, a wide range of options ranging from complete within-family to complete family selection can be considered. For example, Table 9 illustrates a hypothetical situation in which a breed consists of eight families, each of which has four males and four females, from which a total of eight animals of each sex have to be selected based on individual genetic values and/or family means for the traits of interest in the breed. The two extreme options are:

1. selecting the best single individual of each sex from each family (represented by row 1 in Table 9), or, 2. selecting the best animals from the two families with the highest mean value (Option 2).

CGRFA/WG-AnGR-7/12/Inf.6 But there are several intermediate solutions that will differ in the response they yield and the N e they imply. All solutions have to be tested to find the one that yields the desired F.

Table 9. Different ways of selecting individuals from eight families and the expected responses to selection and inbreeding (in percentage) that they imply Distribution of family sizes Options Male/female pairs taken from each family (n) Response F (%) 1 1 1 1 1 1 1 1 1 5.90 7. 2 4 4 0 0 0 0 0 0 17.42 42. 3 4 3 1 0 0 0 0 0 18.17 35. 4 4 2 2 0 0 0 0 0 17.87 33. 5 4 2 1 1 0 0 0 0 17.78 30. 6 3 3 2 0 0 0 0 0 17.30 30. 7 3 3 1 1 0 0 0 0 17.21 27. 8 4 1 1 1 1 0 0 0 16.38 27. 9 3 2 2 1 0 0 0 0 16.91 24. 10 3 2 1 1 1 0 0 0 16.24 21. 11 2 2 2 2 0 0 0 0 14.91 21. 12 2 2 2 2 1 1 0 0 14.85 18. 13 3 1 1 1 1 1 0 0 14.23 18. 14 2 2 1 1 1 1 0 0 13.56 15. 15 2 1 1 1 1 1 1 0 10.83 11. Source: Toro and Prez-Enciso (1990).

It must be emphasized that the real number of animals that need to be kept to ensure the desired N e is affected by a combination of factors that include selection model, mating ratio and the size of the families (see Table 10).

Table 10. The minimum number of sires to be used per generation to achieve an effective population size of 50 or more, for different mating ratios and expected family sizes, and assuming h2 = 0. Mass selection Within Mating Lifetime offspring Random family ratio 4 8 12 16 20 36 selection selection 5 21 23 25 27 28 30 15 4 to 5 21 25 27 28 29 32 16 3 to 4 23 26 28 30 31 35 17 2 to 3 25 29 32 34 36 40 19 1 to 2 31 38 43 46 48 55 25 Source: Woolliams (2007).

Option 4. Implement weighted selection Notice that implementing Options 2 and 3 reduces F at the cost of a lower response than is obtained through mass selection with a fixed number of males and females selected (Option 1). Ideally, it would be desirable to control F without losing response. According to the rules of strict truncation selection, the selected individuals should contribute the same number of offspring to the next generation.

However, if this condition is relaxed and differential contributions are allowed, more individuals can be selected without losing selection intensity, and a larger N e obtained (see Box 50 for an example).

This is possible because the best individuals are allowed to contribute relatively more, with their contribution proportional to their genetic value (phenotype or estimated breeding value). This methodology is called weighted selection, because more weight is given to the better individuals.

The disadvantages of weighted selection are the need to keep more individuals as selection candidates, which implies increased costs for the maintenance of these extra animals, and somewhat greater complexity than strict truncation selection.

CGRFA/WG-AnGR-7/12/Inf.6 Box Weighted selection an example A recent study by Moreno et al. (2011) used simulated data to compare weighted selection versus truncation selection in a small population (32 animals of each sex). Under truncation selection, the 32 individuals of each sex were evaluated per generation and 8 were selected as parents. Each selected individual contributed four sons and four daughters to maintain the census size of the population. This process resulted in a selection intensity of 1.235 and the N e was 19.8. When weighted selection was implemented, the optimal scheme corresponded to selecting the best individuals of each sex, but the number of offspring obtained from each of them was allowed to vary. Specifically, the 12 selected animals were allowed to produce 6, 4, 4, 3, 3, 3, 2, 2, 2, 1, 1, and 1 offspring each, respectively, ordered from highest to lowest in genetic value (i.e. the best animal of each sex produced six offspring, whereas the 12th best animal produced only a single offspring).

In this scenario, the selection intensity was exactly the same as in the truncation selection scenario (1.235), but N e was nearly doubled (31.5), as more individuals contributed offspring.

Source: Moreno et al. (2011).

Option 5. Apply optimum contribution strategy for selection Weighted selection determines particular individuals contributions to the next generation based exclusively on their genetic value for the selected trait(s). However, the simple approach described in Box 49 is only optimal if the genetic relationships between animals are equal for all pairs. This condition is not realistic in animal breeding, as differences in the relationships between pairs will almost certainly be present. When pedigree information is available, a superior solution, called the optimum contribution strategy is possible. The optimum contribution strategy accounts for the coancestry of candidates as part of the decision criteria and is thus a logical approach to minimizing inbreeding for a given level of genetic response. This methodology is recommended as the most powerful way of dealing with genetic gain and inbreeding at the same time (Meuwissen, 1997). The aim of this approach is to vary the numbers of offspring produced by selected individuals so that they are proportional not only to their genetic value for the selected trait (as with weighted selection), but also to their degree of relationship with the rest of the population.

In the animal breeding context, the degree of relatedness is usually expressed as the additive relationship, which is twice the coancestry between any couple of individuals. Following the optimum contribution strategy, if there is a group of relatives that have high values for the trait of interest, not all of them will be allowed to contribute offspring. Not surprisingly, however, as was the case with minimum coancestry contributions (Section 6), the implementation of optimum contributions requires a highly controlled production system, several generations (at least four) of complete pedigree information, and the use of complex mathematical procedures (see Box 51).

Action 3. Implement and monitor the chosen breeding programme Once the desired programme is chosen, extensive cooperation with breeders and other stakeholders will be required for implementation. All of the options listed in Action 2 will require recording of performance information for the traits upon which selection will be based and all but Option 1 require some pedigree data (knowledge of parents as a minimum).

CGRFA/WG-AnGR-7/12/Inf.6 Box Optimum contributions strategy of selection To better account for the two opposing forces, genetic response and genetic variability (F), both of them should be included in the objective function but with opposite sign (+response and -F). The expected mean value for the selected trait of the next generation components can be estimated as the product of the value of parents by the number of offspring they contribute. Expected inbreeding is calculated by multiplying contributions and coancestries. Therefore, the objective function to optimize is c i v i - c i c j f ij, where c i is the contribution of individual i, v i is its genetic value for the selected trait and f ij is the coancestry between individuals i and j, which is considered for every possible pair of animals. In practice, the term regarding F is treated as a restriction, and the algorithm searches for the solution (i.e. combination of offspring contributed by each individual) with the highest G but not exceeding the desired value for F. Several methods have been proposed to solve this optimization problem, all of which require the use of computer programs. The program EVA 15 (Berg et al., 2006) is one of the software available to manage a selection programme with restriction on the inbreeding levels.

Cross-breeding for enhanced production Rationale The potential of cross-breeding The use of cross-breeding as part of a conservation effort may seem counterintuitive, but it can be a valuable option in certain situations. The concept of using limited cross-breeding for genetic rescue of an extremely endangered population with small N e is introduced in Box 38. There are, however, other instances where cross-breeding may be able to play a role in a conservation programme. Cross-breeding can be particularly beneficial when the objective of the conservation programme is to use the beneficial genes of a breed at risk without having to obtain high economic returns from the pure-bred population.

Cross-breeding provides the opportunity to combine the genetic characteristics of different breeds. It is recommended when there are multiple breeding-goal traits that have antagonistic genetic relationships, such as between production and fertility or between production and quality of the product. It can be difficult to improve such traits simultaneously in a single breed. For example, combining the adaptive traits of a locally adapted breed with the production traits of an introduced international transboundary breed might be attractive. However, cross-breeding is only effective and sustainable if the breeding system is carefully chosen and well planned. The breeds used should be available in the long term and the plans should be strictly followed by the livestock keepers. A section in the guidelines on Breeding strategies for sustainable management of animal genetic resources (FAO, 2010) is devoted to cross-breeding.

Cross-breeding strategies One simple strategy for maintaining local animal genetic resources is to apply cross-breed to the non recorded and low-producing surplus local females, while maintaining pure-breeding (and within-breed selection) among the best animals of the local breeds. This procedure of limited and targeted cross breeding not only saves the more productive well-defined local breeds but also maximizes the contribution of the lower-producing animals to local commodity production and food security.

Breeds that have been characterized for production potential and determined to be economically sustainable should be managed through breeding systems with the purpose of enhancing their productivity, and their female animals should generally not be used for cross-breeding. These breeds should instead be improved through selective breeding within the pure breed. Breeds that have been characterized as low producing, or those that are already crosses of local breeds, are logical candidates to have their production improved by using genetically superior germplasm either from local well http://eva.agrsci.dk/index.html CGRFA/WG-AnGR-7/12/Inf.6 defined improved breeds or from international transboundary breeds that are relevant for the local production systems. The decision to cross-breed should then be based on economic factors (costs versus expected returns) and degree of agreement by the keepers of the local animals.

Unregulated and unmonitored cross-breeding can rapidly erode the numbers and genetic integrity of any breed that is used widely for cross-breeding. The utility of many breeds comes specifically from their role in organized cross-breeding systems, so attention must be given to maintaining a sufficiently large and well-managed pure-bred population to ensure the continued availability of animals for the cross-breeding system.

Information required for planning cross-breeding systems When a breed is used in a cross-breeding programme, specific information should be collected during breed surveys and characterization. Useful information about the breeds role in cross-breeding includes population numbers and the current proportion of pure-bred as opposed to cross-bred breeding. Recording the number of females mated pure quickly captures this aspect of the breeds dynamics. The data collected should also include the ultimate fate of cross-bred and pure-bred offspring, and whether these are terminal (i.e. marketed without producing offspring) or used for further reproduction. Assessments should include the relative quality (high, medium, low) of the animals used in pure-breeding and those used in cross-breeding. It is important to describe the role of each sex in the cross-breeding production system (e.g. are males used for cross-breeding with other breeds or are females used in this role?). Ideally, pure-bred populations will be undergoing selection for enhanced performance as measured in both pure-bred and cross-bred offspring.

Implementing cross-breeding systems The planning and implementation of a cross-breeding programme should be based upon a clear understanding of what is wanted as an outcome of the programme. If the objective is to increase production in a local breed, cross-breeding with an introduced breed may be considered. A fairly common and simple approach used to improve production is to cross a local breed with a high-output international transboundary breed. This can be done with one of two goals in mind, 1. replacing the local genetics, i.e. by making continual successive crosses to the introduced breed, or 2. upgrading the local breed, i.e. by crossing to the introduced breed until the population contains a high proportion (usually >75 percent) of introduced-breed genetics.

The replacement strategy is clearly not conservation and frequently fails in tropical or other environmentally stressful regions because the resulting animals are less well adapted to the local conditions than the original population were. Thus, before embarking on a strategy that will lead to the elimination of local animal genetic resources through replacement breeding, the consequences of such a strategy must be thoroughly investigated. Local animal genetic resources can usually make a valuable contribution to the local production system in the long term, in which case their survival and availability must be ensured. At the very least, the local animal genetic resources to be replaced in vivo should be cryoconserved. Breed introduction should usually not even be considered unless the enhanced production (locally realized and not only potentially possible) can be expected to be at least 30 percent greater than that obtained from the pure local breed (FAO, 2010). When this is the case, a system that involves producing F1 animals and conserving the pure local population should receive primary consideration. As noted above, a sound strategy is to cross-breed the relatively lower-producing portion of the local population and to reserve the most productive local animals for pure-bred breeding.

According to Schmidt and Van Vleck (1975) two main classes of cross-breeding system can be distinguished:

1. systems that require maintenance of the pure-breeds (pure-bred and rotational crosses);

and 2. development of a new (synthetic) breed by systematically mating cross-bred females and cross-bred males.

CGRFA/WG-AnGR-7/12/Inf.6 Pure-bred crosses Pure-bred crosses involve the mating of pure-bred animals from different breeds for one or two generations so as to produce cross-bred animals that terminate the breeding system. Such strategies are generally defined by the number of breeds involved:

Two-way crosses: individuals of two pure breeds are mated and the offspring are used only for production (i.e. not for breeding). For example, the dairy cows with the lowest breeding values for milk production in a herd are not selected as dams for producing replacement dairy animals, but are mated to a bull from a beef breed to produce offspring that have better capacity for beef production than pure dairy calves.

Three-way crosses: two-way cross females are mated to a sire from a third breed to produce offspring used for the production goal. For example, in pork production, two breeds with high fertility and maternal traits are occasionally crossed and cross-bred sows then mated to a sire from an excellent meat-producing breed to obtain a large number of piglets with the characteristics desired for pork production. Sometimes the two-way cross females are mated back to a sire of one of the parent breeds this is known as a backcross. Sexed semen can be used to enhance such a cross-breeding programme if animals from one sex are more desirable for production purposes (see Box 52).

Four-way crosses or double two-way crosses: two-way cross females are mated to two-way cross males to produce the animals used for the production goal. For example, this type of cross-breeding is the preferred breeding method employed by multinational breeding companies for specialized egg and broiler production.

Box Effect of sexed semen in producing a final cross in dairy cattle The availability of sexed semen in dairy cattle has been eagerly anticipated for many years, and recent developments in fluorescence-activated cell sorting have brought this technology to commercial application. In recent years, a number of AI companies have started to offer sexed semen to their farmers. Semen sexing provides the potential to increase the numbers of offspring of one sex in a closed population, thereby increasing the intensity of selection for that sex. Semen sexing enhances the farmers' ability to obtain a larger number of replacement heifers from their own herds. In a herd with a stable herd size, semen sexing could be used to breed replacement heifers from the cows with the highest genetic merit. This will create a one-time lift of the genetic level of the herd. The largest economic benefit of using sexed semen in pure-bred herds would come from the ability to use the remaining dairy cows for the production of cross-bred animals for meat production Semen sexing can be used to increase the efficiency of producing F1 dairy hybrids. For an F1 scheme to be sustainable, part of the pure-bred population needs to be mated to bulls of the same breed to produce replacements.

The number of cows that need to be mated for breeding replacements can be nearly halved by the use of sexed semen. In addition, the number of F1-females that are produced can be nearly doubled by using sexed semen. In other words, the number of pure-bred cows that need to be kept for the production of F1 hybrids can be reduced by 60 to 75 percent, depending on the sex ratio resulting from the use of sexed semen. The economic benefit of this reduction is largest when pure-bred cows and cross-bred cows are competing for the same resources. Benefits are smaller in a stratified cross breeding system, such as that used in Brazil where dairy farms buy replacement F1 females, than in the poultry or pig industry. The replacement females are produced in areas where land is less expensive, using Holstein semen on Brazilian dairy zebu breeds.

Source: Van Arendonk (2010).

When implementing two-breed crosses involving the use of imported animal genetic resources, the use of the local breed as the source of pure females and the exotic breed as the source of sires is strongly recommended. Two-way crosses require only a limited number of sires, so maintaining a population CGRFA/WG-AnGR-7/12/Inf.6 solely for the production of males could result in a greatly reduced census size, which would increase the risk of extinction.

Breaking the cycle in which low census numbers limit the potential for within-breed improvement is difficult. Small population sizes limit the selection intensity that can be applied and/or increase inbreeding. If a higher census size can be coupled with good record keeping and selection, then progress can be made in increasing productivity, which will subsequently increase the breeds value and help secure its sustainability in commercial settings. If the only perceived value of a breed is as a component of a cross-bred population, then securing the breed in sufficient numbers for pure-bred selection will be difficult. Moreover, breeds with small population sizes are likely to be overlooked as resources for commercial purposes, and therefore remain in low numbers and at risk of extinction (see Box 53).

Box Two-tiered demand for Criollo Saavedreo cattle in the Plurinational State of Bolivia The use and conservation of Breeds that excel in cross-breeding involve complicated issues.

Temperate breeds such as the Holstein and Brown Swiss have been imported into the Plurinational State of Bolivia in an effort to increase milk production, but pure-bred cows belonging to these breeds have had difficulty surviving in the environment of the Bolivian Tropical Lowlands. To address this problem, the Criollo Saavedreo cattle breed was created under the guidance of Dr John V. Wilkins from the British Tropical Mission in Bolivia with the purpose of providing bulls to be mated to the temperate cows to produce offspring better adapted to the local conditions. The Criollo Saavedreo was created by selecting bulls from Criollo breeds throughout Latin America that had already been selected for improved milk production.

While the development of this breed was successful and the Saavedreo bulls meet with a brisk demand for use in cross-breeding, the pure-bred cows are much less in demand than the cross-bred cows. This has meant that the number of pure-bred animals has remained relatively low (some few hundred head mostly at a single government installation). As a consequence, selection within the breed remains somewhat lower than that achieved in breeds that have larger population sizes.

Production is unlikely to diminish, but selection differentials are unlikely to be high enough to quickly increase genetic merit for productivity. This situation gives rise to a cycle in which low population size prevents within-breed progress in selection for production, which in turn ensures that the population remains small.

Provided by German Martinez Correal.

Rotational crosses There are three general types of rotational crosses:

Crisscrosses: a two-way cross female is mated to a sire of one of the two breeds used to produce the original two-way cross and their female offspring are mated to a sire of the other breed. This alternating pattern of sire-breed usage is then continued in subsequent generations.

Three-way rotation: Sires of three breeds are used in successive alternating generations on the cross-bred dams of the previous generation.

Multibreed rotation: such rotation schemes can be extended to the use of four breeds (four way rotation) or to the continued use of sires from new breeds (indeterminate rotation).

An advantage of rotational crosses is that they do not require exchange females between herds or villages, which decreases costs and reduces disease transmission. Only the sires of the breeds involved have to be purchased by the owners of the females. AI will eliminate even the need to purchase the sires. Another advantage is that the rotational systems maintain high heterosis, 67 percent with a two breed rotation and even more when additional breeds are involved. A disadvantage is that the producing and reproducing offspring will eventually represent different generations and therefore different combinations of breeds and thus can show high variation in phenotype. Also, if one of the breeds is an international transboundary breed, maintaining the rotation may require continual CGRFA/WG-AnGR-7/12/Inf.6 importation of new germplasm. Rotational schemes involving a high number of breeds can be problematic in terms of monitoring and require the availability of a wide variety of germplasm. Three way rotations may be the most efficient compromise.

Composite breeds In some unfortunate cases, either the population size of a breed at risk will be too small to avoid an extinction vortex (see Section 6) or its production potential will be too low to justify the establishment of a breeding programme for its conservation. In such cases, an option that may be considered is to sacrifice the breed as an independent entity, but conserve its genes by crossing it with another breed (or breeds) to create a new composite breed (also known as a synthetic breed). If two or more breeds are at high risk of extinction, they can be combined together to form a composite breed. Box describes how the genetics of now-extinct cattle breeds in Sweden have contributed to contemporary populations. Bennewitz et al. (2008) have proposed a method (based on genetic markers) for determining which breed to match with an at-risk breed so as to conserve the maximum diversity among all the breeds within a country. Complementarity of phenotypes may also be used as a basis for matching breeds, especially in situations where molecular information is not available.

Box Genes of extinct Swedish cattle breeds conserved in todays populations The data for Sweden in DAD-IS at the end of 2011 listed 22 cattle breeds, including four that are classified as extinct. Although animals of these four breeds can no longer be found, the history of the breeds suggests that many of their genes are conserved in current populations. During the late nineteenth and early twentieth centuries, the populations of three of these breeds, the Herrgrd, Smland and Skne, were grouped together and used to form the fourth breed, a composite called the Rdbrokig Svensk Boskap (RSB or Red-pied Swedish). The RSB continued to evolve as well and in 1928 this breed was merged with the Swedish Ayrshire to form yet another composite, the Svensk Rd och Vit Boskap (SRB). Since then, although its name has not changed, the SRB has remained dynamic, incorporating and contributing genes to and from similar breeds in other Scandinavian countries.

Source: Bett et al. (2010).

Many composite breeds have been developed in the past 50 to 100 years (e.g. Shrestha, 2005). One particularly common strategy for tropical environments has been to create composite breeds by inter se mating of cross-bred animals that have resulted from an initial cross of an international transboundary breed to a locally adapted breed (or more complex combinations of more than two breeds). Selection usually stabilizes the exotic inheritance at around 50 percent, because in most cases any exotic influence above this results in a decline in most important economic traits because of poor adaptation. The long-term objective of producing a composite breed should be to stabilize the proportions of the foundation breeds to achieve a combination that is well adapted to the local production environment. Although formation of a composite breed can effectively conserve the genes of a breed at risk and yield a new genetic resource of potentially higher value and sustainability, the process is not simple and has its disadvantages and potential pitfalls (see Box 55).

Cross-breeding and conservation All three of the cross-breeding systems (pure-bred crosses, rotational crosses and development of a composite breed) may be considered when developing a cross-breeding programme for conservation, although the composite breed approach is likely to involve the loss of at least one breed during the creation of another. It must be stressed that any cross-breeding approach requires a great deal of management to achieve the desired results. Indiscriminate cross-breeding is a major threat to local breeds (Tisdell, 2003) and has often yielded unsatisfactory results in terms of increasing productivity.

CGRFA/WG-AnGR-7/12/Inf.6 Box Potential difficulties and pitfalls in the development of composite breeds Increased complexity in the initial years of a composite breeding programme (i.e. before a stabilized population is reached), animals of different generations may be present within the same herd or other breeding group. Animal identification and pedigree recording are necessary in order to ensure that animals with the desired proportions of each breed are mated. This factor is especially important when more than two breeds are involved or if the desired final proportions differ from 50 percent per breed.

Decreased uniformity with matings involving cross-bred parents, proportions of the genes of foundation breeds in the offspring can theoretically range from 0 to 100 percent, resulting in a wide variation in appearance and performance.

Decreased productivity heterosis in matings of cross-bred parents will usually be less than in crosses of pure breeds. Thus, performance may appear to decrease in the generations between the F and the stabilized composite, which may disappoint and discourage breeders.

Need for pure-bred populations ideally, the pure-bred foundation breeds should remain available for infusion of genetic diversity if needed. However, this will be impossible in situations where the breed at risk is entirely integrated into the composite breed. Cryoconservation may be an option that allows this problem to be overcome.

Loss of cultural value although the genes of breeds that are exclusively conserved in a new composite breed will be maintained, the breed itself will cease to exist and, therefore, some of the cultural significance of the breed is likely to be lost.

Ambivalence of breeders unless they are closely involved in the planning and enthused about the idea, breeders loyalty to the breeding scheme may be lower than it would be to a scheme based on a breed that they have a long history of keeping. They may therefore be more inclined to abandon the programme if success is not readily apparent. Alternatively, if close involvement gives them a sense of ownership of the breeding scheme, they may be proud of it and regard themselves as pioneers and innovators.

Objective: To develop a stable cross-breeding system that conserves an animal genetic resource.


1. A breed at risk for which development of a cross-breeding programme is a viable option for conservation;

2. Information about the breed at risk, including its population size and risk status, its strengths and weakness, and the opportunities and threats that may affect its long-term sustainability;

3. A description of breeds production system(s) including markets for products;

and 4. Inventory and characteristics of other relevant local breeds and exotic breeds. This should include the breeds production characteristics as well as breeding programmes for their maintenance and improvement, and their roles in cross-breeding systems.


A sustainable cross-breeding programme for maintenance of an animal genetic resource, either as a pure breed that contributes animals to a subsequent cross, or by incorporating beneficial genes into a synthetic breed.

Task 1. Develop a system for cross-breeding to conserve an animal genetic resource Cross-breeding programmes that are not well planned are likely to fail, or at least not to reach their desired objectives. A comprehensive plan for should thus be devised before commencing any cross breeding activities. The National Advisory Committee on Animal Genetic Resources may take responsibility for this plan, or may choose to form a special ad hoc committee. The committee should include key stakeholders.

CGRFA/WG-AnGR-7/12/Inf.6 Action 1. Outline the desired outcomes of the cross-breeding system The primary goal of any conservation programme will be to maintain the targeted animal genetic resources (as pure breeds or in the form of their important genes). Secondary objectives that support this main goal should be formulated. Examples of such secondary objectives may include improving the livelihoods of the livestock keepers and meeting local demand for the products of the animals. The products to be produced should be considered and overcoming constraints of the production systems must be accounted for. For example, the cross-bred animals will ideally have a greater genetic potential for production, but they may also have a greater demand for inputs. Factors that may restrict availability of such inputs may limit the feasibility of the cross-breeding programme.

Action 2. Evaluate the status of the targeted breed The activities described in Sections 1 to 3 will provide most if not all of the information needed to make informed decisions on the establishment of a cross-breeding programme for conservation. Among the most important pieces of information are the census size of the population and its N e, the breeds strong and weak traits, and the particular threats to its survival. Awareness of the breeds main stakeholders and some indication of their willingness to participate in a cross-breeding programme are also crucial.

Action 3. Evaluate the other breeds that are potentially available for inclusion in the cross-breeding plan Cross-breeding will only be viable if genetic material from the complementary breeds in the cross is readily available in sufficient and sustainable quantities. A list should be made of all such breeds. Both other breeds from which live animals are available, and breeds for which only semen is available should be considered. The adaptability and productivity of these breeds in the local production environment should be determined through literature review and/or studies that document their phenotypic characteristics and performance levels (FAO, 2010). Special attention should be given to unique genes or traits that affect the complementarity of these breeds with the breed targeted for conservation.

Action 4. List the cross-breeding systems that are relevant for the production system A critical initial decision will be whether or not the breed targeted for conservation can realistically be maintained as a pure-bred population. All pure-bred crossing and rotational systems require the maintenance of a population of pure-bred animals. Pure-bred crossing systems will require the largest populations, because they require the availability of two groups of females: one to maintain the pure population and another to produce F1 animals. Rotational cross-breeding systems will generally only require the production of sires (or access to preserved semen) to provide germplasm for cross-breeding.

As described in previous sections, the N e of the pure-bred population should be 50, excluding the pure-bred females that are crossed to produce F1 animals in pure-bred crossing systems. Larger population sizes are preferable, obviously, to allow for greater selection within the pure breed. When N e is significantly less than 50, incorporating the population into a synthetic breed may be the most practical option.

Maintaining pure-bred populations will also require the availability of stakeholders (either livestock keepers or government institutions) that are willing to maintain the breed, even though its production potential will likely be less than that of cross-breeds.

Action 5. Describe the function of the target breed and complementary breeds in the cross-breeding system Attributes of local breeds that can be exploited through cross-breeding usually include characteristics such as disease resistance and stress resistance, quality and composition of animal products, adaptation to particular environments or production systems, and the ability to utilize coarse roughage and crop residues. The complementary breeds often are chosen to increase production.

Action 6. Choose the optimal cross-breeding system Develop cross-breeding to enhance the performance of low-producing local animals, and reserve high producing local animals for use in pure-breeding systems. Establish protocols that ensure some use of high-producing females from the target breed for pure-bred breeding. Select a group of breeding males that can be widely used locally. Determine how many breeds (usually between two and four) are needed to attain the final mix of traits that will provide the desired economic performance. Also determine the gender of the animals to be contributed by each breed.

CGRFA/WG-AnGR-7/12/Inf.6 Action 7. Present the plan to a wider group of stakeholders for final approval Although various stakeholders, including key livestock keepers, should be intimately involved in the planning of a cross-breeding programme, the final plan should be presented to a wider group of stakeholders for discussion, revision if necessary and final approval. In particular, large numbers of livestock keepers that will be implementing the programme and subject to its costs and benefits must be consulted.

Task 2. Organize the logistics, implement and monitor the cross-breeding plan Once the genetic plan for cross-breeding has been developed and agreed upon by stakeholders, the next step is to organize, launch and operate the plan, including procedures for monitoring its success.

These activities are described in detail in Breeding strategies for sustainable management of animal genetic resources (FAO, 2010). A summary is presented here.

Action 1. Prepare the plan for the start of the cross-breeding programme Before a cross-breeding programme can be launched, various factors have to be accounted for. For example, specialized personnel may need to be appointed to manage the programme. Infrastructure for communication and transport of animals may be needed. A financial analysis of the programme may be warranted, especially if substantial investments are required.

Action 2. Establish the financial and organizational structures If outside investment is needed, these funds will have to be secured most likely from the government or a specialized NGO. The cross-bred animals may require management that is different from that used to raise the original pure-bred animals, so training activities for livestock keepers may be required.

Action 3. Implement the cross-breeding programme The cross-breeding programme will require continual attention and monitoring to detect and resolve unexpected problems. The appointment of a committee of particularly competent livestock keepers to aid in providing advice to their contemporaries and feedback to the National Advisory Committee for Animal Genetic Resources is recommended. Extension services should be established or strengthened and used to disseminate solutions to problems encountered.

Action 4. Organize the delivery of cross-breeding services Cross-breeding programmes may require systems for exchange of germplasm that are more complicated than those for pure-breeding programmes. For pure-bred crosses, F1 animals may be produced on one or more farms and distributed to others. For rotational systems, breeders will need to have access to males of a variety of breeds, either as live animals or through AI. Programmes for synthetic breeds will likely benefit from the establishment of a new breeders association and AI services. Support for research on ways to improve the programme will likely be beneficial.

Action 5. Improve the cross-breeding services and promote uptake Promotion of the cross-breeding programme will help increase the number of livestock keepers involved, which will likely improve its success through various economies of scale and thus improve the sustainability of the targeted animal genetic resources. Programmes for animal identification, performance and pedigree recording will also contribute to the genetic improvement of animals and aid in general management of mating systems, as well as providing documentation for the evaluation of the programme.

Action 6. Evaluate the cross-breeding programme for benefits obtained and sustainability The programme will need to be evaluated periodically to determine if its objectives are being met. In particular, programmes established to contribute to conservation need to be evaluated in terms of their effects on the targeted breed. The results of these analyses should be reported to all stakeholders, including livestock keepers, policy makers and any funding agencies.

CGRFA/WG-AnGR-7/12/Inf.6 References Bennewitz, J., Simianer, H. & Meuwissen, T.H.E. 2008. Investigations on merging breeds in genetic conservation schemes. Journal of Dairy Science, 91: 25122519.

Berg P., Nielsen J. & Srensen M.K. 2006. Computing realized and predicting optimal genetic contributions by EVA. Proceedings of the 8th World Congress on Genetics Applied to Livestock Production, Belo Horizonte, Brazil.

Bett, R.C., Johansson, K., Zonabend. E., Malmfors, B., Ojango, J., Okeyo, M. & J. Philipsson.

2010. Computing realized and predicting optimal genetic contributions by EVA. Proceedings of the 9th World Congress on Genetics Applied to Livestock Production, Leipzig, Germany.

EURECA. 2010. Local cattle breeds in Europe, edited by S.J. Hiemstra, Y. Haas De, A. Mki-Tanila & G. Gandini. Wageningen, the Netherlands, Wageningen Academic Publishers (available at http://www.regionalcattlebreeds.eu/publications/documents/9789086866977cattlebreeds.pdf).

FAO. 2003. Community-based management of animal genetic resources. Proceedings of the workshop held in Mbabane, Swaziland 711 May 2001. Rome (available at www.fao.org/DOCREP/006/Y3970E/Y3970E00.htm).

FAO. 2010. Breeding strategies for sustainable management of animal genetic resources. FAO Animal Production and Health Guidelines No. 3. Rome (available at http://www.fao.org/docrep/012/i1103e/i1103e.pdf).

Gowe, R.S., Robertson, A. & Latter, B.D.H. 1959. Environment and poultry breeding problems. 5.

The design of poultry strains. Poultry Science, 38: 462471.

Kosgey, I.S., Baker, R.L., Udo, H.M.J. & van Arendonk, J.A.M. 2006. Successes and failures of small ruminant breeding programmes in the tropics: a review. Small Ruminant Research, 61: 1328.

Meuwissen, T.H.E. 1997. Maximizing the response of selection with a predefined rate of inbreeding.

Journal of Animal Science, 75: 934940.

Mirkena, T., Duguna, G., Haile, A., Tibbo, M., Okeyo, A.M., Wurzinger, M. & Slkner, J. 2010.

Genetics of adaptation in domestic farm animals. A review. Livestock Science, 132: 112.

Moreno, A., Salgado, C., Piqueras, P., Gutirrez, J.P., Toro, M.A., Ibez-Escriche, N. & Nieto, B. 2011. Restricting inbreeding while maintaining selection response for weight gain in Mus musculus. Journal of Animal Breeding and Genetics, 128: 276283.

Perezgrovas, R., Castro, H., Guarn, E. & Parry, A. 1997. Produccin de velln sucio y crecimiento de lana en el borrego Chiapas. I. Estacionalidad. IX Congreso Nacional de Produccin Ovina. AMTEO. Quertaro, Mexico.

Schmidt, G.H. & Van Vleck, L.D. 1975. Principles of dairy science. New York, USA, W.H.

Freeman and Company.

Shrestha, J.N.B. 2005. Conserving domestic animal diversity among composite populations. Small Ruminant Research, 56: 320.

Sonesson, A., Janss, L.L.G. & Meuwissen, T.H.E. 2003. Selection against genetic defects in conservation schemes while controlling inbreeding. Genetics Selection Evolution, 35: 353368.

Tisdell, C. 2003. Socioeconomic causes of loss of animal genetic diversity: analysis and assessment.

Ecological Economics, 45: 365376.

Toro, M.A. & Prez-Enciso, M. 1990. Optimization of selection response under restricted inbreeding. Genetics Selection Evolution 22: 93107.

Valle Zrate, A. 1999. Livestock biodiversity in the mountains/highlands opportunities and threats.

Paper presented at the International symposium on Livestock in Mountain/Highland Production Systems: Research and Development Challenges into the Next Millennium, 7 December, 1999, Pokhara, Nepal.

CGRFA/WG-AnGR-7/12/Inf.6 Van Arendonk, J.A.M. 2010. The role of reproductive technologies in breeding schemes for livestock populations in developing countries. Livestock Science, 184: 213219.

Woolliams, J. 2007. Genetic contributions and inbreeeding. In K. Oldenbroek, ed. Utilization and conservation of farm animal genetic resources, pp. 147165. Wageningen, the Netherlands, Wageningen Academic Publishers.

Wurzinger, M. Solkner, J. & Iniguez, L. 2011. Important aspects and limitations in considering community-based breeding programs for low-input smallholder livestock systems. Small Ruminant Research, 98: 170175.

CGRFA/WG-AnGR-7/12/Inf.6 VIII. OPPORTUNITIES TO INCREASE THE VALUE AND SUSTAINABILITY OF BREEDS IN IN SITU CONSERVATION PROGRAMMES Section 7 describes how selective breeding can be used to improve an at-risk breed in terms of genetic merit for production, and thereby improve its abilities to compete financially with other breeds. Although selection for increased production is usually recommended (as part of a comprehensive breeding goal) for endangered and vulnerable breeds with sufficiently large N e, genetic improvement may not be enough to make the breed economically competitive. In most instances, the difference between the production levels of local breeds and high-output transboundary breeds is large. As genetic improvement is a rather slow process, many years may be required before a low-output breed achieves a competitive level of production. In other instances, a breed may be uniquely adapted to its environment and selection may upset this balance, with detrimental effects on both the breed and its environment. Therefore, although genetic improvement is usually recommended, conservation programmes should include additional, short-term approaches aimed at increasing the value of the targeted breeds.

Opportunities for sustainable use of breeds targeted for conservation Rationale There are several measures that can, depending on the circumstances, be taken to promote the sustainable use of breeds targeted for conservation by stabilising or increasing the incomes of their owners (Oldenbroek, 2007):

Safeguarding the production environment in which the breed is found or the traditional lifestyle of its keepers.

Improving the management of the animals at farm level. The production level of animals is affected by their genetic ability and the by their management (e.g. quantity and quality of the feed provided, housing and disease control). Although improving management requires investment, it will usually provide greater economic returns.

Developing high-quality products for niche markets. Breeds have genetic differences in production potential and in the quality of their products. In general, selection for high production has a negative effect on the quality of products. Breeds targeted for conservation may have lower production potential than other breeds, but they may be the source of high quality products (e.g. cheese, cured meats and textiles) that can be sold in niche markets where per-unit prices are higher, and this can compensate for the lower amount of product obtained.

Promoting high-quality products by highlighting their connectedness to their places of origin. Such factors can be used to promote the products, for example through labelling schemes. Such activities require collaboration with breeders, producers and marketers to realize the enhanced price for the high-quality products, as well as with consumers, to ensure that sufficient demand exists.

Marketing products may be based on social concerns for improved animal welfare. Intense selection for high production has decreased fitness traits in many high-output international transboundary breeds. These weaknesses are typically magnified if the breeds are kept environments to which they are not adapted. Breeds raised in their traditional production systems are likely to be healthier than animals introduced to these production systems.

Marketing products based on concerns about breed conservation, i.e. because buying them helps to ensure the future existence of an at-risk breed.

Drawing on the ecological functions of species and breeds in nature management to obtain an additional source of income. In many areas of the world, natural grasslands, wetlands or heathlands would become forest or low-value scrub if the vegetation were not shortened regularly. Governments may be willing to pay for the service of maintaining these areas.

CGRFA/WG-AnGR-7/12/Inf.6 Grazing herbivores are already used to conserve such habitats in many countries. Well-adapted breeds of cattle, sheep, goats and horses can be conserved in large numbers to fulfil this task.

Drawing on governmental support or incentive payments from other source to sustain the societal and cultural functions of livestock species and breeds, including their roles in promoting tourism. This opportunity is often discussed, but it seems to be very difficult to realize. However the potential touristic value of some breeds linked to their appearance, farming systems or folklore traditions could be rather easily exploited by the tourism industry with beneficial effects for livestock keepers.

Objective: To document opportunities for conserving breeds.


1. List of breeds by species and characteristics of each breed.


A description of the opportunities to secure the position of breeds or to generate more income.

Task 1. Select the opportunities for utilizing the targeted breed Action 1. Determine the relevant opportunities and threats affecting the species to which the breed belongs in the country as a whole or in the local area As part of characterization efforts (see Sections 1 and 3), the importance of the breed to the local area should be documented, as it may reveal opportunities that can be exploited to promote its sustainable utilization. The analysis of threats should include not only the effects of low productivity, but also other factors such as loss of access to the animals production environment or cultural factors taking the livestock keepers away from farming.

If low productivity is found to be a weakness of the breed, improving livestock management should be among the first opportunities considered, as this will usually yield the fastest results, can be applied in almost all situations, and can complement other approaches. Not all opportunities are relevant for all species. For example, organic farming is usually not a good option for horses, and management of natural areas provides limited opportunities for chickens. Cheese production is a real opportunity for milk-producing cattle, sheep and goat breeds.

Action 2. List the characteristics of the breed and combine them with different opportunities (improving management, specialized or niche production, hobby farming, use in nature conservation, etc.) For example, a table can be created in which the breeds characteristics are listed down the left margin and opportunities listed across the tops of the columns. Cells in the table can then be marked to show which opportunities are relevant to which characteristics.

The matching of characteristics and opportunities can be done as part of the SWOT analysis of breeds described in Section 1, by identifying a strategy that combines Strengths and Opportunities (SO Strategy). Box 56 shows a list of the strengths of the Nguni cattle breed of South Africa and opportunities to exploit these strengths.

Action 3. Describe the realistic opportunities for the breed and make a plan to exploit each of these opportunities Based on the relevant combinations of characteristics and opportunities identified in Action 2, specific measures to take advantage of these opportunities can be proposed. The activities required should be outlined and relevant stakeholders identified. The strengths and weakness of the opportunities should be noted, along with any obstacles that need to be overcome. The chances of success of a programme to exploit the opportunity should be assessed.

CGRFA/WG-AnGR-7/12/Inf.6 Box Strengths of the South African Nguni cattle breed and opportunities to increase its value The Nguni is a local South African cattle breed that has been raised by indigenous communities for nearly 1500 years. Because of its relatively small size and prejudicial preference for their own breeds, the Nguni was regarded as inferior by the settlers that colonized the country in the latter part of the last millennium. At different stages throughout history, the government supported specific programmes that favoured exotic breeds, leading to the near annihilation of the Nguni through grading up and breed substitution. The negative perception of the breed persisted until about years ago when local officials began to appreciate the positive aspects of the breed. Well-designed characterization studies demonstrated that the Nguni was quite competitive with exotic breeds when compared in the same production environment. Having evolved in the area, the Nguni is well adapted to the prevailing climatic conditions and endemic diseases and pests. The government has realized that the Nguni could be a valuable animal genetic resource, especially for resource poor farmers, and the Nguni breed is now making a comeback, based both on its value for meat production in harsh environments and by on other strengths. The table below outlines the breeds strengths and the corresponding opportunities to add value.

Strength Opportunity adaptability low-cost production in marginal areas meat quality branded products, cross-breeding unique coat patterns specialty leather products tick resistance higher-quality hides (no tick-bite damage) sloping rump structure easy calving, lower production costs Source: Ramsay et al. (2000).

Preparing a Biocultural Community Protocol for documenting a breeds characteristics and the indigenous knowledge of its keepers Rationale In many instances, particularly in developing countries, breeds at risk have been developed and are largely kept by members of a specific community. These communities will often have a very strong cultural tie to their breed and strong interactions are likely to exist among the community, the breed and the surrounding environment in which the community and its animals exist. The community will often hold a wealth of indigenous knowledge on how to sustainably co-manage the animals and the local environment. In such cases, the survival of the breed will depend not only on its characteristics, but also on the continued existence of the community and the ability of community members to maintain satisfactory livelihoods in the face of external forces. The endangerment of the breed may be an indirect result of threats to the existence of the community itself. For example, loss of access to grazing lands or sources of water may hinder the ability of pastoral communities to continue their traditional lifestyles, including raising their particularly adapted livestock. Documenting the functions of a breed and the cultural practices and indigenous knowledge of the community that keeps it can inform policy-makers about their importance to society in general, including their role in the conservation of biological diversity.

One approach to gathering, organizing and disseminating this information is to develop a Biocultural Community Protocol a concept originally developed by the NGO Natural Justice of Natural Justice, 2009;

LPP and LIFE Network, 2010). BCP are formal documents prepared on the basis of consultations between members of livestock-keeping communities (or other types of communities) and lawyers and experts in indigenous knowledge. They record (among other things) information about communities breeds, their roles in the livelihood of the communities, the indigenous knowledge of the community and the role of the community in managing diversity. Community protocols are recognized by the CBD and referred to in the Nagoya Protocol on Access and Benefit Sharing CGRFA/WG-AnGR-7/12/Inf.6 (Nagoya Protocol) 16. Box 57 describes the development of a Biocultural Community Protocol for the Samburu community in Kenya.

Presenting a BCP to policy makers will raise awareness of the activities of the community with respect to preservation of agricultural biodiversity and may help encourage the development of policies that are favourable to the continued existence of the community keeping the breed at risk and thus favourable to the survival of the breed. In addition, the activities through which the BCP is developed are usually educational to the livestock keepers themselves, increasing their awareness of the value of their resources and of their rights and responsibilities. The process of developing the BCP generally serves to empower the livestock keepers and their community.

Objective: To develop a BCP for keepers of a breed at risk.


1. An indigenous community with a breed at risk;

2. Knowledge of the characteristics of the breed, traditional practices in its management, its cultural significance and its interaction with the environment;

and 3. A facilitating organization and a team of legal and other experts.


A BCP relating to the breed at risk and its community of livestock keepers;

and Community members informed about their rights and the value of their breed.

Task 1. Gather stakeholders and discuss the concepts underlying the content of the proposed BCP Action 1. Establish a working relationship between the facilitating organization and the local community Ideally, BCP are initiated and developed by the communities themselves. This is not realistic, however, in most situations, as many communities will not even be aware of the existence of BCP and not have access to the precise legal expertise required. Thus, the process is usually facilitated by an NGO or other external organization, ideally with an ongoing relationship with the community. If no such organization exists, then time must be spent in achieving familiarity between the facilitators and the community. Background research on the community should be undertaken before the process is initiated. Even if an ongoing relationship with the community has been already established, the process should be done very deliberately, at a pace set by the community rather than the facilitators.

Pages:     | 1 |   ...   | 2 | 3 || 5 |

2011 www.dissers.ru -

, .
, , , , 1-2 .