diff --git "a/data/part_4/05e0c3d1197b76d20bc3e46e4ef81463.json" "b/data/part_4/05e0c3d1197b76d20bc3e46e4ef81463.json"
new file mode 100644--- /dev/null
+++ "b/data/part_4/05e0c3d1197b76d20bc3e46e4ef81463.json"
@@ -0,0 +1 @@
+{"metadata":{"id":"05e0c3d1197b76d20bc3e46e4ef81463","source":"gardian_index","url":"https://cgspace.cgiar.org/rest/bitstreams/29bef3c1-429a-48f8-bc5d-5e36634e80f6/retrieve"},"pageCount":115,"title":"Design of community based breeding programs for two indigenous goat breeds of Ethiopia","keywords":[],"chapters":[{"head":"List of tables","index":1,"paragraphs":[]},{"head":"Abbreviations and Acronyms","index":2,"paragraphs":[{"index":1,"size":45,"text":"Boison a PhD student at BOKU in SNPs data analysis is highly indebted. I am also thankful to current and past BOKU livestock division PhD students and staff; Lina, Romana, Anamarija, Gabor, Marko, Kessang and Sophie for their help and great time we had together."},{"index":2,"size":51,"text":"I would like to thank the development agent at Metema Biruk, Mulugeta and the chairman of the kebele administration Gashaw for their valuable help during data collection. I would like to appreciate and acknowledge the farmers who participated in the study for providing their time and their animals without any payment."},{"index":3,"size":64,"text":"Finally, my special thanks and love goes to my wife Birhan Atena and our children Kirubel and Wdassie for their love, prayer and their patience in my long absent. Birsh I know how difficult is to take-care two little kids in the absence of father but you did it well. I thank you again for your great job you did in this difficult time."},{"index":4,"size":6,"text":"xii Solomon Abegaz GUANGUL PhD Thesis"}]},{"head":"Abstract","index":3,"paragraphs":[{"index":1,"size":273,"text":"The objectives of this study were to describe the production systems, identify the breeding objectives traits, describe the morphological and molecular characteristics of Western Lowland and Abergelle goat breeds of Ethiopia to design community based breeding programs. 120 goat keepers were interviewed. Phenotypic characters of 534 Abergelle and 476 Western Lowland goats were measured. Genetic diversity of the two Ethiopian and three Nigerian goat types were assessed based on 47K Single Nucleotide Polymorphism markers. The breeding objective traits were investigated through own and group animals ranking experiments. Community based one tier breeding schemes with four different alternatives for the top three most important traits were simulated. The survey results indicated that goats were kept for multifunctional roles in both areas. Phenotypic characterization showed high variability within and between the studied breeds in qualitative and quantitative traits. Western Lowland goats are on an average not only bigger than Abergelle goats but also show considerably higher variation in body size. The genetic diversity analysis revealed that the studied goat populations were well differentiated based on their geographical location. Production and reproduction traits such as body size, twinning and milk yield were identified as important breeding objective traits in own flock ranking experiment while in group ranking experiment the observable characters like body size, body conformation and coat color were identified as breeding goal traits. Simulation results gave an acceptable range of genetic gains with little difference across the alternatives. Thus, the community based breeding programs with a few traits in the recording are considered feasible for genetic improvement of goats in the study areas and similar agro-ecological zones. Schlagwörter: Äthiopien, Ziegen, Charakterisierung, Zuchtziele, SNP, Zuchtprogramm"}]},{"head":"Introduction","index":4,"paragraphs":[{"index":1,"size":177,"text":"Ethiopia is a country in East Africa where agriculture is the main stay of the economy. More than 85% of the population depends on agriculture for their livelihoods. It provides 80 % of total employment and 85 % of export earnings. The livestock sub-sector has a share of 12-16% of the total Gross Domestic Product (GDP), 30-35% of agricultural GDP and more than 85% of farm cash income (IBC, 2004;Benin et al., 2006). Goat production is one of the integral parts of livestock farming activities of the country. Based on phenotypic and molecular characterization, there are four families and 12 different types (FARM-Africa, 1996;Tesfaye, 2004) and 29.9 million goats in Ethiopia (CSA, 2010a) which are distributed throughout the country. The majority of the goat population is found in large flocks in the arid and semi-arid Lowlands. Goats in the highlands are widely distributed in the mixed crop-livestock production systems with very small flock size (Tsegahun et al., 2000). Almost all goat population is managed by resource poor smallholder farmers and pastoralists under traditional and extensive production systems."},{"index":2,"size":57,"text":"In traditional production systems, small ruminant are not bred for a specific purpose rather they are kept for multipurpose functions. They provide multiple roles for their owners such as source of income, food (meat and milk), manure, insurance against crop failure and cultural value (Jaitner, et al., 2001;Herpa and Adane, 2008;Legesse et al., 2008;Assen and Aklilu, 2012)."},{"index":3,"size":100,"text":"The growing demands of meat products at the domestic as well as international markets also increase the importance of goat in the national economy of the country. According to CSA (2012) out of 5,187,044 slaughtered animals in the year 2012/2013, 1,771,527 are goats. More than 90% of export trade value of live animal/meat and skin and hide also comes from small ruminants. Special features of goats among the livestock species are small body size, less space requirement, low feed requirement, use poor quality forage and fast turnover make them widely acceptable species in tropical harsh climatic condition (Peacock, 2005;Mekasha, 2007)."},{"index":4,"size":67,"text":"Despite of the large population of goats and the roles of goats at household and national level, the productivity and the contribution of goat to the country economy is far below the potential. At optimum level the county has a potential of annual production of 1.1 million goats for domestic market and 2 million goats for international market but the current annual off take is only 35%"},{"index":5,"size":36,"text":"with the average 10 kg of carcass weight (Herpa and Adane, 2008). Goat production in Ethiopia is constrained by many biological, environmental and socio-economical factors. Among them, lack of systematic breeding programs is an important constraint."}]},{"head":"PhD Thesis","index":5,"paragraphs":[{"index":1,"size":198,"text":"Therefore, there is a need to design and implement the appropriate breeding strategies to improve the livelihoods of the small holder farmers and to satisfy the growing demand of meat for domestic consumption and international market. However, there is no systematic goat breeding program is in place and goat is the most neglect livestock species in research and development endeavors (Tsegahun, et al., 2000). There have been a few attempts of genetic improvement program of goats through upgrading the exotic genetic blood levels. The noticeable example is the FARM-Africa dairy goat development project in south and eastern part of the country. The aim of the project was to improve the milk yield of the local breeds through crossing with exotic Anglo-Nubian goats (Gebremeskel, 2000). However it was reported that crossbred goats did not perform better than indigenous goats if both groups were kept in similar management levels (Ayalew et al., 2003). In general, many small ruminants cross breeding programs in tropical country were not successful because of the incompatibility of the genotype with the farmers breeding objectives, management methods and the prevailing environment of the tropical low input production systems (Ayalew et al., 2003;Wollny, 2003;Kosgey et al., 2006)."},{"index":2,"size":102,"text":"Thus, selective pure breeding of the adapted indigenous breeds is the best possible option of genetic improvement in the tropical countries. Indigenous breeds in harsh tropical environmental conditions have special adaptive features such as tolerance of a wide range of disease, water scarcity tolerance and ability to better utilize the limited and poor quality feed. This makes them survive and be productive in the prevailing environment (Baker and Gray, 2004;Kosgey and Okeyo, 2007). To efficiently utilize these special features of indigenous breeds, there is a need of planning and implementing viable breeding programs that fit to the existing low input production systems."},{"index":3,"size":128,"text":"The recent approach of establishing community based breeding programs is advocated for low input traditional smallholder farming systems (Sölkner et al., 1998;Kahi et al., 2005;Haile et al., 2009, Wurzinger et al., 2011). This is because community based breeding programs take into account the indigenous knowledge of the communities on breeding practices and breeding objectives (Gizaw et al., 2013).The community-based breeding strategies also consider the production system holistically and involve the local community at every stage, from planning to operation of the breeding program ( Baker and Gray, 2004). Breeding programs involve the description and decisions about a series of interacting components. Among them the most important components to be considered in breeding program design are: description of production environment and production system, characterization of the available local genotype,"}]},{"head":"PhD Thesis","index":6,"paragraphs":[{"index":1,"size":40,"text":"definition of breeding objectives, identification of traits to be selected, decision about breeding methods and breeding population and understanding of structure and organization of people involved (Iñiguez, 1998;Sölkner et al., 1998;FAO, 2010). Thus, this study was aimed to PhD Thesis"}]},{"head":"Literature review","index":7,"paragraphs":[]},{"head":"Goat genetic resources in Ethiopia","index":8,"paragraphs":[{"index":1,"size":163,"text":"Goats (Capra hircus) are believed to have been the first ruminant animal domesticated. It is also believed that the first goats reached Ethiopia from the North between 2000 and 3000 B.C. (Rege and Lebbie, 2000). Due to a great variation in climate and topography and proximity to the historical root of livestock domestication of Africa, Ethiopia endows large and very diverse farm animal genetic resources (Ayalew, 2004). The goat population of Ethiopia is estimated at 29.9 million (CSA, 2010a). It is believed that these goats have evolved through a process of natural selection that resulted in goats selected for adaptation and survival rather than production per se. (Peacock, 1996;Abegaz et al, 2008).  (Devendra, 1978). According to the FARM-Africa (1996) goat breed survey report, indigenous Ethiopian goats have been phenotypically classified into four families and 12 types (Table 1). However, Tesfaye (2004) reported only eight distinctively different breeds based on microsatellite markers analysis: Arsi-Bale, Gumuz, Keffa, Long-Ear Somali, Woyto-Guji, Abergelle, Afar and Highland Goats."}]},{"head":"Solomon Abegaz GUANGUL","index":9,"paragraphs":[{"index":1,"size":2,"text":"PhD Thesis "}]},{"head":"Ethiopian small ruminant production system","index":10,"paragraphs":[{"index":1,"size":45,"text":"Livestock production systems in tropical countries are complex by nature and show great variation within and between regions. They depend on integration with crop production, climatic condition, management practice, local resource availability, production objectives of the owners, availability of technologies and government policy (Othere, 1998)."},{"index":2,"size":121,"text":"Goat production and livestock systems at large in Ethiopia have evolved largely as a result of natural production environments and socio-economic circumstances of farmers/pastoralists (Gizaw, et. al., 2010). Ethiopian small ruminant production systems are broadly classified into \"modern\" and \"traditional\" (Tibbo, 2006;Legesse, 2008). The \"modern\" system is practiced only in few places such as government ranches and in small scale urban production systems while most of small ruminant production depends on the traditional extensive system of production (Tibbo, 2006;Gizaw, et al., 2010a). Common features of traditional production systems are limited number of animals per unit area, low productivity per animal, relatively limited use of improved technology and use of on farm by products rather than purchased inputs (Gizaw et al., 2010a)."}]},{"head":"PhD Thesis","index":11,"paragraphs":[{"index":1,"size":65,"text":"According to the degree of integration with crop production and contribution of livelihood, level of input and intensity of production, agro ecology, length of growing period and relation to land and type of commodity to be produced, mobility and duration of movement, the traditional production system is sub divided into three systems (Abegaz et al., 2008). These are mixed crop-livestock, pastoral and agro pastoral system."},{"index":2,"size":123,"text":"Mixed crop-livestock system is commonly practiced in the most crop dominant area of high land and mid-altitude of the country, with altitude ranges of 1500 to 3000 m. The area receives good amount of rainfalls and has moderate temperature. Goats are kept by smallholders and graze together with sheep and/or other livestock species like cattle. The integration and the importance of small ruminants (goat) in the system varies from place to place. The integration is lower in south part of the country where the perennial crop production is more important and small ruminants are less important. In the dry highland area of the Northern part of the county, goat plays a great role where crop production is unreliable (IBC, 2004;Gizaw et al., 2010a)."},{"index":3,"size":103,"text":"The pastoral system is practiced by pastoral people in very dry parts of the country at altitudes below 1500 m. The areas are not suitable for crop production and receive less than 500mm of precipitation. The livelihoods of the pastoral people depend entirely on livestock and more than 50% of the household income and 20% of the food comes from the livestock or livestock related activities. Goats are kept by nearly all pastoralists with higher flock size, often in mixed flocks with sheep. High mobility of animals in search of feed and water is common in the system (IBC, 2004;Abegaz et al., 2008)."},{"index":4,"size":30,"text":"Agro pastoral system is practiced in the semi-arid part of the country. Comparing to the pastoral system the area receives relatively higher rain and people and animals are less mobile."},{"index":5,"size":43,"text":"The system is characterized by high degree of dependency on milk and meat production and 10-50% of the income is derived from livestock production. In this system there is some crop agriculture practice along with the livestock production (IBC, 2004;Abegaz et al., 2008)."},{"index":6,"size":7,"text":"Differently from the above classification, Legesse (2008)  "}]},{"head":"Performance of indigenous goats","index":12,"paragraphs":[{"index":1,"size":36,"text":"Evaluations of the performance of economically important traits of the livestock are very useful inputs for planning a breeding program. The most important traits of livestock are broadly classified into two categories: production and reproductive traits."}]},{"head":"1 Growth performance","index":13,"paragraphs":[{"index":1,"size":116,"text":"Growth performance traits are the most important traits for meat production. To increase economic return from goat production requires improvements in market weight of kids and mature goats. Growth performance may be separated in pre-weaning (birth weight, weaning weight and pre weaning growth rate) and post weaning (six months weight, yearling weight and mature weight). The growth performance of goats is affected by many genetic and non genetic factors. The early stage of growth performance of kids is largely influenced by genotype and the milk yield of the does. Parity, type of birth, sex, season and year of birth also influence the growth performance of goats (Dadi et al., 2012;Bedhane et al., 2013;Derbie and Taye, 2013)."},{"index":2,"size":22,"text":"The mean birth weight, weaning weight, six months and yearling weight of some indigenous Ethiopian goat breeds are presented in Table 2. "}]},{"head":"Reproductive performance","index":14,"paragraphs":[{"index":1,"size":167,"text":"Reproductive performance is an important criterion when evaluating the structure of the strength and weakness of the breeds in particular production environments (Browing et al., 2006)  4). The litter size is largely influenced by ovulation rate. The ovulation rate of the does is highly influenced by the breed and improvement could be achieved by selection (Ibrahim, 1998). Age at fist kidding is an indication of the overall flock productivity. The lifetime production can be increased by decreasing first kidding age. A wide range of 375 to 854 day of age at first kidding (Table 3) were reported in different management and breeds of Ethiopian goats which is influenced by genotype, management, season and type of birth (Derbie 2008; Kebede et al. 2012a). Kidding interval is the interval between two kidding. A doe with long kidding interval has lower overall production index (Ibrahim, 1998). Mean litter size, age at first kidding and kidding interval of some of Ethiopian goat breeds from different references are summarized in Table 3. "}]},{"head":"3 Milk production","index":15,"paragraphs":[{"index":1,"size":146,"text":"Goat provides milk mainly for the resource poor farmers. In the central rift valley, in Eastern, Southeastern and Northeastern part of the country, goat milk is consumed by farming community (Workneh et al., 2004). Very limited information is available about milk performance characters of Ethiopian goats. A comparison made between Somali goat breeds and their crosses with Anglo-Nubian goats showed that the mean daily milk was higher (330 vs 837ml) for crossbred goats (FARM-Africa, 1995). Bedhane et al. (2012) reported 209 gram of daily milk yield, 86 days of lactation length and 18 kg of lactation milk yield for Arsi-Bale goats under station management. In a comparison study between pure breed Adal and Quarterbred with Saanen, quarterbred Saanen gave more lactation milk yield (31 kg) than the Pure Adal goat (24 kg) and 84 days lactation length were reported for both genotypes (Banerjee et al., 2000)."}]},{"head":"Population parameters of tropical goats","index":16,"paragraphs":[{"index":1,"size":42,"text":"Variations in the performance of traits within and between breeds are important raw materials in animal breeding. Heritability and additive genetic correlation of the traits are the most important population parameters of within breed variation in animal breeding (Rege et al. 2006)."}]},{"head":"Heritability","index":17,"paragraphs":[{"index":1,"size":199,"text":"The heritability (h 2 ) of a trait, a central concept in quantitative genetics, is the proportion of variation among individuals in a population that is due to variation in the additive genetic effects (i.e., breeding values of individuals that determine reproductive efficiency of goat production. Falconer and Mackay, 1996). A reliable estimate of heritability will help to decide which breeding program should be used. If the trait heritability is high, mass selection with little PhD Thesis pedigree records may be enough for rapid selection response. On the other hand, if the heritability of a trait is low, response from selection on individual records will be slow. It needs accurate pedigree records, family selection or even progeny testing (Legates and Warwick, 1990).The heritability estimates of traits performance of Ethiopian goat is very scant and only available for Arsi-Bale goat breeds. Bedhane et al. (2013), Bedhane et al. (2012) and Kebeda et al. (2012a) estimate heritability of growth traits, milk traits and reproductive traits of Arsi-Bale goat in Adami-Tulu research center, respectively. They also estimate the genetic and phenotypic correlation of these traits. The heritability estimates of different traits for some of African goat breeds are summarized in Table 4."}]},{"head":"Genetic and phenotypic correlations","index":18,"paragraphs":[{"index":1,"size":145,"text":"Correlation is the measure of association between two characteristics or traits. Both genetic and phenotypic correlations are of particular interest to animal breeders. They are an integral part of breeding program design and analysis. The phenotypic correlation is an estimate of the association between two visible characteristics in the current flock. The genetic correlation is the correlation between breeding values. It is an estimate of the way in which selection of parents for one trait will cause a change in a second trait in the progeny. Observations of traits on related animals are used to estimate genetic correlations. Genetic correlations are an indication of the proportion of genes that affect both traits in the direction of the sign (positive or negative (Simm, 1998). Genetic and phenotypic correlation estimates of some of important traits of Ethiopian and other tropical goat breeds are summarized in Table 5.  "}]},{"head":"Breed characterization","index":19,"paragraphs":[{"index":1,"size":49,"text":"A good understanding of breed characteristics is the base for decision making in livestock development and breeding programs (FAO, 2007). Breed characterization includes all activities related with the description of the origin, development, structure, population, quantitative and qualitative characteristics of the breeds in defined management and climatic conditions (Ayalew,"}]},{"head":"Solomon Abegaz GUANGUL","index":20,"paragraphs":[{"index":1,"size":20,"text":"PhD Thesis 2004; Rege, et al., 2006;Gizaw et al., 2011). Breeds can be characterized by morphological (phenotypic) and molecular tools."}]},{"head":"Morphological or phenotypic characterization","index":21,"paragraphs":[{"index":1,"size":144,"text":"According to FAO (2012) phenotypic characterization is defined as the process of identifying distinct breed populations and describing their external and production characteristics in a given environment and under given management, taking into consideration the social and economic factors that affect them. Phenotypic characterization is description of breeds in terms of external characteristics (such as coat color, ear type and shape, horn shape and type), linear body measurements (such as height at wither, heart girth, body length, ear length), production traits (body weight, milk yield) and reproductive traits (such as age at first kidding, litter size) (FAO, 1986;Tesfaye, 2004;FAO, 2012). Phenotypic characterization is a comparatively easy and cheap tool of breed characterization but phenotypic characters are highly influenced by environmental effects and by sometimes strong genetic and environmental correlations and interaction. Therefore, it should be supported by molecular characterization (FAO, 2011;Gizaw et al., 2011)."}]},{"head":"Molecular characterization","index":22,"paragraphs":[{"index":1,"size":123,"text":"Molecular characterization involves describing and classifying of livestock breeds and species at molecular level by measuring frequencies of genotypes and alleles, degrees of polymorphism, allelic diversity (observed and expected hetrozygosity) and genetic distances ( Toro et al., 2009;Gizaw et al., 2011). Tools for molecular analysis are biochemical (protein) polymorphisms and molecular DNA. Protein (Allozymes) polymorphisms were the first markers used for genetic studies in livestock. However, the number of polymorphic loci that can be assayed, and the level of polymorphisms observed at the loci are often low, which greatly limits their application in genetic diversity studies (Toro et al., 2009). in genetic diversity analysis (Baumung et al., 2004: FAO, 2011). However, microsatellites and SNPs are the currently most used and recommended (FAO, 2011)."}]},{"head":"PhD Thesis","index":23,"paragraphs":[]},{"head":"Microsatellites","index":24,"paragraphs":[{"index":1,"size":51,"text":"Microsatellites are simple tandem nucleotide repeats, interspersed throughout the genome and usually found in non-coding part of the genome (FAO, 2011). Because of their high polymorphism, high abundance, co-dominant inheritance, simplicity to analyze and ease of scorine, microsatellites have been the markers of choice until very recent and have been used"},{"index":2,"size":69,"text":"for genetic diversity studies of many livestock species (Arif and Khan, 2009;Baumang et al., 2004). They have been successfully utilized for genetic analysis of different breeds of goats worldwide (Agha et al., 2008;Fatima et al., 2008;Li et al., 2008;Missehou et al., 2011;Hassen et al., 2012). Microsatellites have some limitations in genetic diversity study such as null alleles, interpretation difficulty of allele calling and size homoplasy (Pariset et al., 2009) "}]},{"head":"Single nucleotide polymorphisms (SNPs)","index":25,"paragraphs":[{"index":1,"size":110,"text":"A SNP is a DNA sequence variation that occurs through the substitution of one nucleotide by another at a single location within the genome of a species or breed (Beuzen et al., 2000;Vignal et al., 2002;FAO, 2011). SNPs occur about once every 1 kb, within the coding and non-coding regions. For a variation to be considered a SNP, it must occur in at least 1% of the population (Vignal et al., 2002;Mburu and Hanotte, 2005). SNP markers have promising advantages over microsatellite markers such as being prevalent and providing potential markers near or in any locus of interest. Some SNPs are located in coding regions and directly affect protein function."},{"index":2,"size":97,"text":"SNPs have lower mutation rates than microsatellites, making them more suited as long term selection markers and SNPs are more suitable for different genotyping techniques and have strong potential for automation (Beuzen et al., 2000;Vignal et al., 2002;Herraz et al., 2005;Nigrini et al., 2008). However, as SNPs have bi-allelic nature, the information content per SNP marker is lower than that of microsatellite markers. Around 5-6 SNPs markers are equivally informartive as one microsatellite marker (Beuzen et al., 2000;Toro et al., 2009). Therefore, large numbers of SNPs are required for genetic diversity analysis to get the accurate results."},{"index":3,"size":62,"text":"Due to the growing availability of mapped SNPs of different species, SNP markers are now beinig used for molecular breed characterization of different species in different areas such as cattle (Mckay et al., 2008;Nigrini et al., 2008;Edea et al., 2013), sheep (Kijas et al., 2009), goat (Pariset et al., 2009;Kijas et al., 2012;Hykai et al., 2013) and horses (Petersen et al., 2013)."},{"index":4,"size":12,"text":"Molecular diversity studies based on SNPs for African goat are not available."}]},{"head":"Genetic improvement strategies","index":26,"paragraphs":[{"index":1,"size":90,"text":"The main objective of animal breeding is to genetically improve population of livestock which is achieved through selecting the best individuals of the current generation and using them as parents of the next generation. Genetic improvement aimed to exploit the present within and between breed variations (ILCA, 1994). Any genetic improvement programs falls under the three major genetic improvement pathways: 1.within breed selection; 2. Selection between breeds (breed substitution) and 3 crossbreeding. To be successful in genetic improvement of livestock, appropriate breeding programs need to be planned, implemented and maintained."},{"index":2,"size":160,"text":"Breeding program is defined as the organized structure that is set up in order to realize the desired genetic improvement of the population. Small ruminant breeding programs in tropical country are less organized and largely fragmented. Many of the programs were based on the upgrading of the indigenous animal to exotic breeds. The crossbred animals were often good in terms of growth and some other production traits under on station evaluation. Yet, the performance of the crossbred animals at smallholder level is frequently low (Hassen et al., 2002 ;Ayalew et al., 2003;Kosgey et al., 2006). In general most of the crossbreeding activities were not successful and sustainable due to incompatibility of the breeding objectives and the management approaches of the existing production system of the area (Ayalew et al., 2003;Kosgey et al., 2006;Wollny, 2003). Furthermore, Haile et al. (2009)   et al., 2011). Pure breeding and other breeding programs involve the description of and decisions about a series of interacting components."},{"index":3,"size":55,"text":"Among them the most important components to be considered in breeding program design are: description of production environment and production system, characterization of the available genotype, definition of breeding objectives, identification of traits to be selected, decision about breeding methods and breeding population and understanding people, structure and organization (Iñiguez, 1998;Sölkner et al., 1998;FAO, 2010)."}]},{"head":"PhD Thesis","index":27,"paragraphs":[{"index":1,"size":44,"text":"According to Kosgey et al. (2006) review work, the small flock size, single sire mating, lack of performance and pedigree recording, low level of literacy, and lack of organizational structure hinders within breed selection in tropics. To solve these problems, nucleus (open and closed)"},{"index":2,"size":24,"text":"breeding schemes are the most used and recommended tools for small ruminant genetic improvement programs in tropical countries (Wolly, 2003;Tibbo et al., 2006;Muller, 2006)."},{"index":3,"size":33,"text":"Community based breeding program is a recently advocated option for tropical traditional low input livestock production systems (Sölkner et al., 1998;Kahi et al., 2005;Gizaw et al., 2009;Haile et al., 2011;Wurzinger et al., 2011)."}]},{"head":"Nucleus breeding schemes","index":28,"paragraphs":[{"index":1,"size":110,"text":"The principle of nucleus breeding program is bring the best breeding males and females from the participants (population) to a central place to create elite breeding animals and to make strong selection there. The selected animals (mostly male animals) will be distributed to participating farmers to disseminate the genetic superiority obtained at the nucleus to the whole population. Open nucleus breeding scheme allow the flow of animals in both directions from the nucleus to the population and vice versa while the closed scheme allows only the flow of animal from the nucleus to the population (Tibbo, et al., 2006;Philipsson, et al. 2011)  Wollny, 2003;Kosgey et al., 2006;Kosgey and Okeyo, 2007)."}]},{"head":"2. Community (village) based breeding schemes","index":29,"paragraphs":[{"index":1,"size":52,"text":"Community based breeding programs are a new approach of genetic improvement program proposed for the low input traditional smallholder farming system (Sölkner et al., 1998;Kahi et al., 2005;Haile et al., 2011;Wurzinger et al., 2011). Sölkner et al. (1998) defined community based breeding programs operating in the following conditions: ''Village breeding programs are"}]},{"head":"Solomon Abegaz GUANGUL","index":30,"paragraphs":[{"index":1,"size":58,"text":"PhD Thesis carried out by communities of smallholder farmers (villagers), often at subsistence level. The availability of feed for the animals is far from optimal with large seasonal variations and variations between years (e.g., droughts and floods in the tropics, summer and winter in extreme mountain regions of Asia). The pressure from diseases may be high (tropical regions)."},{"index":2,"size":239,"text":"The level of organization is low, hierarchical structures with good flow of information between levels of the hierarchy cannot always be assumed to work. Data recording in the sense used by animal breeders in the developed countries will often be missing''. According to (Haile et al., 2009) community is defined as the group of people having social, cultural and economic relation based on common interest, goal, problems or practices shared interest and living in a well defined area. Different from the conventional top down approach, community based breeding programs takes in account the indigenous knowledge of the communities on breeding practices and breeding objectives (Gizaw et al., 2013). The community-based breeding strategies also consider the production system holistically and involve the local community at every stage, from planning to operation of the breeding program (Baker and Gray, 2004). The breeding structure of such a program is commonly single-tiered with no distinction between the breeding and production units, i.e., the farmers and pastoralists are both breeders and producers (Gizaw et al., 2013). The basic steps in community based breeding program are selection of the target community and breeds, description of the production system, definition of breeding goals in participatory manner, assessment of alternative schemes and implementation of feasible schemes (Haile, et al., 2011) The following summarized reasons have been identified as an advantage of the community-based breeding programs for sustainable genetic improvement in tropical regions (Wollny, 2003;Kahi et al., 2005): "}]},{"head":"PhD Thesis","index":31,"paragraphs":[{"index":1,"size":21,"text":" The community has the capacity to run the initiative through relevant training and visits to other initiatives in other areas."}]},{"head":"Breeding goal definition","index":32,"paragraphs":[{"index":1,"size":317,"text":"Breeding goal definition is the first step to be made in designing of breeding program. A clear understanding of production objective and breeding goal of the farmers (Beneficiaries) is an important component of planning of breeding programs. The breeding goal identifies the animal traits that farmers would like to be improved. Breeding objectives must to be set at national (macro), regional or local level by stakeholders (and not by outsiders) to truly reflect the real needs of the area; farmers must support the direction of change (Ahuya et al., 2005;Kosgey et al., 2006;FAO, 2010). The ultimate goals of a breed at the macro-level should be expressed by the agricultural development policy, market, production system of the country, region or locality (FAO, 2010). At the micro level the definition of breeding objectives means that for the given production environment the relative importance of improvement of different traits of the breed must be identified (Phillpson, 2011). Breeding objectives are affected by many factors and have to consider the needs and priorities of the animal owners or producers, the consumers of animal products, the food industry, and increasingly also the general public. In smallholder and pastoral communities, breeding goals are multi-functional than and include many aspects other than high productivity (Taye, 2006). Thus the breeding goal definition in subsistence system needs to take account the diversity of traits (Muller, 2006). Therefore, the breeding objective and the selection criteria (traits), on which the livestock keepers wish to improve and base their selection should be identified through the full participation of pastoralist and smallholder farmers. Lack of participation of farmers in defining the breeding objective was the main reason for failure of many livestock improvement programs in tropics (Kahi et al., 2005;Wurzinger et al., 2011). identify the breeding objectives and the breeding objective traits in pastoral and smallholder subsistence system. Their advantages and short-comings of these tools are summarized in Table 6."}]},{"head":"Solomon Abegaz GUANGUL","index":33,"paragraphs":[{"index":1,"size":2,"text":"PhD Thesis "}]},{"head":"Methods of optimizing alternative breeding programs","index":34,"paragraphs":[{"index":1,"size":75,"text":"Breeding planning includes all steps for developing and optimizing breeding programs (Herold et al., 2012). Optimizing the different scenarios of the breeding program is useful procedure to look the quality of breeding schemes through predicting selection response for the breeding goal traits and the economic return of the given scenario. SelAction (Rutten et al., 2002) predict response to selection within the selection group and rate of inbreeding of livestock improvement programs using deterministic simulation methods."},{"index":2,"size":23,"text":"However it does not give the option to evaluate the breeding programs in terms of discounted returns and profits (Herold et al., 2012)."},{"index":3,"size":2,"text":"PhD Thesis"}]},{"head":"Materials and Methods","index":35,"paragraphs":[]},{"head":"The study areas","index":36,"paragraphs":[{"index":1,"size":28,"text":"The study was conducted in Metema and Abergelle districts of the Amhara National Regional State of Ethiopia (Figure 1). Metema and Abergelle districts were purposively selected for this "}]},{"head":"Solomon Abegaz GUANGUL","index":37,"paragraphs":[{"index":1,"size":2,"text":"PhD Thesis "}]},{"head":"Molecular diversity study using SNPs marker","index":38,"paragraphs":[{"index":1,"size":35,"text":"Understanding of the within and between breed diversity are very helpful for future breed improvement and conservation planning. Molecular genetic diversity and homozygous segments of both breeds were studied using 47K genome wide SNPs markers."}]},{"head":"3. 1. Sample collection and genotyping","index":39,"paragraphs":[{"index":1,"size":129,"text":"A tissue sample of 54 and 41 animals were collected from Abergelle and Western Lowland goat (Gumuz) breeds, respectively. To avoid sampling of closely related animals, the samples were collected from different households with a maximum number of three unrelated animals per households. The tissue samples of individual animals were taken from the ear by tissue puncher tubes using Alflex tissue applicator (http://www.allflexusa.com) (Appendix 11). Genomic DNA was extracted from the tissue samples following the standard manufacturers procedure of Qiagen Puregene tissue extraction method (http://www.qiagen.com/gentra-puregene-tissuekit).The genomic DNA was genotyped using Illumina 47K SNPs bead chip technology (Illumina, 2013). Additional data sets of (47K SNPs and 25 animals from each breed) three Nigerian goat population (West African Dwarf, Red Sokoto and Sahel goat) were provided by USDA-ARS for comparison purpose."}]},{"head":"PhD Thesis","index":40,"paragraphs":[]},{"head":"2. Quality control and data management","index":41,"paragraphs":[{"index":1,"size":33,"text":"Animals with greater than 10% missing genotype, SNPs with greater than 10% missing genotype and SNPs assigned for X chromosome were excluded from the data set using PLINK software (Purcell et al., 2007) "}]},{"head":"Participatory identification of breeding objectives traits","index":42,"paragraphs":[{"index":1,"size":49,"text":"Identification of the breeding objectives traits in participatory manner are a recommended approach for the sustainable breed improvement programs in tropics (Sölkner et al., 1998;Gizaw et al., 2010b;Wurzinger et al., 2011). In the present study, participatory own flock ranking and group ranking methods adapted from (Mirkena, 2011) were applied."}]},{"head":"Own flock ranking methods","index":43,"paragraphs":[{"index":1,"size":88,"text":"In own flock ranking experiment, sixty and thirty households, respectively, from Metema and Abergelle areas were visited. The household members were asked to choose their first best, second best, third best and the most inferior does among the breeding does in their flocks. The reasons of ranking and life history of the ranked animals (age, number of kidding, number of kids born per kidding, number of kids weaned) were inquired and recorded. The live body weight and some linear body measurement of the ranked animals were also taken."}]},{"head":"Group animal ranking","index":44,"paragraphs":[{"index":1,"size":170,"text":"Twelve breeding does and twelve breeding bucks from western Lowland goats and fifteen does and fifteen bucks from Abergelle goats were randomly selected. Similar to the own flock ranking experiment, the life history of does and the life history of bucks (age, birth type, libido and PhD Thesis temperament) were inquired from the owners. The live weight of the selected animals was also measured. The selected animals were brought to the central place and randomly assigned into groups. The selected animals were grouped into four in western Lowland goats and five in Abergelle goats for each sex with three animals per group. Fifteen farmers for Western Lowland goats and 10 farmers for Abergelle goats who have not known the selected animals before were invited to rank the animals. Each farmer was asked to rank the animals in each group and the reasons of ranking. After a first round of ranking, farmers were provided with the life histories attached to individual animals and asked whether they would consider re-ranking the animals."},{"index":2,"size":8,"text":"This procedure was continued until all groups covered."}]},{"head":"Designing and optimizing of alternative breeding programs","index":45,"paragraphs":[]},{"head":"Population structure and Selection pathways","index":46,"paragraphs":[{"index":1,"size":111,"text":"The community based one tier selection scheme was considered for both breeds as the optimal breeding program for both of the study areas. The community based breeding program is believed to be a more convenient breeding program for such type of production systems which are characterized as low-input system with poorly developed infrastructures (Sölkner et al., 1998;Kahi et al., 2005;Gizaw et al., 2009). The flocks from 30 households with the average of 26 breeding does per household were considered as one breeding unit for Abergelle goat, While the flocks from 60 households with the average of 5 breeding does per household was considered as one breeding unit for Western Lowland goats."},{"index":2,"size":176,"text":"Four selection groups were defined to indicate the selection pathways. A selection group is defined by both, type of parents (one sex) passing genes and type of offspring receiving their genes. The selection groups were Bucks to produce Bucks (B>B), Bucks to produce Does (B>D), Does to produce Bucks (D>B) and Does to produce Does (D>D). Strong selection of male animals was assumed. The assumption was that the genetic gain obtained through selection would be disseminated by the selected bucks. The female animals would be selected only for the replacement. The young bucks at the age of six months would be selected based on their own performance of growth and the information from their dams for others traits. Breeding bucks were assumed to be in use for two time units (two years). Early age of selection (six months weight) was assumed to be more appropriate for the existing production systems of the study areas where negative selection is very common. Farmers tend to sale the fast growing animal at early age to get attractive market price."}]},{"head":"PhD Thesis","index":47,"paragraphs":[{"index":1,"size":111,"text":"The important input parameters of the two breeds for modeling (running ZPLAN) are shown in table 7. The information for the input parameters were derived from the production system, morphological characterization and own flock ranking results of this study and published report of on-farm monitoring studies (Derbie, 2008;Tsegaye, 2009;Derbie and Taye, 2013). The number of proven (candidate) animals in each time unit (year) were projected using the reproductive parameters and survival rate of the breeds. For instance the numbers of proven male animals for Western Lowland goat were calculated as follows: Assuming that sixty participant farmers with the average of 5 breeding does and the total of 300 breeding does in  "}]},{"head":"Alternatives breeding programs","index":48,"paragraphs":[{"index":1,"size":105,"text":"Four different alternatives for each breed were proposed for evaluating optimal breeding program (Table 8). The alternatives were based on the variation of the number of the traits in the selection index (recording) while keeping all traits in aggregate breeding goal. The important considerations of the alternatives were to see the effect of the variation of the number of traits in the recording scheme (selection criteria) on the genetic gains of the individual traits as well as the aggregate response. Since the selection program will operate at village level, inclusion of all traits in recording scheme might not be feasible in technical and economical terms."}]},{"head":"Solomon Abegaz GUANGUL","index":49,"paragraphs":[{"index":1,"size":2,"text":"PhD Thesis "}]},{"head":"Genetic and phenotypic parameters","index":50,"paragraphs":[{"index":1,"size":186,"text":"The genetic and phenotypic parameters are presented in Table 9. Due to the population parameters of the study breeds lacking, the weighted heritability estimates of the traits from published reports of other local and exotic goats were used (Odubate et al., 1996;Bosso et al., 2007;Valencia et al., 2007;Rashidi et al., 2008;Chun-yan Zhang et al., 2009;Alade et al., 2010;Faruque et al., 2010;Mantaldo et al., 2010;Kebede et al., 2012a). The genetic and phenotypic correlations of the traits were obtained from published reports on sheep (Abegaz, 2002;Matika et al., 2003;Gizaw et al., 2007;Afolayan et al., 2009). The SAS (2009) program was used to describe the survey data. Chi-square or t-test was employed when required to test the independence of categories or to assess the statistical significance. Indices were calculated for ranked variables (reasons of goat keeping, selection criteria and production constraints). Indices were computed as: sum of (3x for rank 1 + 2x for rank 2 + 1x for rank 3) given for a given reason divided by the sum of (3x for rank 1 + 2x for rank 2 + 1x for rank 3) for overall reasons. Where:"}]},{"head":"Morphological data","index":51,"paragraphs":[{"index":1,"size":21,"text":"Y ijk = the observation on body weight, chest girth, body length and height at withers μ = the overall mean;"},{"index":2,"size":51,"text":"A i = the fixed effect of age (i = ≤ 6 months, 6-12 months, 1PPI, 2PPI, 3PPI, 4PPI) B j = the fixed effect of breed ( j =Western Lowland, Abergelle) (A × B) ij =the interaction effect of age with breed e ijk = the effect of random error."}]},{"head":"Solomon Abegaz GUANGUL","index":52,"paragraphs":[{"index":1,"size":2,"text":"PhD Thesis"}]},{"head":"SNP data","index":53,"paragraphs":[]},{"head":"Within-breed diversity","index":54,"paragraphs":[{"index":1,"size":50,"text":"Allelic frequencies, number of polymorphic loci (MAF>0.05), observed heterozygosities (Ho), and expected heterozygosities (He) were obtained using Arlequin version 3.1 (Excoffier et al., 2005). The same package was used to test deviation from Hardy-Weinberg equilibrium using a Markov chain of steps100 batches, 5,000 iterations per batch, and 10,000 dememorization step."}]},{"head":"Population differentiation and cluster analysis","index":55,"paragraphs":[{"index":1,"size":40,"text":"Analysis of molecular variance (AMOVA) was calculated using the Arlequin version 3.1 software (Excoffier et al., 2005). Population differentiation was measured using F-statistics by calculating F IS (inbreeding within the population), F ST (the degree of gene differentiation among population)"},{"index":2,"size":85,"text":"and F IT (inbreeding in a group of populations). F statistics (F IS , F ST and F IT ) known as fixation index were calculated according to the methods of Weir and Cockerham (1984) using GENEPOP package version 4.2.1 (Rousset, 2008). The significance of pair wise population differentiation and significance of fixation index was tested using permutation test (10000 permutations). Reynolds standard population pair-wise genetic distance matrix (Reynolds et al., 1983) was computed as implemented by Arlequin version 3.1 software (Excoffier et al., 2005)."},{"index":3,"size":118,"text":"Neighborhood-joining dendrogram was constructed based on the Reynolds genetic distance using Unweighted Pair Group Method Arithmetic Averages (UPGMA). MEGA software version 5.2 was used to construct the tree (Tamura et al., 2008). Principal components analysis (PCA) was performed by GENABEL package in R (http://cran.r-project.org) and the principal components were calculated from all allele frequencies for each population. The population structure and admixture was evaluated based on a Bayesian clustering analysis by employing the STRUCTURE program (Pritchard et al., 2000). The samples were clustered with number of genetic clusters, K ranging from 2 to 5, applying 5 independent runs for each of the different values of K, with burn-in period of 5000 iterations and run length of 10,000 iterations."}]},{"head":"Runs of homozygosity (ROH)","index":56,"paragraphs":[{"index":1,"size":92,"text":"ROH are contiguous lengths of homozygous genotypes that are present in animal due to both parents transmitting identical haplotypes to their offspring (Purfield et al., 2012). The degree and frequency of these may inform on the ancestry of an individual and its population. The longer ROH segments are generated by inbreeding to a recent ancestor while the shorter ROH segments may also inform on the presence of remote ancestral inbreeding. (Purfield et al., PhD Thesis 2012;Ferencakovic et al., 2013a). In this study, ROH was detected by using CgaTOH (Zhang et al., 2013)  "}]},{"head":"Participatory identification of breeding objectives traits","index":57,"paragraphs":[{"index":1,"size":67,"text":"The statistical software SAS ( 2009) was used to analyzed the data from the own flock ranking and group ranking experiments. The proportion of traits preferred by the farmers in own flock ranking and group ranking experiment were analyzed by the frequency procedure. The body measurements and other traits from the life history were analyzed by glm procedure fitting the rank as fixed effects in the model."}]},{"head":"Optimizing alternative breeding programs","index":58,"paragraphs":[{"index":1,"size":112,"text":"Alternative breeding schemes were designed and evaluated using the computer program ZPLAN (Willam, et al., 2008). Using the gene flow method and selection index procedures, the program enables to simulate different breeding plans by deterministic approach. The program calculates genetic gain for the aggregate breeding value, the annual response for each trait and discounted return and discounted profit for a given investment periods. ZPLAN cannot consider reduced genetic variance due to selection (Bulmer effect) and inbreeding. Rate of inbreeding per generation (∆F) were calculated using a formula relating effective population size to use number of male (N m ) and number of female (N f ) breeding animals (Falconer and Mackay 1996);"},{"index":2,"size":2,"text":"PhD Thesis"}]},{"head":"Results and Discussion","index":59,"paragraphs":[]},{"head":"Production system study","index":60,"paragraphs":[]},{"head":"General household characteristics","index":61,"paragraphs":[{"index":1,"size":28,"text":"The majority (91.7 %) of the respondents in both study areas were male. The mean (±standard deviation) age of the respondents was 42.50±12.01 and 42.00±12.29 years for Western"},{"index":2,"size":58,"text":"Lowland and Abergelle goat keepers, respectively. The mean family size was 5.40±1.85 and 6.30±2.34 for Western Lowland and Abergelle, respectively. This was higher than the national average of 4.80 (CSA, 2010b). In Western Lowland goat keepers the majority of respondents (56.6%) are able to read and write, whereas in Abergelle only 18 out of 60 farmers were literate."},{"index":3,"size":43,"text":"The relatively higher proportion of literate household heads for Western Lowland goat owners would be a good opportunity to implement a goat improvement program as it might be easier for them to record performance and pedigree information. The average land holding (5.03±2.78 ha)"},{"index":4,"size":35,"text":"of Western Lowland goat owners was significantly higher (P<0.05) than land holding of Abergelle goat owners (1.00±1.47 ha). These figures include only privately owned land for crop production. For grazing, communal grazing areas are used."}]},{"head":"Purpose of keeping goats","index":62,"paragraphs":[{"index":1,"size":255,"text":"Table 10 shows the purpose of keeping goats and their respective rank by study areas. Better understanding of the purposes of keeping goats is a prerequisite for defining breeding goals (Jaitner et al., 2001). The purpose of goat keeping identified in this study is in line with previous studies from Ethiopia and other African countries (Harpe and Abebe, 2008; Kosgey et al., 2008;Legesse et al., 2008;Assen and Aklilu, 2012). The role of goat as source of cash income was found to be the primary reason of keeping goats in both study areas with index values of 0.5 and 0.4 for Western Lowland goat keepers and Abergelle, respectively. Milk production was ranked as the second most important role in Abergelle, while consumption of goat milk was considered as a cultural taboo in Western Lowland goat breeders. The value of manure was ranked third in Abergelle, whereas for Western Lowland goat breeders it was ranked fourth. The highest index value means the highest importance Western Lowland goat breeders gave higher priority to meat production (rank 2) and savings (rank 3) compared with Abergelle goat keepers. These results clearly show that goat rearing is seen as an option to generate income through sale of slaughter animals, but also contributes to the household consumption through meat and milk production. Based on the above, the conclusion is made that the main breeding goal of Western Lowland goat breeders is to increase meat production for marketing and consumption whereas Abergelle goat breeders wish to increase meat as well as milk production."}]},{"head":"Livestock holding and flock structure","index":63,"paragraphs":[{"index":1,"size":53,"text":"This study revealed that farmers keep mixed livestock species.  The lower the proportions of the kids in the Abergelle area were due to seasonal kidding there. As the area is drought prone area, most of kidding was happened between November and December following the active mating at the wet season (June and July)."},{"index":2,"size":36,"text":"The ratio of breeding buck to breeding does was 1:7 for Western Lowland goat and 1:12 for Abergelle, which was higher than the recommended ratio of 1:25 for tropical traditional production system (Wilson and Durkin, 1988)."}]},{"head":"Selection criteria and breeding practice","index":64,"paragraphs":[{"index":1,"size":10,"text":"Selection criteria for breeding does and bucks are summarized in  "}]},{"head":"PhD Thesis","index":65,"paragraphs":[{"index":1,"size":208,"text":"In both study areas, mating was uncontrolled and random, since bucks were mixed with the does throughout the year. Most of the goat keeper respondents (91.67%) in Western Lowland practiced mixing of their flock during grazing period on average with 5 other flocks. However, in Abergelle only 15 percent of the respondents allowed their flocks to mix with other flocks during grazing. As explained by Kosgey et al. (2006) uncontrolled mating together with small flock size would increase the level of inbreeding. On the other hand, practice of mixing flocks would minimize the problem of inbreeding by increasing the chance of mating of unrelated animals (Jaitner et al., 2001). The implication of these results is that cooperative village level breeding scheme would be appropriate for Western Lowland goat breeders, while selection within individual flocks could be possible in Abergelle goat given the individual flock grazing practice and the large flock size. There is a significant (X 2 , P<0.05) difference in buck ownership between the two communities. Only 40% of the respondents of Western Lowland goat keepers had their own buck, however higher proportions of (86.67 %) Abergelle goat keepers had their own buck. The farmers, who had no buck, used bucks from their neighbors and grazing lands."},{"index":2,"size":15,"text":"Regardless of the communities, farmers kept bucks mainly for mating and later fattening and slaughter."},{"index":3,"size":127,"text":"Castration of bucks after mating/service was common practice in both study areas. Fattening was the most important reason of castration (77.5% for Western Lowland goat keepers and 82.32% in Abergelle). Castration to control mating and temperament were reported by a few respondents. The average age (2.10± 0.68 years) of castration for Western Lowland goat was significantly (P<0.001) lower than that of Abergelle (4.40 ± 1.05 years). Keeping of intact male in the flock for a prolonged period would increase the hazard of inbreeding through increasing the chance of mating of bucks with their daughters. The practice of castration reported in both communities would be good for implementing village level selection through avoiding of mating of unwanted bucks and it would also increase the value of culled bucks."}]},{"head":"Reproductive performance","index":66,"paragraphs":[{"index":1,"size":89,"text":"The average reproductive performances of goats as reported by the respondents are given in Table 13. There was a significant (P<0.001) difference between the two breeds for all aspects of reproductive performance considered. The better performance of Western Lowland goats may be due to the genetic superiority of the breed and/or better feed situation of the area. Age at first kidding reported in female Western Lowland goat (12.4 months) and Abergelle (15.5 months) goat were comparable to the report of 13.6 months for Metema area (Tsegaye, 2009) and 14.9"}]},{"head":"Solomon Abegaz GUANGUL","index":67,"paragraphs":[{"index":1,"size":48,"text":"PhD Thesis months for Abergelle (Deribe, 2008). The kidding interval, 6.29 months for Western Lowland and 8.28 months in Abergelle, observed in this survey was lower than given in earlier reports of 8.4 months for Western Lowland goat (Tsegaye, 2009) and 11.31 months for Abergelle goat (Deribe, 2008). "}]},{"head":"Production constraints","index":68,"paragraphs":[{"index":1,"size":180,"text":"A good understanding of the existing production constraints in the study regions is essential for planning appropriate interventions. In both study areas high prevalence of disease and parasites were mentioned by the goat owners as the most limiting factor for goat production. All respondents complained about the low efficiency of veterinary service provided by the government. Feed shortage and recurrent droughts were also identified as important constraints for Abergelle goat owners. Goat keepers moved their goats to other areas where enough feed was available as a possible mitigation strategy. Feed shortage was mentioned by only a few goat owners of Western Lowland. This is because the area receives good rain and there is a relatively large area of communal grazing land. Predators, input (mostly veterinary service), lack of improved genotypes, labour and capital, theft, lack of market and lacking extension service were also reported as limiting factors of goat production in both study areas. This result is in line with goat production constraints reported for Southern Ethiopia (Tibbo, 2000;Legesse et al., 2008) and Northern Ethiopia (Tsegaye, 2009;Assen and Aklilu, 2012)."}]},{"head":"Morphological characteristics","index":69,"paragraphs":[]},{"head":"Qualitative characteristics","index":70,"paragraphs":[{"index":1,"size":33,"text":"Qualitative characters observed for female and male goats of the two breeds are presented in Tables 14 and 15. The study revealed that the two breeds have a wide range of coat colors."},{"index":2,"size":267,"text":"Most of (54%) Abergelle goats have a plain coat pattern, while most (60%) of Western Lowland PhD Thesis goats show a mixture of different colors with patchy and spotted patterns. Red brown, brown and the combination of these colors with other colors are the predominant coat colors observed in Abergelle goat (Figures 2 and 3). White and the combination of white with other colors were the major coat colors of Western Lowland goat (Figure 4 and 5). Irrespective of breeds and sex groups, all observed goats had short and smooth hair. There is a very small number (3.36%) of animals of the Western Lowland goat breed, which have a long and coarse hair type. Wattles were found in Western Lowland goats (24.53%) and Abergelle goats (10.11%). A variation in the existence of ruff was observed between breeds and sex groups. Only 8.0% of males of Abergelle goats have a ruff and 42.28 and 5.11 % of males and females of Western Lowland goats have a ruff, respectively. Almost all males and females of Abergelle goats had horns and around 5% of Western Lowland goats were polled. Most of the qualitative characters of both breeds obtained in this study were in agreement with the results of FARM-Africa (1996).   There was a significantly higher within breed variation of body weight of Western Lowland goats compared to the Abergelle goats in natural scale for most of age categories. However there was no significant difference for many of the age categories except for age group 3PPI after transformation to the log scale (Table 17). The relatively higher variation observed in Western"}]},{"head":"PhD Thesis","index":71,"paragraphs":[{"index":1,"size":21,"text":"Lowland goat could be a larger scope for genetic improvement of Western Lowland goats through selection compared to the Abergelle goats. "}]},{"head":"Molecular diversity 4. 3. 1. SNP polymorphism and genetic diversity within breeds","index":72,"paragraphs":[{"index":1,"size":73,"text":"The proportion of polymorphic loci, number of loci deviate from HWE and genetic variability (i.e observed and expected heterozygosities) are presented in Table 18. All populations showed high levels of polymorphism. The level of SNP polymorphism found in this study is in agreement with the previous report for four goat populations based on 50k SNP analysis (Kijas et al., 2012). However, Kijas et al. (2009) reported lower levels of polymorphism for ovine breeds."},{"index":2,"size":262,"text":"Similarly Edea et al. (2013) found less than 86 % level of polymorphism for Ethiopian cattle population based on the analysis of 8K SNPs. The higher level polymorphic SNPs found for goat population than other livestock species could be explained that goats have higher diversity than others (Kijas et al., 2012). Alternatively, different choices for the selection of SNPs during quality control might strongly impact these results. In all studied populations, the average expected heterozygosity values were slightly higher than the observed heterozygosity. The average heterozygosity observed was 0.379, 0.384, 0.375, 0.355 and 0.378 for Abergelle, Western Lowland, Red Sokoto, West African Dwarf and Shale, respectively. Shale, the Nigerian goat and Western Lowland the Ethiopian goats showed higher diversity with the highest observed (0.387 and 0.384) and expected (0.392 and 0.384) heterozygosity levels, respectively. The lowest genetic diversity (Ho= 0.355 and He=0.300) was observed in West African Dwarf goat population. Differently from the present findings, higher hetrozygosity values were reported in various molecular diversity studies of goat population in tropics and elsewhere based on microsatellite markers (Hassen et al, 2012;Missohou et al., 2011;Agha et al., 2008;Fatima et al., 2008;Li et al., 2008). This variation is due to the difference in the nature of the markers used. Microsatellite markers have much higher numbers of alleles than SNPs. However, the heterozygosities values found in this study are comparable with what found in SNPs based studies on Asian native goat by Lin et al. (2013), on four goat populations by Kijas et al. (2012) and on sixteen European goat breeds by Pariset et al. (2009)."},{"index":3,"size":125,"text":"The majority of the SNP markers were at Hardy Weinberg Equilibrium. A 3.2, 3.3, 3.6, 3.0 and 3.4 percent of SNPs markers in Abergelle, Western Lowland, Red Sokoto, Sahel and West African Dwarf goats were found significantly (P≤0.05) deviated from HWE, respectively. The measure of loss of heterozygosities (F IS ) value revealed that low inbreeding coefficients within population. Ethiopian goats had lower inbreeding coefficients (Abergelle=0.0046, Western Lowland=0.0052) than the three Nigerian goat populations. The inbreeding coefficient (F IS ) value observed in Ethiopian goat populations in this study was lower than that reported for Ethiopian goats in study using microsatellite markers (Hassen et al., 2012). This deviation could be because of the difference in markers and sampling frame. However, the F IS values obtained"}]},{"head":"PhD Thesis","index":73,"paragraphs":[{"index":1,"size":24,"text":"for Nigerian goat are in the range of previously reported values for goats in the same region (Traore et al., 2009;Missohou et al., 2011)."}]},{"head":"2. Breed differentiation","index":74,"paragraphs":[{"index":1,"size":57,"text":"The analysis of molecular variance (AMOVA) revealed as a low but significant (p<0.05) overall F ST 0.063. That is, a high proportion of the variation (93.7 %) is derived from within population and only 6.3 % variation comes from across populations. The overall within population inbreeding coefficient (F IS =0.012) was found insignificant (P>0.05) while significant (P<0.001)"},{"index":2,"size":42,"text":"global level loss of heterozygosity (F IT =0.075) was found.  (Hassen et al., 2012). The small differentiation between the breeds within the same region might be explained by less selective breeding, uncontrolled matting and transfer of genetic material due to animal movement."}]},{"head":"Population structure and clustering","index":75,"paragraphs":[{"index":1,"size":83,"text":"The Neighbor-joining dendogram (Figure 6) cluster analysis based on Reynolds' standard genetic distance was consistent with the geographical location of the breeds. Two clear clusters PhD Thesis  The principal component analysis was performed based on allele frequencies to see the population cluster of the five goat population (Figure 7). Except for a few outliers, the membership of each cluster coincided with their geographical origin. The first axis explains 11% of the total variation and separated the Nigerian goat population from Ethiopian goat population."},{"index":2,"size":18,"text":"The second axis contributed 2.2% of total variation. It separated the goat populations within the same geographical location."}]},{"head":"Solomon Abegaz GUANGUL","index":76,"paragraphs":[{"index":1,"size":68,"text":"PhD Thesis shown in Table 20. 99 % and 88 % of the two Ethiopian goat populations Abergelle and Western Lowland were included in the same cluster, respectively. The West African Dwarf goat was grouped in its own cluster with an estimated membership of 80%. The Red Sokoto and Sahel goats were grouped in the same cluster but with the high level of admixture; 67 % and 70"},{"index":2,"size":61,"text":"PhD Thesis % of estimated membership, respectively. In general, the cluster analysis result from PCA, Neighborhood joined tree and STRUCTURE program reveled that the studied goat population were less differentiated according to their type and morphological classification. They were more differentiated based on their geographical location. This finding is supported by previous studies report (Chenyambuga et al., 2004;Missohu et al., 2006).  "}]},{"head":"Runs of homozygosity (ROH)","index":77,"paragraphs":[{"index":1,"size":54,"text":"A ROH is a long continuous homozygote segment at diploid state. Detecting autozygous segments is the most common tool for association study of disease in human genomics. And it is also useful in farm animals to calculate the inbreeding coefficients at different ancestral population especially when pedigree information is lacking (Ferencakovic et al., 2013a)."},{"index":2,"size":95,"text":"Moderat to high correlations were reported between pedigree inbreeding coefficients and genomic inbreeding in different studies of cattle populations (Ferencakovic et al., 2011;Ferencakovic et al., 2013a). ROH at different length categories (>1MB, >2MB, >4MB, >6MB, >8MB, >10MB and >16MB) were analysed for all breeds to detect autozygous segments over the whole genome. The proportions of animals with ROHs at a given length are given in Figure 9. At ROH length >1MB all animals in all five population showed ROHs segments. As the ROH run length increased, the proportion of animals with ROH segments substantially decreased."},{"index":3,"size":29,"text":"For  The genomic inbreeding coefficients of individual animals at different run lengths were calculated as the ratio of total ROH segments of individual animals to the total genome length."},{"index":4,"size":326,"text":"The average genomic inbreeding coefficients were very low for all breeds at all run lengths, ranging from 0.6 % to 5% (Table 22). For all breeds, as the length of run increased the genomic inbreeding decreased. The F ROH at 8 MB and 10 MB for all studied population were very close to the values of loss of heterozygosity (F IS ) obtained from F statistics analysis. According to Ferencakovic et al. (2013b) the F ROH estimate in the runs less than 4MB from low density SNPs (50K) are not very accurate because of capturing short segments that are not truly homozygous when analyzed at higher SNP density.   Among the studied breeds the Nigerian goats WAD and RSK had relatively highest genomic inbreeding coefficient compared to other breeds at all runs lengths. The small number of ROH segments and low F ROH at long runs found in this study indicated that the populations had no the recent inbreeding problems. Even though the F ROH estimates at shortest runs are less reliable, all studied populations had relatively higher ancestral generation inbreeding coefficients than recent generation inbreeding. The ROH length found in this study is much lower than what was reported for different cattle breeds. (Ferencakovic et al., 2013a;Ferencakovic et al., 2011;Purfield et al., 2012). This could be because of the differences in the procedure of data collection and the species and population differences. In the present study, the samples were purposely collected from distantly related animals to avoid biased results in genetic diversity study.  The list of preferred does traits by farmers from own flock ranking experiments are described in Table 23. Various traits were mentioned by the farmers as preferred traits of breeding does. The diverse traits as the selection criteria for breeding female animals for tropical countries are also well documented in many research reports (Jaitner et al., 2001;Bett et al., 2009;Alexandre et al., 2009;Duguma et al., 2011;Mirkena, 2011;Berhanu et al., 2012;Gebreyesus et al., 2013)."},{"index":5,"size":184,"text":"The traits of preference by the farmers reflect the multi-functional role of goat. Milk yield, drought resistance, body size, kid growth and twinning rate were frequently mentioned as the preferred traits of breeding does by Abergelle goat owners. Twinning rate (multiple birth), kids growth, body size, mothering ability and kidding interval were reported as important traits of preferences for breeding does by Western Lowland goat owners. Although there was similarity of the trait preference between the two systems, there was a big difference in preferences for some traits. For instance, Milk yield (20.47 %) and drought resistance (14.96 %) were mentioned as important traits by Abergelle goat owners but these traits were not mentioned at all by Western Lowland goat keepers. This result clearly associated with the breeding objectives and the agro ecology of the study area where milk from goat is an important human food and where moisture stress is prevalent and feed is scarce for most parts PhD Thesis of the year. Similar finding for goats and sheep in comparable environment have been reported (Duguma et al., 2011;Mirkena, 2011;Gebreyesus et al., 2013)."},{"index":6,"size":121,"text":"The preference of big body size and fast kid growth as the preferred attributes in both study areas are expected when the main purpose of keeping goat is for cash source. The animals with big size are highly demanded in market and fetch good local market prices. Higher preference values of body size for breeding animals were reported by similar studies in Ethiopia and elsewhere in the tropics (Berhanu et al., 2012;Mbuku et al., 2006;Kebede et al., 2012b;Duguma et al., 2011;Jaitner et al., 2001). Relatively higher twinning rate (20.26 %) as the preferred traits for Western low land goat might be the availability of enough feed throughout the year that support many animals compared with the dry highland of Abergelle area."},{"index":7,"size":133,"text":"Since the western Lowland goat breeder do not use goat milk, in this area does with high milk are considered as good mothers for their kids. Thus mothering ability (like; nursing behavior and good attachment with their offspring) has relatively higher proportion (15.42%) as ranking criteria for Western Lowland goat than for Abergelle goat owners. Reproductive performance such as kidding interval was also mentioned as important traits 9.45 % and 14.10% for Abergelle and Western Lowland, respectively. Shorter kidding interval will increase flock productivity by providing many animals for marketing and replacement. It would be also helpful for genetic improvement program by increasing selection intensity. However, the improvement of kidding interval through selection may be slow because of the low heritability of the trait and very seasonal kidding, particularly in the Abergelle area."},{"index":8,"size":268,"text":"Table 24 describes mean ± SE values for some reproduction and production traits and age of the ranked does. Age of the does, doe weight, number of kidding, number of kids born, number of kids weaned and twinning rate significantly (P<0.001) influenced the ranking decision of the farmers in both study areas. Milk yield also significantly (P<0.001) influenced the ranking decision of Abergelle goat owners. This result revealed that the farmers' decisions for ranking of breeding does were highly correlated with the performance of the given animals. In both areas, there was a logical trend in the mean values of the traits between 1 st best, 2 nd best, 3 rd best and inferior does. For instance in Abergelle goats, the magnitude difference between the 1 st best and inferior does in live weight, number of kids weaned and milk yield were 6.76kg, 5.10, and 0.42l /day, respectively. In Western Lowland goats, the differences between the two groups were 9.14 kg of live weight, 8.17 numbers of kids weaned; and 0.31 litter sizes. Appreciations PhD Thesis and using of farmers knowledge for selecting the best animals is possible option to start the breeding program where performance recording totally lacking.  ) were mentioned as important traits for ranking of bucks in Abergelle area. In both site, coat color has got higher value for buck selection than for does. This is in agreement with previous breeding objective traits identification studies where color and beauty related traits were considered as important traits of preference for male breeding animals than female animals (Wurzinger et al., 2006;Duguma;2011;Kassie et al., 2009;Berhanu et al., 2012). "}]},{"head":"Comparisons objective traits ranking before and after the provision of life history","index":78,"paragraphs":[{"index":1,"size":168,"text":"Table 27 summarizes the proportion of rank altered after the provision of the life history of the ranked animals. Lower rank alterations were observed in male animals. In Western Lowland goat: 91.66%, 91.66% and 90% bucks were kept their rank as 1 st , 2 nd and 3 rd , respectively. The corresponding proportion in Abergelle bucks were 98%, 90% and 98%. More substantial rank changes based on life history information were observed in female animals. In Western Lowland does, 16.66% and 6.66% change their ranks from 1 st to 2 nd and 1 st to 3 rd , respectively. The corresponding values in Abergelle goats were 10 % and 40%. The probable reason for this is that apart from physical observation (body size, color and body conformation) farmers tend to select female animals based on their reproductive performance and mothering ability. This revealed that the attached performance of individual animals highly influenced rank alteration breeding does compared to breeding bucks. Similar finding were reported in identification of"}]},{"head":"PhD Thesis","index":79,"paragraphs":[{"index":1,"size":121,"text":"breeding objective traits studies through phenotypic group ranking methods for sheep breeds in Ethiopia (Mirkena, 2011) and for Ankole cattle in Uganda (Ndumu et al., 2008). In general, in group ranking experiment farmers tend to select the animal based on physical appearance such as color, size and conformation and they overlook the other performance whereas in own flock ranking farmer give more emphasis on performance of the animals such as reproductive performance, mothering ability and milk yield. For tropical small ruminant production systems, identification of the traits related with both subjective and objective criteria are equally important to establish sustainable breeding programs. Thus it could be advisable to use both methods for identification of breeding objectives traits in similar production systems."}]},{"head":"Optimization of alternative breeding programs","index":80,"paragraphs":[]},{"head":"Breeding objectives and selection criteria","index":81,"paragraphs":[{"index":1,"size":155,"text":"The primary step in the evaluation of the efficiency of alternative breeding systems is definition of the breeding objective. As the breeding program would be implemented at community level, for each breed, only three traits with high preference by farmers and easy to measure were considered. A breeding goal with many traits and traits with no reliable data for heritability and phenotypic and genetic correlation seems unrealistic to implement at community based selection schemes (Wurzinger et al., 2008). In this study, the breeding objectives were derived from the preferred traits by the community from the production system study, own flock ranking and group ranking experiments (Table 28). Some of the traits such as drought resistance, kidding interval and color which had higher preference by farmers were intentionally excluded to avoid the complexity during implementation. The identified breeding objective traits from those studies were scaled to hundred percent (unity) and weighted for the ranking. The breeding"}]},{"head":"PhD Thesis","index":82,"paragraphs":[{"index":1,"size":107,"text":"objectives identified for Abergelle goat owners were: Body size, milk yield and mothering ability (kids survival), while the breeding objectives for Western Lowland goat owners were body size, twinning rate and mothering ability (kids survival). Two selection indexes, one for each breed were constructed. Index 1, to reflect the breeding objective of Abergelle goat breeders, included six months weight (for body size), daily milk yield (for milk yield) and proportion of kids weaned (for kid survival). Index 2 to reflect the breeding objective of western Lowland goat six months weight, number of kid born per does per year and proportion of kids weaned per does per year. "}]},{"head":"Economic values","index":83,"paragraphs":[{"index":1,"size":43,"text":"The relative economic weights based of the preference of the community were derived. Thus the relative economic weight were set for the selected breeding objectives trait using the result from production system (PS), own flock ranking (OFR) and group ranking (GR) studies results"},{"index":2,"size":349,"text":"of the breeding does. The proportion of the selected breeding objective traits of does from different study (production system study, own flock ranking and group ranking) were scaled to one hundred percent (unity) and weighted for ranking ((%Ps+%OFR+%GR)/3) (Table 29).The weighted proportion of the given traits were used as the relative economic weights. The phenotypic standard deviation of the traits were estimated from the result of morphological PhD Thesis characterization and enumeration of own flock ranking experiment of this study. The genetic standard deviations of the traits were also estimated by the ratio of phenotypic standard deviation to the heritability of a trait.   30. Those parameters were different among the different alternatives and breeds. For all traits considered, higher genetic gains were predicted for Western Lowland goats than the Abergelle goats. These variations were due to higher phenotypic variation of the traits, lower generation interval and better performance (such as high twinning rate) of Western Lowland goats. The highest genetic gain of 0.3676 kg per year for six month's weight was predicted for Abergelle goats in growth only scheme (alternative 4) while the lowest 0.3599 was obtained in the alternative 2. As expected the highest gain was simulated for six month weight from growth only alternative where only the information of growth was included in the selection index. The highest value 0.8724 kg annual genetic gain of the six months weight was simulated for Western Lowland goats from alternative 3 (growth and survival information in the selection index) whereas the lowest value of 0.8702 kg was simulated from alternative 2 (growth and twinning information in the selection index). The highest gain of six month weight from PhD Thesis Lowland goats. In both breeds, the differences of annual genetic gain of kid survival between different alternatives were very small. This is because of the low heritability of the trait and low correlation with other traits. Comparable results with the range of 0.00-0.007% were predicted from different alternatives for Kenyan dairy goat breeds (Bett et al., 2012) and the range of 0.009-0.01% for Ethiopian Afar sheep breed (Mirkena et al., 2012)."},{"index":3,"size":139,"text":"Very low genetic gains of twining rate were predicted from all alternatives for Western Lowland goats. Even negative gain was predicted from the alternative 3 and 4 where the twinning information was not included in the recording scheme. This is due to the low heritability the trait and low phenotypic and genetic correlation with other traits. In addition to this, selection intensity was mostly derived from the male path of selection thus the twinning rate performance information was obtained only from the dams of young bucks. Since recording of the twinning rate is very simple, it would be worthwhile to include the information of twinning rate in the recording and give more weight in breeding goal to avoid the loss of genetic gain of twinning rates which was reported as the most preferred traits in Western Lowland goat keepers."}]},{"head":"Evaluation criteria","index":84,"paragraphs":[{"index":1,"size":341,"text":"Table 31 depicts the important evaluation criteria simulated by ZPLAN program. The selection accuracies of obtained from different alternatives for both breeds were in the acceptable range 0.481 to 0.512. Relatively higher accuracy of selection 0.504 and 0.512 were obtained from Alternative 1 (all traits in selection index) for Abergelle goats and Western Lowland goats, respectively. This reflects as the information source increased in the selection criteria the accuracy also increased. The annual monetary genetic gains ranged between 16.42 to 17.57 Euro were predicted for Abergelle goats from the different alternatives whereas 25.96 to 26.06 were predicted for western Lowland goats. As the difference between the schemes was only by varying the information source in the selection index, there was no difference between the different alternatives in selection intensity and generation interval within the same breed. The differences of those parameters between the two breeds were connected with the difference of population size of the breeding does and the difference in reproductive performance of the breeds in input parameters. A selection accuracy of 1.99 and a generation interval of 2.88 years were predicted for Abergelle goats while the corresponding values for Western Lowland goats were 2.25 and 2.14 years. The discounted profit found in all alternatives and in both breeds was very high. It might not be appropriate to compare the alternatives in this study based on the discounted profit because the economic value attached to each trait is not in the real monitoring PhD Thesis term and only additional cost to the normal practice were considered as the cost. The relative economic weights based on farmers' preference were assigned as the economic weight. The rate of inbreeding per generation 0.4% and 1.3% were calculated for Abergelle and Western Lowland goats respectively. The higher inbreeding rate for Western Lowland goats could be explained by the small flock size per household. During the implementation period, increasing the participant farmers with in the village or implementing across village selection for Western Lowland goat breeds would be advisable to avoid the problem of inbreeding. "}]},{"head":"Practical implementation","index":85,"paragraphs":[{"index":1,"size":38,"text":"The community level alternative schemes were designed and predicted for smallholder goat farmer conditions. Community based breeding program is the breeding program implemented at the smallholder levels where the infrastructure is poor and low input production system prevails."},{"index":2,"size":164,"text":"Therefore, the organizational structure should be simple and the traits in the recording should also small in number to avoid complexity during implementation (Sölkner et al., 1998;Wurzinger et al., 2008;Gizaw et al., 2009). This study was aimed that to see how much genetic gain and economic return loss in aggregate breeding goals (breeding objectives traits) by varying the number of traits at selection criteria. Even though, relatively higher gain from the alternatives with more traits in the selection criteria, the magnitude of the loss in genetic gains and economic returns from the alternatives with single versus more traits in the selection index were very PhD Thesis small. For instance, the difference in annual monitoring genetic gain between all traits and one trait alternative for Abergelle goats were 1.154%. This result indicates that it might be good to start the community based breeding by using a few traits in the recording schemes with minimum loss of genetic gain of other traits in the breeding goal."},{"index":3,"size":267,"text":"The present prediction study gave an acceptable range of genetic improvement for breeding objectives traits from different alternatives in both breeds at the community level. However implementation of community based breeding program need some organizational set up at the community level and strong participation of the community members (Wurzinger et al., 2011).The alternative schemes should be presented to the farmers to choose the appropriate options and the farmers have to be aware of advantage of genetically improved goats. The willing farmers need to be organized in form of a breed association. Some rules and regulation should be set to run the program smoothly. The rule may include how to manage and use the selected bucks. Giving some incentive like treatment of sick animal will also help for the successful implementation by creating motivation on the farmers. As the response of selection is slow, integration of other livestock improvement technologies such as feed technologies and disease control strategies would be also appropriate. The analysis of similar running projects such as the community based sheep breeding programs in Ethiopia and taking the lesson from them will help for successful implementation ( Haile et al., 2011;Gizaw et al., 2013). The strong technical and financial back up from the public service like research institutes and agricultural extension system is very important especially at the beginning. One of the reasons for unsustainability of the breeding programs in the tropical countries is the interruption of the program immediately after the funding stopped (Kosgey et al., 2006). Therefore, the program should be designed in the way that it can be run by the community."}]},{"head":"Conclusions","index":86,"paragraphs":[{"index":1,"size":137,"text":"This study provides the basic insights into production system, breed characteristics, breeding Phenotypic characterization of this study indicated high variation within and between the studied breeds in qualitative and quantitative traits. Western Lowland goats are on an average not only bigger than Abergelle goats but also show considerably higher variation in body size. This indicates a large scope of genetic improvement by selecting best young males. Molecular diversity analysis from the SNPs markers revealed that substantial amount of the variation of the studied goat populations are explained by within breed variation. The existing higher variability within indigenous goat breeds would be useful for future genetic improvement through selection and breed conservation. In order to have sustainable breed improvements and breed conservation, it would be worthwhile to develop effective breeding methods that reduce the gene flow between breeds."},{"index":2,"size":248,"text":"Participatory identification of the breeding objective traits is a central idea of community based breeding programs. In our approach, live animal ranking, the wide range of traits and with magnitude difference between the study areas were identified as breeding objective traits which are the reflection of multifunctional roles of goats and the differences in relative importance of goats in the study areas. In own flock ranking method farmers tend to prefer the animals based on production and reproduction traits like body size, multiple births, milk yield, litter size and kidding interval whereas in group ranking methods farmers give more emphasis on observable characters such as body size, body conformation and coat color. Therefore, both methods can be used for identification of breeding objective traits in low input system where performance records of the traits are not available. However, it is advisable to use the combination of PhD Thesis methods for identification of breeding objective traits at smallholder level. However, breeding for a minimal number of traits should be performed in community based breeding programs to avoid complex and intractable schemes. The simulation results from different scenarios gave an acceptable range of genetic gains for breeding objective traits with small differences between alternative with more traits in selection index and a few traits in selection index. This indicates that it is possible to start a feasible community based breeding with growth only or very few traits in selection criteria with little loss of genetic gain in breeding goal traits."}]}],"figures":[{"text":"9 Table 4 . 11 Table 5 . management condition ................................................................................................................Heritability estimates for important traits of some of tropical goat population ...............Genetic and phenotypic correlation estimates between traits of some of African goat population .................................................................................................................................12 "},{"text":"Figure 1 -Figure 2 -Figure 3 -Figure 4 -Figure 5 -Figure 6 -Figure 7 -Figure 8 -Figure 9 -Figure 10 -Figure 11 - Figure 1-Maps of the study areas ..............................................................................................22 Figure 2-Adult male of Abergelle goat .......................................................................................40 Figure 3-Adult female of Abergelle goat ....................................................................................40 Figure 4-Young male of Western Lowland goat.........................................................................40 Figure 5--Adult female of Western Lowland goat .......................................................................40 Figure 6-UPGMA neighbor-joining tree based on Reynolds' distance of 5 African goat breeds .46 Figure 7-Principal component analysis of five African goat populations.....................................47 Figure 8-Population partitioning suggested by STRUCTURE program ......................................48 Figure 9-Proportion of the number of animals with ROH segments at different runs..................50 Figure 10-The average total length of ROH at different length of categories .............................51 Figure 11-The relationship of number of ROH segments to the total length in ROH of individual animals .....................................................................................................................................53 "},{"text":" Key words: Goat; Breed; Genetic diversity; SNP; breeding objectives; Breeding Programs; ist die Entwicklung von dörflichen Zuchtprogrammen für zwei Regionen Äthiopiens. In der Studie wurden die Produktionssysteme beschrieben, Selektionskriterien und Zuchtziele der Bauern identifiziert und eine phenotpyische und molekulare Charakterisierung der Western Lowland und der Abergelle Ziege vorgenommen. Insgesamt wurden 120 TierhalterInnen befragt und von insgesamt 534 Abergelle Ziegen und 476 Western Lowland Ziegen Körpermaße genommen. Die genetische Diversität wurde von beiden äthiopischen Ziegenrassen mit Hilfe von SNP-Daten erhoben und mit drei nigerianischen Ziegenrassen verglichen. In einem Ranking-Experiment mit verschiedenen Tieren wurden die Selektionskriterien erhoben. Alternative Szenarien von dörflichen Zuchtprogrammen wurden simuliert. Die Ergebnisse der Umfrage ergaben, dass in beiden Studiengebieten Ziegen vielfältige Funktionen haben. Die phenotypische Charakterisierung zeigte, dass bei beiden Rassen eine hohe Variabilität sowohl innerhalb als auch zwischen den beiden Rassen besteht. Western Lowland Ziegen sind nicht nur größer als Abergelle Ziegen, sondern weisen auch eine höhere Variation in der Körpergröße auf. Die Studie zur genetischen Diversität zeigt, dass die Ziegenrassen sich klar voneinander abgrenzen. Produktions-und Reproduktionsmerkmale wie Körpergröße, Zwillingsrate und Milchleistung wurden als wichtige Merkmale in einem Rankingexperiment in der eigenen Herde identifiziert, während in einem anderen Rankingexperiment mit für die Befragten fremden Tieren andere Merkmale wie Körpergröße, Exterieur und Haarfarbe als wichtiger eingestuft wurden. Die Ergebnisse der Simulation zeigen in allen Varianten einen akzeptablen Zuchtfortschritt in allen Merkmalen. Es konnte gezeigt werden, dass auch mit der Erhebung von nur wenigen Merkmalen Zuchtfortschritte erzielt werden können. "},{"text":" develop appropriate community based breeding programs for Western Lowland and Abergelle goat breeds of Ethiopia with the following specific objectives:  To describe the production system  To describe the morphological characteristics  To characterize the genetic diversity of five African goat populations by genome wide (47K) SNP markers  To identify the breeding objective traits of Western Lowland and Abrgelle goat keepers by participatory live animal ranking approaches  To simulate the appropriate breeding schemes for two indigenous goat breeds in two different agro ecological zones of Ethiopia Thesis outline This thesis has five main sections. The first section is the introduction; it deals with the status of goat production in Ethiopia, limitations of goat production, goat genetic improvement interventions, justification of the study and the objectives of the study. The second part is the literature review which includes the goat genetic resources of Ethiopia and their production system, methods of breed characterization, the performance and the population parameters of important traits of Ethiopian and other tropical goats. The available genetic improvement strategies for tropical countries are also discussed. The materials used in this study and the methods that have been used to collect data, the statistical tools and analytical methods are presented in part three. The results and discussion of the study are provided in part four. The conclusions drawn from this study are presented in section five. The list of references cited and appendix are provided at the end. "},{"text":" With the development of new technologies, Molecular polymorphisms (nuclear DNA) have become the markers of choice for molecular-based surveys of genetic variation. Different types of markers are now available to detect polymorphisms in nuclear DNA. Randomly amplified polymorphic DNA (RAPD), amplified fragment length polymorphisms (AFLP), restriction fragment length polymorphisms (RFLP), microsatellites and single nucleotide polymorphisms (SNPs) have been developed and utilized "},{"text":" in review work summarized the main problems of upgrading tropical breeds with exotic breeds as follows:  Lack of clearly defined breeding policy  No owner participation  Too complicated in terms of logistics, technology and infrastructure  Indiscriminate crossbreeding  Lack of analysis of the different socio-economic and cultural roles that livestock play Therefore, in tropical countries, it is very useful to understand the level of performance and the potential of genetic improvement through selection within the indigenous breeds before implementing any breeding strategy (Philipsson "},{"text":" The breeding flocks are located within the production environment and potential genotype-environment interactions are therefore minimized  Direct farmer participation is possible  The farmers are owners of the initiative and benefit from it  The farmers have a sense of responsibility for the targeted breed, since it is a part of the traditional culture and contributes to their identity and self respect  Keeping of the targeted breed is economically important  Utilization of available feed resources  Maintenance is labor-intensive and not capital intensive  The initiative is self-administered by the community, but is supported by government and other organizations. The community therefore knows where to obtain information and technical advice. "},{"text":"Duguma et al. (2010) and Haile et al. (2011) recommended five participatory tools (Personal interviews, workshops, choice cards, and two types pf ranking experiments with live animals) to "},{"text":" study to address goat production in two different agro-ecological zones, farming systems and goat breeds. Metema district is located in a wet Lowland agro-ecological zone and in the North Western part of the country, 860 km from the capital Addis Ababa. The altitude ranges from 550 to 1608 m and the latitude of 12 o 40' N to 13 o 14' N. The rainfall pattern is unimodal with a mean annual range from 850 to 1100 mm, occurring from June to September(IPMS, 2005).Temperature ranges from a minimum of 22 o C to the maximum of 43 o C (IPMS, 2005). The production system is a mixed crop-livestock system with dominance of crop production as there is a high potential for biomass production. The dominant goat breed is Western Lowland (Gumuz). The second study area, Abergelle district is in the dry/sub-moist highland agroecological zone of the Northern part of the country, 780 km from Addis Ababa. The altitude ranges from 1150 to 2500 m with the latitude of 12 o 18'N to 13 o 06'N. The mean annual rainfall ranges from 250 to 750 mm, falling mainly from July to September(DOARD, 2010). The rainfall pattern is very erratic and uneven. Due to this erratic nature of rainfall, frequent crop failure and drought are common phenomena in the area. The production system is a mixed crop-livestock system with a focus on livestock, mainly Abergelle goat production. "},{"text":"Figure 1 1 . Figure 1-Maps of the study areas 3.2. Data collection 3.2.1. Description of production systems "},{"text":" the village selection scheme. Only 85% of them have kidding within one time unit (year), 0.6 year kidding interval, 1.5 liters per kidding, 1.67 kidding per year (time unit), 80% of survival rate and 50 % sex ratio. It gives 255 male selection candidates (300*.85*1.5*1.67*.8*.5=255).In this study, only the costs of additional activities to the normal management practices were considered as the cost parameters. Those were the cost of performance recording (Enumerator salary, cost of items for animal identification and cost of stationary materials) and cost of drugs.Those costs were calculated per individual breeding does per year. For instance the costs at Western Lowland goat were estimated as follow: a) Labor cost for recording: One enumerator for 300 does: Enumerator salary 24.59 €/month=295.08€/year; 295.08€/300 does; 0.98€/doe/year b) Identification cost: 2 tags/animal/year (a doe and her 2.5 kids/year) 0.2€/tag = 2*3.4*0.2 =1.36 €/does/year C) Drugs: 0.4 €/animal/year = 3.4 (a doe plus 1.13 kids/year)*0.4 =1.36€/does/year Solomon Abegaz GUANGUL PhD Thesis "},{"text":"Figure 2 -Figure 3 -Figure 4 -Figure 5 - Figure 2-Adult male of Abergelle goat "},{"text":" were observed: The first cluster consisted the two Ethiopian goats breed (Abergelle and Western Lowland) and the second group consisted the three Nigerian goats (West African Dwarf, Red Sokoto and Sahel). Furthermore, the Red Sokoto goats were more close to Sahel goats. According to DAGRIS (2007) Red Sokoto and Sahel goats are grouped under the some breed groups of short-eared twisted and subgroup of West African twisted horn. "},{"text":"Figure 6 - Figure 6-UPGMA neighbor-joining tree based on Reynolds' distance of 5 African goat breeds "},{"text":"Figure 7 - Figure 7-Principal component analysis of five African goat populations Figure 8 shows the graphical representation of genetic structure analyzed by STRUCTURE software. Four independent runs for the given K (K=2-5) were done to determine the optimum K-value. The most likely number of clusters was 3. In K=3 across runs, the average ln p(X/K) was maximum and the mean variance in lnp(X/K) was minimum. The Ethiopian goats, Abergelle and Western Lowland, weren not differentiated from each other. Except for a few admixture levels in Western Lowland goat, they clustered together as one group. The Nigerian goats tended to be clustered into two groups. That is, West African Dwarf goats as one separate cluster and admix population of Red Sokoto and Sahel goat as another group. The proportion membership of each predefined population in each of the 3 most likely inferred clusters is "},{"text":"Figure 8 - Figure 8-Population partitioning suggested by STRUCTURE program "},{"text":"Figure 9 - Figure 9-Proportion of the number of animals with ROH segments at different runs "},{"text":"Figure 10 - Figure 10-The average total length of ROH at different length of categories "},{"text":"Figure 11 - Figure 11-The relationship of number of ROH segments to the total length in ROH of individual animals "},{"text":"4. 5 . 3 . schemes to predict the genetic gain and the economic return is very helpful during implementation. It gives the chance to adjust the technical, infrastructural and socio economical issues ahead of the implementation. The predicted annual genetic gains (∆G) of individual breeding objectives traits from different alternative schemes of the two breeds are presented in Table30. Those parameters were different among the different alternatives and breeds. For all "},{"text":" objectives and selection criteria of Abergelle and Western Lowland goats in Ethiopia which are basic elements for community based breeding program planning. From this study it is possible to conclude that goat farming is an important component of the farming activity in the study areas by providing multifunctional roles to their owners. The high economic importance of goat with high flock size in Abergelle area and high potential of biomass production in Metema area would suggest the scope of goat improvements in both study areas. However, poor breeding practice like uncontrolled mating, mixing flock and negative selection of young bucks and high disease prevalence in both study areas and feed shortage in Abergelle area should be addressed. Thus, implementation of holistic community level development approach that considers the multifunctional roles of goats and existing goat production constraints is important. "},{"text":"  "},{"text":"  "},{"text":"  "},{"text":"  "},{"text":"Table 1 . Goat breeds of Ethiopia and their geographical distribution ..........................................5  "},{"text":"Table 2 . Birth, weaning, six months and yearling weight of some of Ethiopian goat breeds.........8  "},{"text":"Table 3 . Reproductive performance of some of Ethiopian goat breeds under different  "},{"text":"Table 6 . Advantages and disadvantages of the five breeding objective identification methods ..19   "},{"text":"Table 7 . Input parameters for modeling alternative breeding programs ....................................27  "},{"text":"Table 8 . Alternative breeding schemes for Abergelle and Western Lowland goats ....................28  "},{"text":"Table 9 . Phenotypic correlation (above the diagonal  "},{"text":" Birhanu, Hunegnaw, Bewket, Alubel, Adane and Natnael for their support during data collection. I would like to acknowledge the Sustainable Natural Resources Management Program in North Gondar for budget support of the field work. I am grateful to Teshome Mulu the head of the program for his encouragement during the field work. I am grateful to USAD-ARS for covering the cost of tissue samples collection, genotyping and providing of the SNPs data from Nigerian groups for accepting me as ILRI fellow student and dealing the tissue samples shipment. Special thanks to Rahel Misganaw, Eshitu Zerihun and Grum for arranging tissue samples shipment. I would like to acknowledge Dr. Barbara Rischkowsky from ICARDA for joining us in field work during tissue sample collection. I would like express my deepest appreciation to my friend Tesfaye Getachew for his unreserved friendship, his advice and valuable comments on every step of my work. He was with me for all my study period. I would like to extend my thanks for my brother Belay Derbie for his accompany during my flied work at Abergelle. I also want to extend my thanks to my friend Mengstie Taye for his willingness supporting me starting from the concept note writing. I would like to thank the former and the present Ethiopian student in Vienna; Dereje, Tigist, Tekeba, Asaminew, Kahsa, Kumela, Lemlem, Atsede, Tesfaye Feyssa, Amlaku, Zewdu, Gemeda, Muluneh, Kiflu, Mahilet, Meiwal, Diriba, Menale, Messeret and Hailemariam, for our memorable social life and making me feel at home. The support of Ana Maria Perez O'Brien and Solomon AMOVA Analysis of molecular variance AMOVAAnalysis of molecular variance CSA Central Statistical Authority CSACentral Statistical Authority DOARD District office of Agriculture and Rural Development DOARDDistrict office of Agriculture and Rural Development FAO Food and Agriculture Organization of United Nation FAOFood and Agriculture Organization of United Nation glm Generalized Linear model glmGeneralized Linear model He Expected heterozygosity HeExpected heterozygosity Ho Observed heterozygosity HoObserved heterozygosity HWE Hardy Weinberg Equilibrium HWEHardy Weinberg Equilibrium IPMS Improving market success and productivity IPMSImproving market success and productivity K Thousand KThousand MAF Minor Allele Frequency MAFMinor Allele Frequency MB Mega Base MBMega Base PCA Principal Component Analysis PCAPrincipal Component Analysis PPI Pair of Permanent Incisors PPIPair of Permanent Incisors ROH Run of Homozygosity ROHRun of Homozygosity RSK Red Sokoto goats RSKRed Sokoto goats SAS Statistical system analysis SASStatistical system analysis SHL Sahel goats SHLSahel goats SNP Single Nucleotide Polymorphism SNPSingle Nucleotide Polymorphism UPGMA Unweighted Pair Group Method with Arithmetic mean UPGMAUnweighted Pair Group Method with Arithmetic mean WAD West African Dwarf goats WADWest African Dwarf goats   x x Solomon Abegaz GUANGUL PhD Thesis Solomon Abegaz GUANGULPhD Thesis  "},{"text":"Table 1 . Goat breeds of Ethiopia and their geographical distribution  Family name Breed name Other Local name Distribution Family nameBreed nameOther Local nameDistribution Nubian family Nubian Barka, Begayit West Tigray Nubian familyNubianBarka, BegayitWest Tigray Rift valley family Afar Adal, Denakil Afar region, Northern and Western Rift valley familyAfarAdal, DenakilAfar region, Northern and Western    Hararghe Hararghe  Abergelle  Along Tekeze river Tigray region, AbergelleAlong Tekeze river Tigray region,    Wag Himra, East Gondar Wag Himra, East Gondar  Arsi-Bale Gishe, Sidamo Arsi, Bale and Western Hararghe Arsi-BaleGishe, SidamoArsi, Bale and Western Hararghe  Woyto-Guji Woyto, Guji, Konso South Omo, Southern Sidama and Woyto-GujiWoyto, Guji, KonsoSouth Omo, Southern Sidama and    Wolayita Wolayita Somali family Hararghe Highland  Hararghe Somali familyHararghe HighlandHararghe  Short -eared Somali Denghier Northern and Eastern Ogaden Short -eared SomaliDenghierNorthern and Eastern Ogaden  Long-eared Somali Degheir, Digodi, Ogaden, Lowland of Bale and Long-eared SomaliDegheir, Digodi,Ogaden, Lowland of Bale and   Melebo Borena MeleboBorena Small east Centeral highland Brown Centeral highlands, West of the rift Small eastCenteral highlandBrownCenteral highlands, West of the rift African family   valley Wollo, Gondar and Shoa African familyvalley Wollo, Gondar and Shoa  Western highland  Highland of South Gondar, Gojam, Western highlandHighland of South Gondar, Gojam,    Wollega and West Shoa Wollega and West Shoa  Western Lowland Gumuz Along the area bordering the Sudan Western LowlandGumuzAlong the area bordering the Sudan  Keffa  Highlands and Lowlands of Keffa KeffaHighlands and Lowlands of Keffa    and South Shoa Zone and South Shoa Zone Source: Farm-Africa (1996); Gizaw et al. (2010a)  Source: Farm-Africa (1996); Gizaw et al. (2010a)  "},{"text":"Table 2 . Birth, weaning, six months and yearling weight of some of Ethiopian goat breeds  Breed Management type BWT WWT SMW YWT Source BreedManagement typeBWTWWTSMW YWTSource Abergelle Traditional 1.91 6.84 9.13 14.25 Deribe and Taye (2013) AbergelleTraditional1.916.849.13 14.25 Deribe and Taye (2013) Western Lowland Traditional 2.28 12.00 NA NA Tsegaye (2009) Western LowlandTraditional2.2812.00NANATsegaye (2009) Central high land Traditional 2.32 7.17 9.30 13.04 Getachew et al ( 2006 ) Central high landTraditional2.327.179.30 13.04 Getachew et al ( 2006 ) Central high land Traditional 2.01 9.02 13.82 20.61 Deribe (2008) Central high landTraditional2.019.0213.82 20.61Deribe (2008) Arsi-Bale Station NA 6.95 9.00 14.31 Dadi et al (2008) Arsi-BaleStationNA6.959.00 14.31Dadi et al (2008) Arsi-Bale Station 1.91 6.65 9.00 14.32 Bedhane et al (2013) Arsi-BaleStation1.916.659.00 14.32Bedhane et al (2013) Arsi-Bale Traditional 2.80 8.39 NA NA Weldu et al. (2004) Arsi-BaleTraditional2.808.39NANAWeldu et al. (2004) Keffa Traditional 2.78 9.00 NA NA Shenkute, (2009) KeffaTraditional2.789.00NANAShenkute, (2009) Somali Station 3.19 11.67 NA NA Zeleke, 2007 SomaliStation3.1911.67NANAZeleke, 2007 Note: BWT= Birth weight; WWT=Weaning weight; SMWT= Sex months weight; YWT= Yearling weight; Note: BWT= Birth weight; WWT=Weaning weight; SMWT= Sex months weight; YWT= Yearling weight; NA= Not available       NA= Not available  "},{"text":"Table 3 . Reproductive performance of some of Ethiopian goat breeds under different management condition  Breed Management type LS AFK (days) KI (days) Source BreedManagement typeLSAFK (days)KI (days)Source Abergelle Traditional 1.04 448 339 Deribe (2008) AbergelleTraditional1.04 448339Deribe (2008) Central high land Traditional 1.42 NA 248 Getachew et al. ( 2006 ) Central high landTraditional1.42 NA248Getachew et al. ( 2006 ) Central high land Traditional 1.16 408 308 Deribe (2008) Central high landTraditional1.16 408308Deribe (2008) Arsi-Bale Station 1.64 854 293 Dadi et al. (2008) Arsi-BaleStation1.64 854293Dadi et al. (2008) Arsi-Bale Station 1.60 575 280 Kebede et al. ( 2012a) Arsi-BaleStation1.60 575280Kebede et al. ( 2012a) Keffa Traditional 1.70 375 237 Shenkute (2009) KeffaTraditional1.70 375237Shenkute (2009) Western Lowland Traditional  408 252 Tsegaye (2009) Western LowlandTraditional408252Tsegaye (2009) LS= Litter size; AFK=Age at first kidding; KI=Kidding interval   LS= Litter size; AFK=Age at first kidding; KI=Kidding interval  "},{"text":"Table 4 . Heritability estimates for important traits of some of tropical goat population  Trait Breed h 2 Country Source TraitBreedh 2CountrySource AFK Arsi-Bale 0.25±0.19 Ethiopia Kebede et al. (2012a) AFKArsi-Bale0.25±0.19EthiopiaKebede et al. (2012a) KI Arsi-Bale 0.06±0.08 Ethiopia Kebede et al. (2012a) KIArsi-Bale0.06±0.08EthiopiaKebede et al. (2012a) LSB Arsi-Bale 0.07±0.05 Ethiopia Kebede et al. (2012a) LSBArsi-Bale0.07±0.05EthiopiaKebede et al. (2012a) LSW Arsi-Bale 0.01±0.05 Ethiopia Kebede et al. (2012a) LSWArsi-Bale0.01±0.05EthiopiaKebede et al. (2012a) AFK Saanen 0.25±0.04 South Africa Muller (2005) AFKSaanen0.25±0.04South AfricaMuller (2005) LSB Zaraibi 0.08±0.01 Egypt Hamed et al.(2009) LSBZaraibi0.08±0.01EgyptHamed et al.(2009) LSW Zaraibi 0.05±0.01 Egypt Hamed et al.(2009) LSWZaraibi0.05±0.01EgyptHamed et al.(2009) LSB WAD 0.32±0.07 Nigeria Odubate (1996) LSBWAD0.32±0.07NigeriaOdubate (1996) KI WAD 0.03±0.01 Nigeria Odubate (1996) KIWAD0.03±0.01NigeriaOdubate (1996) LSB Sahle 0.39±0.09 Nigeria Alade et al. (2010) LSBSahle0.39±0.09NigeriaAlade et al. (2010) BW Arsi-Bale 0.09±0.08 Ethiopia Bedhane et al. (2013) BWArsi-Bale0.09±0.08EthiopiaBedhane et al. (2013) WW Arsi-Bale 0.03±0.08 Ethiopia Bedhane et al. (2013) WWArsi-Bale0.03±0.08EthiopiaBedhane et al. (2013) SMW Arsi-Bale 0.04±0.08 Ethiopia Bedhane et al. (2013) SMWArsi-Bale0.04±0.08EthiopiaBedhane et al. (2013) YW Arsi-Bale 0.02±0.10 Ethiopia Bedhane et al. (2013) YWArsi-Bale0.02±0.10EthiopiaBedhane et al. (2013) BW Draa 0.16±0.07 Morocco Boujenane and Hazzab (2012) BWDraa0.16±0.07MoroccoBoujenane and Hazzab (2012) WW Draa 0.11±0.06 Morocco Boujenane and Hazzab (2012) WWDraa0.11±0.06MoroccoBoujenane and Hazzab (2012) SMW Draa 0.01±0.08 Morocco Boujenane and Hazzab (2012) SMWDraa0.01±0.08MoroccoBoujenane and Hazzab (2012) BW WAD 0.50±0.05 Gambia Bosso et al. (2007) BWWAD0.50±0.05GambiaBosso et al. (2007) WWT WAD 0.43±0.07 Gambia Bosso et al. (2007) WWTWAD0.43±0.07GambiaBosso et al. (2007) YW WAD 0.30±0.07 Gambia Bosso et al. (2007) YWWAD0.30±0.07GambiaBosso et al. (2007) BW Nubian 0.54±0.05 Sudan Ballal et al. (2008) BWNubian0.54±0.05SudanBallal et al. (2008) WW Nubian 0.16±0.12 Sudan Ballal et al. (2008) WWNubian0.16±0.12SudanBallal et al. (2008) LL Arsi-Bale 0.03±0.15 Ethiopia Bedhane et al. (2012) LLArsi-Bale0.03±0.15EthiopiaBedhane et al. (2012) LMY Arsi-Bale 0.22±0.12 Ethiopia Bedhane et al. (2012) LMYArsi-Bale0.22±0.12EthiopiaBedhane et al. (2012) DMY Arsi-Bale 0.26±0.12 Ethiopia Bedhane et al. (2012) DMYArsi-Bale0.26±0.12EthiopiaBedhane et al. (2012) DMY Saanen 0.31±0.04 South Africa Muller (2005) DMYSaanen0.31±0.04South AfricaMuller (2005)  "},{"text":"Table 5 . Genetic and phenotypic correlation estimates between traits of some of African goat population Traits Breed Genetic correlation Phenotypic correlation Source  AFK with KI Arsi-Bale -0.43±0.01 0.05 Kebede et al. (2012a) AFK with KIArsi-Bale-0.43±0.010.05Kebede et al. (2012a) AFK with LSB Arsi-Bale 0.61±0.21 0.11 Kebede et al. (2012a) AFK with LSBArsi-Bale0.61±0.210.11Kebede et al. (2012a) AFK with LSW Arsi-Bale 0.34±0.14 0.05 Kebede et al. (2012a) AFK with LSWArsi-Bale0.34±0.140.05Kebede et al. (2012a) KI with LSB Arsi-Bale 0.69±0.18 -0.06 Kebede et al. (2012a) KI with LSBArsi-Bale0.69±0.18-0.06Kebede et al. (2012a) KI with LSW Arsi-Bale 0.59±0.02 -0.07 Kebede et al. (2012a) KI with LSWArsi-Bale0.59±0.02-0.07Kebede et al. (2012a) LSB with LSW Arsi-Bale 0.81±0.21 0.21 Kebede et al. (2012a) LSB with LSWArsi-Bale0.81±0.210.21Kebede et al. (2012a) LSB with LSW Zaraibi 0.91 0.63 Hamed et al.(2009) LSB with LSWZaraibi0.910.63Hamed et al.(2009) LSB with BW Sahel -0.25 -0.29 Alade et al. (2010) LSB with BWSahel-0.25-0.29Alade et al. (2010) LS with WW Sahel -0.15 -0.12 Alade et al. (2010) LS with WWSahel-0.15-0.12Alade et al. (2010) BW with WW Arsi-Bale 0.70±0.55 0.17 Bedhane et al. (2013) BW with WWArsi-Bale0.70±0.550.17Bedhane et al. (2013) BW with SMW Arsi-Bale 0.64±0.47 0.19 Bedhane et al. (2013) BW with SMWArsi-Bale0.64±0.470.19Bedhane et al. (2013) BW with YW Arsi-Bale 0.10±0.35 0.12 Bedhane et al. (2013) BW with YWArsi-Bale0.10±0.350.12Bedhane et al. (2013) WW with SMW Arsi-Bale 0.94±0.33 0.72 Bedhane et al. (2013) WW with SMWArsi-Bale0.94±0.330.72Bedhane et al. (2013) WW with YWT Arsi-Bale 0.52±0.50 0.56 Bedhane et al. (2013) WW with YWTArsi-Bale0.52±0.500.56Bedhane et al. (2013) SMW with YWT Arsi-Bale 0.57±0.43 0.65 Bedhane et al. (2013) SMW with YWTArsi-Bale0.57±0.430.65Bedhane et al. (2013) BW with WW Draa 0.58 0.27 Boujenane and Hazzab (2012) BW with WWDraa0.580.27Boujenane and Hazzab (2012) BW with SMW Draa 0.28 0.15 Boujenane and Hazzab (2012) BW with SMWDraa0.280.15Boujenane and Hazzab (2012) WW with SMW Draa 0.43 0.51 Boujenane and Hazzab (2012) WW with SMWDraa0.430.51Boujenane and Hazzab (2012) BW with WW WAD 0.74±0.08 0.30 Bosso et al. (2007) BW with WWWAD0.74±0.080.30Bosso et al. (2007) BW with YW WAD 0.73±0.14 0.19 Bosso et al. (2007) BW with YWWAD0.73±0.140.19Bosso et al. (2007) LL with LMY Arsi-Bale 0.43 0.15 Bedhane et al. (2012) LL with LMYArsi-Bale0.430.15Bedhane et al. (2012) LL with DMY Arsi-Bale -0.01 -0.19 Bedhane et al. (2012) LL with DMYArsi-Bale-0.01-0.19Bedhane et al. (2012) LMY with DMY Arsi-Bale 0.31 0.49 Bedhane et al. (2012) LMY with DMYArsi-Bale0.310.49Bedhane et al. (2012)  "},{"text":"Table 6 . Advantages and disadvantages of the five breeding objective identification methods  Properties Personal interviews Workshops Choice cards Ranking of live animals PropertiesPersonal interviewsWorkshopsChoice cardsRanking of live animals     Own animals Unknown for farmers Own animalsUnknown for farmers Advantages -A large number of -Information from -Large sample size -Relatively easy to -Easily done by farmers Advantages-A large number of-Information from-Large sample size-Relatively easy to-Easily done by farmers  persons can be different persons -Enumerator handle -Closer to reality than persons can bedifferent persons-Enumeratorhandle-Closer to reality than  interviewed collected at once introduced bias likely -Closer to reality than choice cards: seeing a live interviewedcollected at onceintroduced bias likely-Closer to reality thanchoice cards: seeing a live  -Possible to verify the -Differences can be to be lower than in choice cards: Seeing a animal is better than a -Possible to verify the-Differences can beto be lower than inchoice cards: Seeing aanimal is better than a  consistency of the directly discussed interviews live animal is better picture consistency of thedirectly discussedinterviewslive animal is betterpicture  responses  -Price can be than a picture  responses-Price can bethan a picture  -Additional information  included as a -Information from  -Additional informationincluded as a-Information from  can be gathered at the  characteristic different family  can be gathered at thecharacteristicdifferent family  same time   members can be  same timemembers can be     considered  considered Disadvantages -Language barrier -Some people (e.g. -Limited number of -There may not be -Large 'pool' of animals Disadvantages -Language barrier-Some people (e.g.-Limited number of-There may not be-Large 'pool' of animals  -Enumerator introduced with higher social animal profile choices enough animals of the often no readily available -Enumerator introducedwith higher socialanimal profile choicesenough animals of theoften no readily available  bias may be high status) might can be made per same category -Hypothetical life history bias may be highstatus) mightcan be made persame category-Hypothetical life history  -Important traits may not dominate the person available in small herds provided with a given -Important traits may notdominate thepersonavailable in small herdsprovided with a given  be mentioned discussion -Visual illustration of  animal may not be be mentioneddiscussion-Visual illustration ofanimal may not be    some traits can be  compatible with the visual some traits can becompatible with the visual    complicated or  appearance according to complicated orappearance according to    impossible  farmers' experience impossiblefarmers' experience Adopted from Haile et al. (2011)     Adopted from Haile et al. (2011)  "},{"text":" . After the above quality control measures46,885 autosomal SNPs   and 53, 41, 23, 22  and 22 animals for Abergelle, Western Lowland, West African Dwarf, Red Sokoto and Sahel goats were available for genetic diversity analysis, respectively. For runs of homozygosity (ROH) analysis, 44,721 SNPs were available after further pruned for MAF (<0.05) and for Hardy Weinberg Equilibrium (HWE=p<0.0001). 16,127 randomly selected SNPs were used for model-based clustering analysis using STRUCTURE software. According to(Frkonja et al., 2012) in a cattle admixture study, a small number of sub set SNPs (4000 SNPs from the 50K SNP chip) were sufficient to study breed composition. PLINK(Purcell et al., 2007) and R (http://cran.r-project.org) programs were used to arrange the SNP data. PGDSpider 2.0.1.4(Lischer and Excoffier, 2012) software was used to convert the data files into different programs formats.  "},{"text":"Table 7 . Input parameters for modeling alternative breeding programs Parameters Abergelle Western Lowland Population parameters  Population size(Does) Population size(Does)  "},{"text":"Table 8 . Alternative breeding schemes for Abergelle and Western Lowland goats  Alternatives Alternatives  "},{"text":"Table 9 . Phenotypic correlation (above the diagonal), genotypic correlation (below the diagonal) and heritability of the traits (along diagonal)  Traits  Abergelle   Western Lowland TraitsAbergelleWestern Lowland  SMW DMY PKW SMW NKB PKW SMWDMYPKWSMWNKBPKW SMW 0.28 0.1 0.1 0.28 0 0.1 SMW0.280.10.10.2800.1 DMY/NKB 0.2 0.32 0.14 0 0.10 0.15 DMY/NKB0.20.320.1400.100.15 PKW 0.3 0.53 .05 0.3 -0.20 0.05 PKW0.30.53.050.3-0.200.05 Note: SMW=Six months weight, DMY=Daily milk yield, PKW=Proportion of Kids weaned, NKB=Number of Note: SMW=Six months weight, DMY=Daily milk yield, PKW=Proportion of Kids weaned, NKB=Number of kids born       kids born  "},{"text":" software by applying the following ROH definition; the minimum number of SNPs needed to define a segment as ROH, 20; number of missing calls allowed, 5; number of needed to define a segment as ROH, 20; number of missing calls allowed, 5; number of heterozygous call allowed, 1; maximum gap between consecutive homozygous SNPs, 1 Mb; heterozygous call allowed, 1; maximum gap between consecutive homozygous SNPs, 1 Mb; minimum length of ROH set for >1, >2, >4, >6, > 8, >10, and >16 MB to see inbreeding at minimum length of ROH set for >1, >2, >4, >6, > 8, >10, and >16 MB to see inbreeding at different ancestral generations. A genomic inbreeding (F ROH ) was derived as the ratio of the different ancestral generations. A genomic inbreeding (F ROH ) was derived as the ratio of the length of genome present in ROH at specific run length to the total length of autozygous length of genome present in ROH at specific run length to the total length of autozygous genome covered by the consensus map of SNPs ( 2402.62MB). The SAS (2009) software was genome covered by the consensus map of SNPs ( 2402.62MB). The SAS (2009) software was used to describe the ROH results. used to describe the ROH results.  "},{"text":"Table 10 . Ranks of purpose for keeping goats  Purpose   Study communities    PurposeStudy communities  Western Lowland goat owners   Abergelle goat owners  Western Lowland goat ownersAbergelle goat owners   Rank    Rank   RankRank 1 st 2 nd 3 rd Index 1 st 2 nd 3 rd Index 1 st2 nd3 rdIndex1 st2 nd3 rdIndex Income 52 8 - 0.500 38 13 7 0.410 Income528-0.500381370.410 Manure - 1 3 0.014 8 14 21 0.205 Manure-130.014814210.205 Meat 3 31 25 0.270 1 4 13 0.067 Meat331250.27014130.067 Milk - - - - 12 28 15 0.300 Milk----1228150.300 Saving 5 20 25 0.220 - - 3 0.010 Saving520250.220--30.010 Skin - - 4 0.010 - - - - Skin--40.010----  "},{"text":"Table 11 . Age structure of goats in flocks of the different study communities  Age class  Study communities     Age classStudy communities Western Lowland goat owners (N=60)  Abergelle goat owners (N=60) Western Lowland goat owners (N=60)Abergelle goat owners (N=60) Mean (number of SD Range % Mean SD Range % Mean (number ofSD Range%MeanSDRange% goats)        goats) Doe 4.2 2.32 1-10 44.79 25.9 36.29 2-240 51.80 Doe4.2 2.321-10 44.7925.9 36.29 2-240 51.80 Buck 0.6 0.92 0-4 4.14 2.8 2.94 0-15 6.65 Buck0.6 0.920-44.142.82.940-156.65 Castrated 0.4 0.80 0-4 2.27 0.6 2.07 0-13 0.56 Castrated0.4 0.800-42.270.62.070-130.56 Young Buck 1.0 1.42 0-5 7.21 4.6 5.38 0-30 9.60 Young Buck1.0 1.420-57.214.65.380-309.60 Young Doe 1.6 1.81 0-8 12.80 6.6 7.33 0-35 12.89 Young Doe1.6 1.810-8 12.806.67.330-35 12.89 Kid 3.1 2.58 0-10 28.75 9.5 14.29 0-90 18.40 Kid3.1 2.580-10 28.759.5 14.290-90 18.40 young does (did not give birth) (12.8%), young bucks (not sexually active) (7.21 %), bucks young does (did not give birth) (12.8%), young bucks (not sexually active) (7.21 %), bucks (4.14%) and castrates (2.27%). A similar pattern was also observed in Abergelle with 51.8% (4.14%) and castrates (2.27%). A similar pattern was also observed in Abergelle with 51.8% breeding does, 18.4% kids, 12.89% young does, 9.60 young bucks, 6.65% bucks and 0.56% of breeding does, 18.4% kids, 12.89% young does, 9.60 young bucks, 6.65% bucks and 0.56% of castrates.        castrates.  "},{"text":"Table 12 . For WesternLowland goat owners, the most important selection criteria for breeding does were multiple PhD Thesis emphasis for multiple births as the preferred trait by Western Lowland goat keeper could be due to the high availability of the feed throughout the year and the breed potential. Around 1.6 litters per kidding were reported by Western Lowland goat breeders. For Abergelle goat owners, milk yield, body conformation and multiple births were ranked as first, second and third important selection criteria with index values of 0.32, 0.21 and 0.12, respectively. Drought resistance, coat colour, kidding interval, kid growth, mothering ability and pedigree information were also described as selection criteria. Body conformation followed by coat color were found as the most important selection criteria of breeding bucks in both study communities with the index values of 0.33 and 0.22 for Western Lowland goat keepers and 0 .31 and 0.25 for Abergelle, respectively. The preferred colors in Western Lowland goat breeders were white, red and patchy of those colors. The preferred colors for Abergelle goat breeders were red brown and red. Plain black was the less preferred color in both communities. Due to the relatively large flock sizes per household in Abergelle goat, farmers gave a high emphasis on sexual activity of breeding bucks. In general, goat owners in both study sites preferred size and other performance traits. The improvement of traits related with growth performance can be achieved easily through village level selection as the traits are easy to measure and have high heritability.  "},{"text":"Table 12 . Selection criteria for breeding does and bucks  Selection criteria Western Lowland goat owners Abergelle goat owners Selection criteriaWestern Lowland goat ownersAbergelle goat owners  1 st Rank 2 nd 3 rd Index 1 st Rank 2 nd 3 rd Index 1 stRank 2 nd3 rdIndex1 stRank 2 nd3 rdIndex Breeding does         Breeding does Body conformation(size) 4 15 13 .156 14 13  .211 Body conformation(size)41513.1561413.211 Twinning 32 9 7 .340 6 10  .124 Twinning3297.340610.124 Milk yield 2 - 1 .019 20 20 10 .317 Milk yield2-1.019202010.317 Mothering ability 7 12 8 .151 2 3 10 .063 Mothering ability7128.1512310.063 Kidding interval 4 7 3 .082 7 3  .086 Kidding interval473.08273.086 Kid growth 2 12 9 .110 3 4  .072 Kid growth2129.11034.072 Color Age of 1 st kidding 2 - 4 - 12 1 .074 .003 1 - 3 - 11 - .057 - Color Age of 1 st kidding2 -4 -12 1.074 .0031 -3 -11 -.057 - Drought resistance - - - - 5 1  .055 Drought resistance----51.055 Pedigree (Ancestor performance) 6 - 2 .056 - 1  .011 Pedigree (Ancestor performance)6-2.056-1.011 Breeding bucks         Breeding bucks Body conformation(Size) 20 25 5 .330 20 18 12 .310 Body conformation(Size)20255.330201812.310 Color 11 11 16 .225 9 24 16 .254 Color111116.22592416.254 Libido 2 3 4 .045 20 5 13 .232 Libido234.04520513.232 Growth rate 13 1 15 .216 3 8  .092 Growth rate13115.21638.092 Pedigree 8 3 11 .116 7 1  .075 Pedigree8311.11671.075 Horn - - 4 .017 - 2  .022 Horn--4.017-2.022 Drought resistance Age at 1 st mating -- 1 15 2 1 .011 .003 1 - 2 - - .030 - Drought resistance Age at 1 st mating--1 152 1.011 .0031 -2 --.030 -  "},{"text":"Table 13 . Reproductive performance of goats as reported by respondents in the surveyed area  Trait Trait  "},{"text":"Table 14 . Qualitative characteristics of Abergelle goat  Characters Male Female Total  CharactersMaleFemaleTotal  N % N % N %  N%N%N% Color Red brown 33 21.85 92 24.66 125 23.85 ColorRed brown3321.859224.6612523.85  White and brown 26 17.22 56 15.01 82 15.65 White and brown2617.225615.018215.65  Brown 36 23.84 67 17.96 100 19.66 Brown3623.846717.9610019.66  Red brown and white 17 11.75 50 13.4 67 12.79 Red brown and white1711.755013.46712.79  Brown and white 7 4.64 28 7.51 35 6.68 Brown and white74.64287.51356.68  Others 32 21.18 92 24.66 124 5.34 Others3221.189224.661245.34 Coat pattern Plain 81 64.8 203 54.42 284 54.2 Coat patternPlain8164.820354.4228454.2  Patchy and spotted 70 35.2 170 45.58 240 45.8 Patchy and spotted7035.217045.5824045.8 Hair type Short and smooth 151 100 373 100 524 100 Hair typeShort and smooth151100373100524100 Wattle Present 12 7.95 41 10.99 53 10.11 WattlePresent127.954110.995310.11  Absent 139 92.05 332 89.99 472 89.89 Absent13992.0533289.9947289.89 Ruff Present 42 27.81 - - 42 8.02 RuffPresent4227.81--428.02  Absent 109 72.19 373 100 492 91.98 Absent10972.1937310049291.98 N= Number of goats observed       N= Number of goats observed  "},{"text":"Table 15 . Qualitative characteristics of Western Lowland goat  Characters Characters  "},{"text":"Table 16 . Least squares means and standard error of body weight, body length and height at withers at different breed and age groups   Body weight(kg) Chest girth (cm) Body length(cm) Height at Wither (CM)  Body weight(kg) Chest girth (cm) Body length(cm) Height at Wither (CM) Level LSM SE LSM SE LSM SE LSM SE N LevelLSMSELSMSELSMSELSMSEN Breed ***  ***  ***  ***   Breed************ WL 24.00 0.19 65.27 0.23 54.80 0.21 62.60 0.22 340 WL24.00 0.1965.27 0.2354.80 0.2162.600.22 340 Abergelle 18.34 0.22 61.03 0.27 51.00 0.24 58.99 0.25 368 Abergelle18.34 0.2261.03 0.2751.00 0.2458.990.25 368 Dentition ***  ***  ***  ***   Dentition************ kids 10.35 a 0.42 48.84 a 0.51 41.26 a 0.46 49.50 a 0.49 95 kids10.35 a0.4248.84 a 0.5141.26 a0.4649.50 a0.4995 young 17.08 b 0.28 58.98 b 0.33 49.55 b 0.30 58.27 b 0.32 157 young17.08 b 0.2858.98 b 0.3349.55 b0.3058.27 b0.32 157 1PPI 21.91 c 0.41 64.62 c 0.50 54.36 c 0.45 62.43 c 0.47 66 1PPI21.91 c 0.4164.62 c 0.5054.36 c 0.4562.43 c0.4766 2PPI 23.07 c 0.41 66.36 c 0.49 55.28 c 0.44 63.42 cd 0.47 72 2PPI23.07 c0.4166.36 c0.4955.28 c0.4463.42 cd0.4772 3PPI 25.50 d 0.38 68.92 d 0.45 57.49 d 0.41 65.12 de 0.43 79 3PPI25.50 d0.3868.92 d0.4557.49 d0.4165.12 de0.4379 4PPI 29.13 e 0.22 71.19 e 0.26 59.44 e 0.24 66.02 e 0.25 239 4PPI29.13 e0.2271.19 e0.2659.44 e0.2466.02 e0.25 239 Age*Breed **  *  NS  NS   Age*Breed***NSNS Kid*WL 12.25 b 0.38 50.64 b 0.45 43.00 0.41 50.95 0.43 76 Kid*WL12.25 b0.3850.64 b0.4543.000.4150.950.4376 Kid*Abergelle 8.46 a 0.76 47.05 a 0.91 39.52 0.82 48.05 0-87 19 Kid*Abergelle8.46 a 0.7647.05 a 0.9139.520.8248.050-8719 Young*WL 20.04 d 0.45 62.13 d 0.55 51.85 0.50 60.22 0.53 52 Young*WL20.04 d0.4562.13 d0.5551.850.5060.220.5352 Young*Abergelle 14.11 c 0.32 55.82 c 0.39 47.24 0.35 56.32 0.37 105 Young*Abergelle14.11 c 0.3255.82 c 0.3947.240.3556.320.37 105 1PPI*WL 24.73 fg 0.63 66.36 e 0.77 56.37 0.69 64.59 0.73 27 1PPI*WL24.73 fg0.6366.36 e0.7756.370.6964.590.7327 1PPI*Abergelle 19.10 d 0.53 62.89 d 0.64 52.35 0.57 60.28 0.61 39 1PPI*Abergelle19.10 d0.5362.89 d0.6452.350.5760.280.6139 2PPI*WL 25.38 g 0.48 68.12 f 0.58 56.81 0.52 65.06 0.55 47 2PPI*WL25.38 g0.4868.12 f0.5856.810.5265.060.5547 2PPI*Abergelle 20.76 de 0.66 64.60 de 0.80 53.76 0.72 61.78 0.76 25 2PPI*Abergelle20.76 de0.6664.60 de0.8053.760.7261.780.7625 3PPI*WL 29.01 h 0.48 71.35 gh 0.58 59.58 0.52 66.89 0.55 47 3PPI*WL29.01 h0.4871.35 gh0.5859.580.5266.890.5547 3PPI*Abergelle 21.99 ef 0.58 66.50 ef 0.70 55.40 0.63 63.35 0.67 32 3PPI*Abergelle21.99 ef 0.5866.50 ef 0.7055.400.6363.350.6732 4PPI*WL 32.62i 0.34 73.02 h 0.41 61.18 0.37 67.88 0.40 91 4PPI*WL32.62i 0.3473.02 h0.4161.180.3767.880.4091 4PPI*Abergelle 25.64 g 0.27 69.35 fg 0.32 57.70 0.29 64.16 0.31 148 4PPI*Abergelle25.64 g0.2769.35 fg 0.3257.700.2964.160.31 148 Column means within each sub-class with different superscript letter are statistically differ. NS=Non Column means within each sub-class with different superscript letter are statistically differ. NS=Non significant, *=P0.05, **=P0.01, ***0.001 PPI=Pair of permanent incisors. SE=Standard error significant, *=P0.05, **=P0.01, ***0.001 PPI=Pair of permanent incisors. SE=Standard error WL=Western Lowland         WL=Western Lowland  "},{"text":"Table 17 . Variability of the body weight at different age groups  Age classes N WL SD Abergelle N SD P Value WL CV Abergelle CV P Value Age classesNWLSDAbergelle N SDP ValueWL CVAbergelle CVP Value Kid 76 2.61 19 1.55 0.0198 0.21 0.18 0.2642 Kid762.61191.550.01980.210.180.2642 Young 52 3.20 106 2.59 0.0748 0.16 0.18 0.7201 Young523.201062.590.07480.160.180.7201 1PPI 27 3.10 39 2.09 0.0342 0.12 0.10 0.3807 1PPI273.10392.090.03420.120.100.3807 2PPI 47 3.63 25 3.05 0.3198 0.18 0.14 0.8935 2PPI473.63253.050.31980.180.140.8935 3PPI 47 4.73 32 2.45 0.0028 0.16 0.11 0.0383 3PPI474.73322.450.00280.160.110.0383 4PPI 94 4.61 148 3.12 <0.0001 0.14 0.12 0.1127 4PPI944.611483.12<0.00010.140.120.1127 WL=Western Lowland goat; N=Number of animals; SD=Standard deviation CV=Coefficient of variation WL=Western Lowland goat; N=Number of animals; SD=Standard deviation CV=Coefficient of variation  "},{"text":"Table 18 . Proportion of polymorphic SNPs, observed heterozygosity(Ho), expected heterozygosity (He), population level loss of heterozygosity (F IS ) of five goats population  Breed N % of polymorphic loci Ho He F IS BreedN% of polymorphic lociHoHeF IS Abergelle 53 97.5 0.379±0.137 0.382±0.127 0.0046 Abergelle5397.50.379±0.1370.382±0.1270.0046 Western Lowland 41 98.5 0.384±0.138 0.387±0.123 0.0052 Western Lowland4198.50.384±0.1380.387±0.1230.0052 Red Sokoto 22 97.8 0.375±0.151 0.389±0.125 0.0302 Red Sokoto2297.80.375±0.1510.389±0.1250.0302 West African Dwarf 23 95.7 0.355±0.160 0.367±0.139 0.0300 West African Dwarf2395.70.355±0.1600.367±0.1390.0300 Sahel 22 98.3 0.387±0.151 0.392±0.123 0.0098 Sahel2298.30.387±0.1510.392±0.1230.0098  "},{"text":"Table 19 . Pair-wise genetic differentiation (FST) values between the five African goat (below diagonal) and Reynolds' genetic distance (above the diagonal) Tables19show the pair-wise genetic differentiation (F ST ) and Reynolds' distance between the five goat breeds. As expected, the pair-wise genetic differentiation (F ST ) values were very low within Ethiopian goats and Nigerian goats. The highest F ST values of 0.123 and 0.112 were observe between West African Dwarf and Abergelle and between West African Dwarf and Western Lowland goats, respectively, whereas the lowest values were obtained between Saheland Red Sokoto (0.001) and between Abergelle and Western Lowland (0.018) goats.Similar F ST values with the range of 0.025 to 0.098 in five Nigerian goats were reported byMissohou et al. (2011), 0.024 to 0.09 in Burkina Faso goats byTraore et al. (2009)  and the overall F ST value of 0.05 was also reported for five goat breeds in northern Ethiopia Population Abergelle Western Lowland Red Sokoto West African Dwarf Sahel PopulationAbergelle Western Lowland Red Sokoto West African Dwarf Sahel Abergelle - 0.017 0.083 0.131 0.066 Abergelle-0.0170.0830.1310.066 Western Lowland 0.018 - 0.071 0.118 0.058 Western Lowland0.018-0.0710.1180.058 Red Sokoto 0.079 0.068 - 0.035 0.007 Red Sokoto0.0790.068-0.0350.007 West African Dwarf 0.123 0.112 0.035 - 0.049 West African Dwarf0.1230.1120.035-0.049 Sahel 0.067 0.056 0.001 0.048 - Sahel0.0670.0560.0010.048-  "},{"text":"Table 20 . Proportion of analyzed goat populations in each of the three clusters (K=3)  Inferred cluster Inferred cluster  "},{"text":" instance, at run length of >16 MB the proportion of animals with ROHs for Abergelle, Western Lowland, Red Sokoto, Sahel and West African Dwarf were 9.4%, 7.3%, 31.8%, 30.4% Western Lowland, Red Sokoto, Sahel and West African Dwarf were 9.4%, 7.3%, 31.8%, 30.4% and 9.1 %, respectively. The descriptive statistics of the number of ROH segments >1MB and and 9.1 %, respectively. The descriptive statistics of the number of ROH segments >1MB and total length of ROH of the five goat populations are presented in Table 21 and the relationship total length of ROH of the five goat populations are presented in Table 21 and the relationship between number ROH segments and total ROH length of individual animals is depicted in between number ROH segments and total ROH length of individual animals is depicted in Figure 10 . At the ROH >1Mb West African Dwarf goats had the highest average number of Figure 10 . At the ROH >1Mb West African Dwarf goats had the highest average number of segments (42.48±42.48) and it also cover the longest ROH segments as well (120.17MB) while segments (42.48±42.48) and it also cover the longest ROH segments as well (120.17MB) while Shale goat had the lowest number of ROH segments (20.5±10.01) and the shortest ROH length Shale goat had the lowest number of ROH segments (20.5±10.01) and the shortest ROH length (61.68 MB). (61.68 MB).  "},{"text":"Table 21 . Descriptive statistics of number of ROH segments and total length of ROH of the studied goat population  Breed Breed  "},{"text":"Table 22 . Genomic inbreeding coefficients of five goat population at different runs length  Breed F ROH1 F ROH2 F ROH4 F ROH6 F ROH8 F ROH10 F ROH16 BreedF ROH1F ROH2F ROH4F ROH6F ROH8F ROH10F ROH16 Abergelle 0.0288 0.0136 0.0098 0.0088 0.0083 0.0072 0.0055 Abergelle0.02880.01360.00980.00880.00830.00720.0055 Western Lowland 0.0266 0.0142 0.0108 0.0085 0.0078 0.0067 0.0048 Western Lowland0.02660.01420.01080.00850.00780.00670.0048 Red Sokoto 0.0479 0.0388 0.0341 0.0326 0.0304 0.0282 0.0226 Red Sokoto0.04790.03880.03410.03260.03040.02820.0226 West African Dwarf 0.0500 0.0307 0.0268 0.0250 0.0242 0.0229 0.0191 West African Dwarf0.05000.03070.02680.02500.02420.02290.0191 Sahel 0.0257 0.0166 0.0138 0.0126 0.0114 0.0104 0.0080 Sahel0.02570.01660.01380.01260.01140.01040.0080  "},{"text":"Table 23 . List of traits of does from own flock ranking methods  Traits  Abergelle  Western Lowland TraitsAbergelleWestern Lowland  Freq % Freq % Freq%Freq% Milk yield 26 20.47 - - Milk yield2620.47-- Body size 18 14.17 35 15.42 Body size1814.173515.42 Drought resistance 19 14.96 - - Drought resistance1914.96-- Kid growth 15 11.81 42 18.50 Kid growth1511.814218.50 Twining 13 10.24 46 20.26 Twining1310.244620.26 Sex of kid 1 0.79 1 0.44 Sex of kid10.7910.44 Kidding interval 12 9.45 32 14.10 Kidding interval129.453214.10 Mothering ability 10 7.87 35 15.42 Mothering ability107.873515.42 Weight of kid at birth 8 6.30 7 3.08 Weight of kid at birth86.3073.08 Temperament 2 1.57 - - Temperament21.57-- Beauty 3 2.36 2 0.80 Beauty32.3620.80 Body length - - 8 3.52 Body length--83.52 Color - - 3 1.32 Color--31.32 Ear length - - 5 2.20 Ear length--52.20 Tail shape - - 1 0.44 Tail shape--10.44 Tail length - - 6 2.64 Tail length--62.64 Udder size   4 1.76 Udder size41.76  "},{"text":"Table 24 . Means ± SE of body weight and traits from the life history of the ranked animals  Breed Traits p  Rank   BreedTraitspRank    1 2 3 Inferior 123Inferior Western Lowland Age *** 5.47±0.19 a 3.88±0.18 b 2.90±0.18 c 2.61±0.18 c Western LowlandAge***5.47±0.19 a3.88±0.18 b2.90±0.18 c2.61±0.18 c  BW *** 34.04±0.65 a 31.02±0.64 b 27.17±0.64 c 24.90±0.64 c BW***34.04±0.65 a31.02±0.64 b27.17±0.64 c24.90±0.64 c  NK *** 5.83±0.22 a 3.73±0.22 b 2.78±0.22 c 2.18±0.22 c NK***5.83±0.22 a3.73±0.22 b2.78±0.22 c2.18±0.22 c  NKB *** 10.71±0.43 a 6.15±0.43 b 4.06±0.44 c 2.81±0.43 c NKB***10.71±0.43 a6.15±0.43 b4.06±0.44 c2.81±0.43 c  NKW *** 9.78±0.41 a 5.16±0.41 b 3.13±0.41 c 1.61±0.41 d NKW***9.78±0.41 a5.16±0.41 b3.13±0.41 c1.61±0.41 d  Twinning *** 1.83±0.04 a 1.61±0.04 b 1.37±0.05 c 1.19±0.04 c Twinning***1.83±0.04 a1.61±0.04 b1.37±0.05 c1.19±0.04 c Abergelle Age *** 6.33±0.29 a 4.93±0.29 cb 5.89±0.30 ab 4.73±0.29 c AbergelleAge***6.33±0.29 a4.93±0.29 cb5.89±0.30 ab4.73±0.29 c  BW *** 32.26±0.59 a 30.14±.59 b 30.40±0.61 b 25.50±0.61 b BW***32.26±0.59 a30.14±.59 b30.40±0.61 b25.50±0.61 b  NK *** 5.36±0.25 a 3.76±0.25 cb 4.41±0.26 b 3.06±0.25 c NK***5.36±0.25 a3.76±0.25 cb4.41±0.26 b3.06±0.25 c  NKB *** 6.76±0.39 a 4.33±0.39 b 4.58±0.40 b 3.10±0.39 c NKB***6.76±0.39 a4.33±0.39 b4.58±0.40 b3.10±0.39 c  NKW *** 6.40±0.36 a 3.90±0.36 b 3.96±0.37 b 1.30±0.36 c NKW***6.40±0.36 a3.90±0.36 b3.96±0.37 b1.30±0.36 c  Twinning *** 1.24±0.04 a 1.14±0.04 a 1.03±0.04 b 0.96±0.04 b Twinning***1.24±0.04 a1.14±0.04 a1.03±0.04 b0.96±0.04 b  Milk yield *** 0.58±0.03 a 0.48±0.03 ba 0.40±0.03 b 0.16±0.03 c Milk yield***0.58±0.03 a0.48±0.03 ba0.40±0.03 b0.16±0.03 c BW=Body weight; NK=Number of kidding; NKB=Number of kids born; the means in the same row with BW=Body weight; NK=Number of kidding; NKB=Number of kids born; the means in the same row with different superscripts are significantly different from each other   different superscripts are significantly different from each other 4.4       4.4  "},{"text":".2. Doe traits identified in group-animal ranking experiments Table25describes the list of preferred traits of breeding does in the group ranking experiment. The most important traits identified in Western Lowland goats were body size, body The most important traits identified in Western Lowland goats were body size, body conformation, coat color, twinning and udder size, which were accounted 21.71%, 15.43, conformation, coat color, twinning and udder size, which were accounted 21.71%, 15.43, 10.86% and 6.86%, respectively. The farmers in Abergelle mentioned body conformation (19.31 10.86% and 6.86%, respectively. The farmers in Abergelle mentioned body conformation (19.31 %), color (17.24%), mothering ability (17.24 %) and body size (13.79 %) as the most important %), color (17.24%), mothering ability (17.24 %) and body size (13.79 %) as the most important preferred traits of the breeding does. preferred traits of the breeding does.  "},{"text":"Table 25 . List of does traits in group ranking experiment   WL  Abergelle  WLAbergelle Traits Freq. % Freq % TraitsFreq.%Freq% Body conformation 27 15.43 28 19.31 Body conformation2715.432819.31 Body length 9 5.14 11 7.59 Body length95.14117.59 Body size 38 21.71 20 13.79 Body size3821.712013.79 Color 22 12.57 25 17.24 Color2212.572517.24 Beauty 2 1.14 2 1.38 Beauty21.1421.38 Height 8 4.57 12 8.28 Height84.57128.28 Ear length 3 1.71 - - Ear length31.71-- Horn shape - - 5 3.45 Horn shape--53.45 Horn length 2 1.14 2 1.38 Horn length21.1421.38 Tail shape 1 0.57 - - Tail shape10.57-- Tail length 7 4 1 0.69 Tail length7410.69 Mothering ability 6 3.43 25 17.24 Mothering ability63.432517.24 Milk yield - - 13 8.97 Milk yield--138.97 Kidding interval 7 4 1 .69 Kidding interval741.69 Twinning 19 10.86 - - Twinning1910.86-- Udder size 12 6.86 - - Udder size126.86-- Age 9 5.14 - - Age95.14-- WL=Western Lowland goat     WL=Western Lowland goat 4.4     4.4  "},{"text":".3. Buck traits identified in group ranking experiment  Table 26 shows the list of traits mentioned by farmers for ranking of breeding bucks and does in Table 26 shows the list of traits mentioned by farmers for ranking of breeding bucks and does in group ranking experiment of Western Lowland goats. The important attributes listed in group group ranking experiment of Western Lowland goats. The important attributes listed in group ranking of bucks for Western Lowland goat owners were coat color, body size, body ranking of bucks for Western Lowland goat owners were coat color, body size, body conformation and height with the magnitude of 23.86%, 21.02%, 10.23% and 9.09%, conformation and height with the magnitude of 23.86%, 21.02%, 10.23% and 9.09%, respectively. Similarly coat color (31.88%), body conformation (17.39 %) and body size respectively. Similarly coat color (31.88%), body conformation (17.39 %) and body size (14.29% (14.29%  "},{"text":"Table 26 . List of bucks traits in group ranking experiment   WL  Abergelle  WLAbergelle Traits Freq. % Freq % TraitsFreq.%Freq% Body conformation 24 17.39 18 10.23 Body conformation2417.391810.23 Body length 16 11.59 2 1.14 Body length1611.5921.14 Body size 19 14.29 37 21.02 Body size1914.293721.02 Color 43 31.88 42 23.86 Color4331.884223.86 Activeness 5 3.62 4 2.27 Activeness53.6242.27 Ear length - - 4 2.27 Ear length--42.27 Fast growth 2 1.45 14 7.95 Fast growth21.45147.95 Horn shape 2 1.45 - - Horn shape21.45-- Horn length 7 5.07 5 2.84 Horn length75.0752.84 Height 11 8.62 16 9.09 Height118.62169.09 Leg length 4 2.90 4 2.27 Leg length42.9042.27 Beauty - - 8 4.55 Beauty--84.55 Libido - - 1 0.57 Libido--10.57 Tail shape - - 4 2.27 Tail shape--42.27 Tail length - - 11 6.25 Tail length--116.25 Temperament - - 1 0.57 Temperament--10.57 WL=Western Lowland goat     WL=Western Lowland goat  "},{"text":"Table 27 . Rank proportion before and after provision of additional information in group ranking  Breed Rank 1  Rank 2 Buck   Rank 2 Does BreedRank 1Rank 2 BuckRank 2 Does   1 2  3 1 2 3 123123 Western lowlan 1 55(91.66%) 2(3.33%) 3(5%) 46(76.66%) 10(16.66%) 4(6.66%) Western lowlan 155(91.66%) 2(3.33%)3(5%)46(76.66%) 10(16.66%) 4(6.66%)  2 1(1.6%) 55(91.66%) 4(6.6%) 0(0) 46(76.66%) 14(23.33%) 21(1.6%)55(91.66%) 4(6.6%) 0(0)46(76.66%) 14(23.33%)  3 3(5%) 3(5%) 54(90%) 4(6.6%) 11(18.33%) 45(75%) 33(5%)3(5%)54(90%) 4(6.6%)11(18.33%) 45(75%) Abergelle 1 49(98%) 1(2%) 0(0%) 25(50%) 5(10%) 20(40%) Abergelle149(98%)1(2%)0(0%)25(50%)5(10%)20(40%)  2 5(10%) 45(90%) 0(%) 14(28%) 33(66%) 3(6%) 25(10%)45(90%)0(%)14(28%)33(66%)3(6%)  3 1(2%) 0(0%) 49(98%) 13(26%) 12(24%) 25(50%) 31(2%)0(0%)49(98%) 13(26%)12(24%)25(50%) 1 = Rank before additional information given 2 =Rank after provision of additional information 1 = Rank before additional information given2 =Rank after provision of additional information  "},{"text":"Table 28 . List of traits for breeding does from own flock ranking, group ranking and production system study  Traits  Abergelle   Western Lowland TraitsAbergelleWestern Lowland  OWFR PS GR WR OWFR PS GR WR OWFRPSGRWROWFRPSGRWR Milk yield 20.47 (2) .317(1) 8.97(3) 1.5 (2) -   Milk yield20.47 (2) .317(1) 8.97(3)1.5 (2) - Body size 32.28 (1) .283(2) 48.97(1) 1.25(1) 40.52(1) .307(2) 47.42(1) 1.5 (1) Body size32.28 (1) .283(2) 48.97(1) 1.25(1) 40.52(1) .307(2) 47.42(1) 1.5 (1) Drought resistance 14.96 (3) .055(6)  4.5(4) - - - - Drought resistance 14.96 (3) .055(6)4.5(4)---- Twining 10.24 (4) .124(3)  3.5 (3) 20.26(2) .340(1) 10.86(3) 1.25(2) Twining10.24 (4) .124(3)3.5 (3) 20.26(2) .340(1) 10.86(3) 1.25(2) Kidding interval 9.45(5) .086(4)  4.5 (4) 14.10(4) .082(5) 4.00(4) 4.5 Kidding interval9.45(5).086(4)4.5 (4) 14.10(4) .082(5) 4.00(4)4.5 Mothering ability 7.87(6) .063(5) 17.24(2) 5.5 15.42(3) .151(3)  3(4) Mothering ability7.87(6).063(5) 17.24(2) 5.515.42(3) .151(3)3(4) Temperament 1.57    -   Temperament1.57- Beauty 2.36    0.8   Beauty2.360.8 Color -    1.32  12.57(2) 2(3) Color-1.3212.57(2) 2(3) Ear length -    2.2   Ear length-2.2 Tail shape -    0.44   Tail shape-0.44 Tail length -    2.64   Tail length-2.64 Udder size     1.76   Udder size1.76 Note: OWFR= Own flock ranking, PS= Production system, GR=Group ranking, WR=Weighted ranking Note: OWFR= Own flock ranking, PS= Production system, GR=Group ranking, WR=Weighted ranking  "},{"text":"Table 29 . Economic weight and variance component of the selection criteria (traits)  Breeding  Unit REW   BreedingUnitREW objective traits Selection criteria     objective traitsSelection criteria Abergelle      Abergelle Body size Six month weight Kg 54% 1.45 2.74 Body sizeSix month weightKg54%1.452.74 Milk yield Milk yield kg 30 % 0.13 0.23 Milk yieldMilk yieldkg30 %0.130.23 Kid survival      Kid survival (mothering ability) Proportion of kids weaned/does/year % 16% 0.089 0.40 (mothering ability) Proportion of kids weaned/does/year%16%0.0890.40 Western Lowland      Western Lowland Body size Six months weight Kg 55% 1.99 3.76 Body sizeSix months weightKg55%1.993.76 Twinning Number of kid born /doe/year  31% 0.14 0.45 TwinningNumber of kid born /doe/year31%0.140.45 Kid survival      Kid survival (mothering ability) proportion of kids weaned /does/year % 14% 0.13 0.60 (mothering ability) proportion of kids weaned /does/year%14%0.130.60 RW-Relative economic weight; a -Additive genetic standard deviation; p -phenotypic standard  RW-Relative economic weight; a -Additive genetic standard deviation; p -phenotypic standard deviation      deviation  "},{"text":"Table 31 . Important evaluation criteria simulated from different alternative in Abergelle and Western Lowland goats  Alternative Criteria Abergelle Western Lowland AlternativeCriteriaAbergelleWestern Lowland 1 Accuracy of selection 0.503 0.512 1Accuracy of selection0.5030.512  AMGG 17.57 26.06 AMGG17.5726.06  Discounted profit/doe 138.85 213.29 Discounted profit/doe138.85213.29 2 Accuracy of selection 0.504 0.511 2Accuracy of selection0.5040.511  AMGG 17.51 26.05 AMGG17.5126.05  Discounted profit/doe 138.48 212.83 Discounted profit/doe138.48212.83 3 Accuracy of selection 0.484 0.511 3Accuracy of selection0.4840.511  AMGG 16.58 26.01 AMGG16.5826.01  Discounted profit/doe 133.24 212.99 Discounted profit/doe133.24212.99 4 Accuracy of selection 0.481 0.510 4Accuracy of selection0.4810.510  AMGG 16.42 25.93 AMGG16.4225.93  Discounted profit/doe 132.32 212.41 Discounted profit/doe132.32212.41  "}],"sieverID":"0c2b41ce-1a69-4ac9-9a51-15a5b53c5bfe","abstract":""}
\ No newline at end of file