Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Pedigree-based analyses of changes in genetic variability in three major swine breeds in Taiwan after a disease outbreak

Pedigree-based analyses of changes in genetic variability in three major swine breeds in Taiwan... Translational Animal Science, 2022, 6, 1–9 https://doi.org/10.1093/tas/txac043 Advance access publication 13 April 2022 Animal Genetics and Genomics Pedigree-based analyses of changes in genetic variability in three major swine breeds in Taiwan after a disease outbreak † ‡ ‖ † †,1 Ruei-Syuan Wu, Hsu-Chang Wang, Chan Liang Su, Pei-Hwa Wang, and En-Chung Lin Department of Animal Science and Technology, National Taiwan University, 10617 Taipei, Taiwan National Animal Industry Foundation, 10070 Taipei, Taiwan Dairy Association of Taiwan, R. O. C., 10644 Taipei, Taiwan Corresponding author: eclin01@ntu.edu.tw ABSTRACT Pedigree analysis was performed in three major Taiwanese swine breeds to evaluate the genetic variability in the current population and determine the main reason for genetic diversity (GD) loss after the occurrence of foot-and-mouth disease (FMD) in Taiwan. The pedigree files of the Duroc, Landrace, and Yorkshire breeds, containing 60,237, 87,177, and 34,373 records, respectively, were analyzed. We divided the population into two subpopulations (pre-1998 and post-1998) to determine the role of FMD in GD loss. Pedigree completeness and related indicators were analyzed to evaluate the pedigree quality, and several parameters were used to measure the levels of GD and further used to determine the major cause of GD loss. The pedigree completeness indexes for the different breeds were higher than 0.60, and the trend was enhanced after the FMD outbreak. The estimated proportion of random genetic drift in GD loss increased in all breeds over time (from 62.64% to 78.44% in Duroc; from 26.26% to 57.99% in Landrace; and from 47.97% to 55.00% in Yorkshire, respectively). The effective population size of Duroc and Landrace were increased by the time (Duroc: from 61.73 to 84.75; Landrace: from 108.70 to 113.64); however, it shows opposite trend in Yorkshire population (decline from 86.21 to 50.00). In summary, the occurrence of FMD led to the major loss of GD loss by random genetic drift. Therefore, for the recovery of GD, breeders in Taiwan should increase the effective population size with newly imported genetic materials and adjust the breeding strategy to reduce the inbreeding rate. Key words: genetic diversity, pedigree, random genetic drift, swine INTRODUCTION of genetic resources for future improvement and application. For instance, other countries have built up registry databases Genetic variability is used to represent the different genetic to collect comprehensive information for evaluating the ge- characteristics within or between breeds. In livestock, breeders netic variability of their national purebred swine populations typically use within-breed genetic variability as an indicator (e.g., USA: National Swine Registry and Canada: Canadian for allowing sustained genetic improvement for economi- Centre for Swine Improvement). cally important traits to enhance performance (Reist-Marti A purebred swine registry system was also established et al., 2003), and it is helpful to make breeding objectives in Taiwan in 1975, beginning with a primary focus on re- to meet particular market (Notter, 1999). However, due to cording superior imported livestock and continuing to collect the modification of production systems and breeding strategy, records of their mating, farrowing, progeny, and performance genetic diversity (GD) has been lost in swine breeds (FAO, testing to construct a complete pedigree with performance in- 2000; Barker, 2001). The GD, or the expected heterozygosity, formation. This information is useful for breeders (and some is a basic criterion for evaluating the genetic variability (Nei, producers) to select or cull of their breeding stocks. At the 1973). The reduction of GD is associated with various events, same time, tracking the selection strategy based on pedigree such as inbreeding, which is usually associated with a negative and performance information has also strengthened the func- influence on fitness-related traits ( Fernández et al., 2005) and tionality of the registry system (Sung, 1993). The registration limit the response of selection (Falconer and Mackay, 1996). of purebred herds has become routine in Taiwan today, and In closed populations, inbreeding leads to a higher rate of loss until 2017, there had been more than 200,000 registered pigs of alleles (Woolliams, 2007) due to unequal founder contri- (Fig. 1). However, there still has not been an in-depth study of bution. In contrast, even without inbreeding, the frequency the within-breed genetic variability of the Taiwanese purebred of alleles in a certain generation may randomly increase, swine population. Moreover, in 1997, there was an outbreak decrease, or even disappear in the subsequent generation. of foot-and-mouth disease (FMD) in Taiwan, which resulted Genetic drift can eliminate alleles, and the disappearance of in the culling of around 2.73 million pigs (25.5% of the entire such genetic variation can only be restored by mutation or number of heads raised) that year. Noticeably, this cull in- migration (Lacy, 1989). Therefore, understanding the genetic cluded about 470,000 purebred animals (31.9% of the pure- variability within a breed is necessary for confirming the value bred herd) (COA, 1997). Although breeders have continued Received November 10, 2021 Accepted March 31, 2022. © The Author(s) 2022. Published by Oxford University Press on behalf of the American Society of Animal Science. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. 2 Wu et al. to import new genetics since 1998, the total number of pure- Characterization of the Pedigree Information bred animals in the last two decades are only one-fifth of that There are four imported breeds (Duroc, Hampshire, Landrace, before 1998. Rapid and persistent declines in population size and Yorkshire) that have been used by breeders in the are important causes of inbreeding and loss of genetic vari- crossbreeding systems in Taiwan. Duroc and Hampshire are ability (Nei et al., 1975). Therefore, this study analyzes the the primary paternal breeds, and Landrace and Yorkshire are variation in the GD of the swine populations before and after mainly used to produce the commercial parental sow (PS) 1998 to understand the impact of FMD occurrence on the for future commercial herd production. Due to Hampshire- GD of purebred stocks in Taiwan. This study uses pedigree crosses have a higher incidence of pale soft exudative (PSE) data to present the current GD of the three major pig breeds compared to Duroc-crosses (Purchas et al., 1990), which in in Taiwan and identify the reasons for the loss of GD. turn leads to major economic losses for breeders, breeders in Taiwan gradually select Duroc breed as terminal sire breed over the Hampshire breed. Therefore, the registered number MATERIALS AND METHODS of Hampshire swine is significantly fewer than the other three breeds historically (Fig. 2). Thus, in this study, the analyses Animal Care and Use Committee approval was not obtained of genetic variability will focus on the Duroc, Landrace, and for this study because the study did not utilize any live animals. Figure 1. Number of pigs born and farm per year. Figure 2. Number of pigs born within different swine breeds per year. Analysis of three swine breeds from Taiwan 3 Yorkshire breeds. Pedigree records were obtained from the Pedigree Completeness The pedigree completeness local database maintained by the National Animal Industry levels of the two reference populations were evaluated by cal- Foundation and included all pedigree records available for culating (1) the number of fully traced generations, (2) the each breed up to 2017. To understand the influence of FMD maximum number of generations traced, and (3) the equiv- disease on the genetic variation of purebred pigs, the entire alent number of complete generations for each animal in the population was divided into two parts: the subpopulation pre- pedigree data. The first is defined as the number of genera - 1998, which includes pigs born between 1971 and 1997, and tions that separates the offspring from the furthest generation the subpopulation post-1998, which contains pigs born from where their ancestors of the individual are known. The second 1998 to 2017. For comparison of GD, some population with is the number of generations between an individual and its GD management was often selected as a reference point. This is furthest ancestor. Finally, the equivalent complete generation called the reference population, which can be some population is computed as the sum over all known ancestors of the terms in the past or the current population (Meuwissen et al., 2020). computed as the sum of (1/2) , where n is the number of gen- Previous literature has pointed out that average generation in- erations separating the individual from each known ancestor terval (GI) is used for reference population selection, which due (Maignel et al., 1996). Additionally, ENDOG offers a fourth to this reference population would comprises the last gener- indicator (4), the pedigree completeness index (PCI), which ation of data evaluated in each breed (Melka and Schenkel, describes the completeness of each ancestor in the pedigree of 2010). Similarly, present study also used average GI as the the parental generation (MacCluer et al., 1983). basis for the selection of the reference population. Since the Generation Interval The GI was computed for the four average GI of each breed in Taiwan is about 2.5 years, piglets possible selection pathways (sire–son, sire–daughter, dam– born in 1995–1997 and 2015–2017 were defined as reference son, and dam–daughter) as the average age of the parents at population for the pre-1998 and post-1998 subpopulation, re- the birth of the offspring used to replace them. spectively. The completeness of the pedigrees and the genetic variability were found by using pedigree information to calcu- The Parameters for Genetic Variability Evaluation late the associated parameters, such as pedigree completeness Inbreeding and effective population size The in- index, inbreeding coefficient and the parameters derived from dividual inbreeding coefficient (F), which is defined as the probability of gene origin, etc. The number of animals analyzed probability that an individual has two identical genes by de- in the entire pedigree and the reference population of each scent (Wright, 1922), was calculated according to the algo- breed is given in Table 1. Pedigree files collected for the Duroc, rithm described by Meuwissen and Luo (1992). The increase Landrace, and Yorkshire breeds consisted of 60,237, 87,177, in inbreeding (∆F) was calculated for each generation using and 34,373 records, respectively, for subsequent analyses. the classical formula, ∆ F = F F /1 − F , where F and F t— t-1 t-1 t t-1 are the average inbreeding at the tth generation. The effective Pedigree Quality Determination and Genetic population size (Ne) represents the relationships between the Variability Evaluation number of males and females that contribute genetically to the The ENDOG program (version 4.8) was used to identify the population. Due to the increase of inbreeding, this parameter pedigree quality and genetic variability of these three pure- was estimated in this study (Wright, 1931) based on ∆ F, as bred breeds (Gutiérrez and Goyache, 2005). Ne = 1/(2△F), for each generation having F > F , to roughly t t-1 Table 1. Basic description of pedigree datasets, and the parameters of pedigree completeness and inbreeding analysis for all breeds. Breed Duroc Landrace Yorkshire subpopulation Pre-1998 Post-1998 Pre-1998 Post-1998 Pre-1998 Post-1998 No. of animals in whole population 32,858 27,379 51,263 35,914 22,964 11,409 No. of animals in reference population 4,534 2,871 8,766 3,402 3,688 1,083 No. of Inbred animals 3,367 2,623 3,657 2,278 1,531 666 Inbred animals, % 74.26 91.36 41.71 66.96 41.51 61.49 Pedigree Completeness Index, % 0.83 0.86 0.64 0.73 0.61 0.69 Maximum generations traced 17 17 15 16 16 15 Mean equivalent generations 3.50 3.46 2.43 2.63 2.36 2.23 Complete generations 7 7 5 5 5 5 % known ancestors in: 1st generation 92.9 96.7 92.0 91.3 90.7 88.1 2nd generation 85.6 91.0 72.5 82.2 70.7 77.4 4th generation 73.5 74.8 43.6 54.8 34.4 49.3 Mean F, % 4.24 3.62 1.37 1.76 1.86 2.77 ∆ F, % 0.81 0.59 0.45 0.43 0.58 1.00 N 61.73 84.75 108.70 113.64 86.21 50.00 Inbred animals, the animals with F > 0; Mean F, average inbreeding coefficient of all animals in reference population. ∆ F, the increase in inbreeding of all animals in reference population; N , effective population size. e 4 Wu et al. characterize the effects of remote and close inbreeding. Ne 1—GD* represents the loss of GD due to unequal is defined as the number of breeding animals that would contributions by the founders to the population (Caballero increase inbreeding if they contributed equally to the next and Toro, 2000). Therefore, the difference between GD* and generation (Gutiérrez and Goyache, 2005). GD estimates the loss of diversity by genetic drift (Caballero and Toro, 2000; Honda et al., 2004). The probability of gene origin. Total number of founders (f) was defined as ancestors with unknown parents. RESULTS The effective number of founders (f ) means the number of The number of purebred pigs began to grow significantly in equally contributing founders that would be expected to gen- 1980 and remained stable before the outbreak of FMD in erate the same amount of GD as in the studied population Taiwan in 1997 (Fig. 1). However, in addition to the FMD (Lacy, 1989), was calculated using the following formula: outbreak in 1997, entry to the World Trade Organization   (WTO) in 2002 and a worldwide rise in grain prices in 2008 −1 caused a rapid decrease in the total number of pigs and   f = q breeders. Compared with the historical peak, the number of i=1 pigs born and the number of breeding farms have decreased by 66% and 76%, respectively, in the last 3 years. The inter- where q is the genetic contribution of the ith founder to the national situation and the influence of breeding conditions reference population and f is the total number of founders. The have caused Taiwan’s pig breeding industry to become con- f is usually lower than f, indicating the unequal contributions servative and gradually formed a large-scale, centralized, and of founders due to selection, which is the major mating op- corporate feeding operation. eration in the purebred industry. However, f alone may not be an useful indicator for assessing GD because the genetic Pedigree Completeness The PCIs of the reference contributions of the founders would converge after some populations of different breeds are shown in Table 1. Over number of generations (Bijma and Woolliams, 1999), and time, the PCI of the three breeds has increased. Comparison hence, the f would remain constant in the resulting popula- between the breeds indicated that the PCI of Duroc is the tion over time. The founder genome equivalent was defined as highest, while the completeness of the Yorkshire pedigree is the number of equally contributing founders with no random relatively low. Similarly, the maximum generations traced and loss of founder alleles that would be expected to give the same mean equivalent generations of the Duroc breed are also higher level of GD observed in the population under study (Lacy, than the other two breeds. The analysis of known ancestors in 1989), and it was computed as f = 1/(2 × f ), where f is the ge g g different generations shows that in the first generation, except average co-ancestry for the population considered (Caballero for the Yorkshire post-1998 subpopulation, more than 90% and Toro, 2000). The amount of GD in the reference popula- of the ancestors were known in other populations. But by the tion was calculated as (Lacy, 1995): fourth generation, only Duroc is higher than 70%, Landrace has fallen below 70%, and the value is less than 50% for the Yorkshire breed. In summary, the completeness of the GD = 1 − 2fge Duroc pedigree is better than that of the other two breeds. The requirements for pedigree integrity after the occurrence The loss of GD can be due to both genetic drift and unequal of FMD are also significantly higher than they were before the founder contribution; the value expressed as 1 − GD indicates occurrence of FMD. the amount of GD lost in the population since the founder generation due to genetic drift and the unequal founder con- Generation Interval Computed average GI and divided tribution. The amount of GD in the reference population into four selection paths in all breeds are shown in Table 2. accounting for the loss of diversity due to unequal founder As mentioned above, the average GI of all the breeds is over contribution (GD*) was calculated as (Lacy, 1995): 2 years, so the breeding stocks born in the most recent 3 years of each subpopulation are set as the reference popu- lation. The results indicate that the GI of the population be- GD = 1 − 2f fore the occurrence of FMD is shorter than that after the occurrence of FMD in all breeds. This might be due to the Table 2. Average generation intervals in evaluated breeds. Breed Duroc Landrace Yorkshire subpopulation Pre-1998 Post-1998 Pre-1998 Post-1998 Pre-1998 Post-1998 Average generation interval 2.17 2.43 2.30 2.63 2.28 2.60 GI for sire–son path 2.20 2.66 2.20 2.96 2.21 2.78 GI for sire–daughter path 2.19 2.43 2.17 2.58 2.19 2.56 GI for dam–son path 2.13 2.33 2.31 2.62 2.34 2.52 GI for dam–daughter path 2.16 2.41 2.43 2.64 2.38 2.61 GI, generation interval. Analysis of three swine breeds from Taiwan 5 reduction of the scale of production, which reduced the de- post-1998 subpopulation had the lowest N (50 animals), mand for breeding stocks, and lead to the breeding age of while the Landrace had the highest N in both subpopulations. breeding stocks is older. The lowest value of average GI was observed for Duroc breed (2.17 and 2.43 years) in pre-1998 The probability of gene origin. Table 3 summarizes all and post-1998 subpopulation, respectively. In contrast, the of the calculated parameters for the gene origin analysis of the longest GI are all obtained Landrace breed (2.30 and 2.63 reference population. Among the pre-1998 subpopulation, years) in different period. When comparing the different se- the highest number of founders were detected (f) for Landrace lection paths within each breed in pre-1998 subpopulation, (2,921), followed by Yorkshire (1,494), and Duroc (1,378). we can observe that the longest GI was the sire to son path However, in the post-1998 subpopulation, the Landrace and the shortest GI was found for the dam to son path in breed still had the highest number of founders, but the Duroc. On the other hand, there are the longest GI for the number for the Yorkshire breed was reduced to fewer than dam to daughter path and the shortest path for the sire to the Duroc breed. A similar trend also appeared for the effec- daughter path of the two maternal line breed. However, the tive number of founders (f ) and the equivalent number of longest path was all detected for the sire to son path in each founders’ genomes (f ). ge breed in post-1998 subpopulation. The highest difference The highest values of f and f were reached in the e ge among selection paths within one breed was observed in subpopulation of the Landrace breed before 1998 (457 and Landrace in different subpopulation (0.26 and 0.38, respec- 337) and the subpopulation after 1998 (169 and 71). The tively). The most balanced GI was detected for Duroc and lowest effective number of founders and founder genome Yorkshire (0.07 in pre-1998 subpopulation and 0.26 in post- equivalents were found in the Duroc and Yorkshire breeds 1998 subpopulation) when comparing the individual paths (174 and 65 founders; 80 and 36 founders, respectively) in of selection. different subpopulations. After the occurrence of FMD, the f, f , and f of all breeds decreased. In comparing different e ge The Parameters for Genetic Variability Evaluation subpopulations, the ratio of f /f shows a similar trend in Inbreeding and effective population size The sum- the Landrace and Yorkshire breeds; however, it increased in mary statistics for each breed are listed in Table 1. The results Duroc after the FMD outbreak. The highest value occurred showed that the proportion of inbred animals of all breeds in the Yorkshire breed (0.25) and the lowest value in Duroc has increased over time. In the Duroc breed, the proportion (0.13) in the pre-1998 subpopulation. In the post-1998 of inbred animals is significantly higher than in the other two subpopulation, Duroc and Yorkshire breeds had equal values breeds. However, although the proportion of inbreeding in the (0.16), while the Landrace had the lowest value (0.12). The analyzed breeds is high the value of the mean inbreeding co- ratio of f /f decreased in all the breeds in this study over ge e efficient (Mean F) is moderate, ranging from 4.24% (Duroc) time. The lowest values calculated were for the Duroc breed to 1.37% (Landrace) in the pre-1998 subpopulation and in different subpopulations (0.37 in pre-subpopulation and from 3.62% (Duroc) to 1.76% (Landrace) in the post-1998 0.22 in post-subpopulation, respectively). The accumulated subpopulation. With the occurrence of FMD, the average marginal contribution of 100 major ancestors is illustrated in increase of inbreeding (∆F) of each breed presented a different Fig. 3a–c for the different breeds. Independent of breed, more trend. The Duroc breed was reduced, and the Landrace breed ancestors are required to explain the gene origin in the pre- remained almost the same. However, ∆ F showed a dramatic 1998 subpopulation than in the post-1998 subpopulation. increase in the Yorkshire breed (from 0.58% to 1%). This One hundred ancestors can only explain about 50% of result can also be seen in the effective population size (N ). the gene pool in pre-1998 subpopulations of the Landrace Among the subpopulations pre- and post-1998, the Yorkshire and Yorkshire breeds. However, only half this amount is Table 3. Parameters derived from the probability of gene origin in the different reference populations in each breed. Breed Duroc Landrace Yorkshire subpopulation Pre-1998 Post-1998 Pre-1998 Post-1998 Pre-1998 Post-1998 Total number of founders, f 1378 1049 2921 1393 1494 515 Effective number of founders, f 174 167 457 169 369 80 Founder genome equivalent, f 65 36 337 71 192 36 ge f /f ratio 0.13 0.16 0.16 0.12 0.25 0.16 f /f ratio 0.37 0.22 0.74 0.42 0.52 0.45 ge e Number of ancestors to explain: 50% of gene pool 56 35 150 42 105 22 75% of gene pool 169 103 415 133 295 74 100% of gene pool 1143 680 2598 1042 1255 350 GD 0.992 0.986 0.999 0.993 0.997 0.986 1 − GD (GD loss) 0.008 0.014 0.001 0.007 0.003 0.014 Proportion of unequal contributions of the founders in GD loss 37.36 21.56 73.74 42.01 52.03 45.00 Proportion of random genetic drift in GD loss 62.64 78.44 26.26 57.99 47.97 55.00 GD, genetic diversity; The equation of GD calculation is GD = 1 − . 2f ge 6 Wu et al. Figure 3. Accumulated marginal contribution of 100 major ancestors in the Taiwan swine breeds. (a) Duroc breed; (b) Landrace breed; (c) Yorkshire breed. needed for this purpose in the Duroc breed. In the post-1998 some recent pedigree analyses of pigs. The completeness of subpopulation, fewer than 50 ancestors are enough to explain our analyzed Duroc pedigrees was similar to that published 50% of the gene pool in all evaluated breeds. Comparing the by Melka and Schenkel in four Canadian swine breeds and breeds indicated that Landrace contains the largest amount of more complete than the results reported by Tang et al. in three ancestors to explain its gene pool. imported swine breeds in China (Melka and Schenkel, 2010; Tang et al., 2013) However, the Landrace and Yorkshire PCIs were worse than the previous studies. These differences in DISCUSSION the quality of pedigree parameters between breeds found in From 1980 onward, the number of pigs and breeding farms our study might be caused by the dynamics and intensity of increased because Taiwan began to export purebred herds individual breeds used in commercial breeding programs. In to Southeast Asian countries, such as the Philippines and addition, some differences may also be based on the impact Thailand, and was a net exporter of prime pork meat to of imports or the depth of pedigree knowledge of imported Japan, with the offal left for domestic consumption (Sung, animals. These parameters ultimately affect the quality of the 1993). However, the outbreak of FMD led to consequences pedigree and thus the estimated inbreeding and GD. such as the inability to export pigs and pork, the development In general, the generation intervals of males are slightly of Taiwan’s purebred breeding pig industry stalled, and the shorter than that of females (Melka and Schenkel, 2010). population became closed. Therefore, this study uses pedigree Similar results can be found in previous reports (Melka and information and different parameters (such as F, N , f and Schenkel, 2010; Tang et al., 2013; Krupa et al., 2015). In this e e f ) to evaluate the GD of the population pre- and post-FMD study, the pre-1998 subpopulation also showed the same ge to understand the impact on the GD of the sharp decline in trend, except for the Duroc breed. However, in the post-1998 the number of purebred stocks and subsequent population subpopulation, each breed showed the opposite result, the closure. The accuracy of parameters such as inbreeding co- generation interval of males is longer than females. The gener- efficient, effective population size, and GD depends on the ation interval of both males and females increases over time. quality of the population pedigree. This study shows that This may be because after the occurrence of FMD, Taiwan’s the PCI of each breed can reach more than 60%, and the breeding pig industry shrank and could not be exported, de- percentage of known ancestors in the fourth generation mand for breeding stock has declined, so the mating strategy was higher than 70% in the Taiwan Duroc breed indicating has shifted to the purebred population was centralized derived a reasonable estimation of other pedigree parameters. from certain animals, which leading to older breeding age of Moreover, the correlated parameters of pedigree integrity males and females in the herd. This is because breeders have of the subpopulation in the breeds in this study before and no confidence in using new breeding animals, therefore would after the occurrence of FMD are very similar, so subsequent rather the current breeding stocks keep in production than comparisons can be made. Furthermore, the pedigree quality allow new generation to join or even replace them. This is of the Taiwanese pig population is moderate compared to evident in the individual selection paths, especially in the sire Analysis of three swine breeds from Taiwan 7 to son path of each breed. This situation might be improved the Yorkshire breeders have recently switched to the strategy through on-farm testing, which has gradually matured in of constructing the population by import instead of breeding recent years in Taiwan. By performing performance testing, native breed (the percentage of unknown parents in the regis- calculating estimated breeding values and establishing the se- tered Yorkshire population is equal to 13%) and centralizing lection index for various traits of purebred progeny, breeders by using this imported livestock to multiply the nucleus can identify the best animals and mating them in younger age. herd. The newly introduced breeding stocks are often mated It will helpful to shorten the generation distance and reduce with their progeny, causing their ∆ F rate to increase much breeding costs. faster than the other two breeds so that the usage of the The observed values of the mean inbreeding coefficient Yorkshire breed in Taiwan is very limited. However, a pre- over the breeds evaluated in this study were similar to the vious study demonstrated that the reproductive performance previous report. Krupa et al. (2015) have indicated a higher of Yorkshire × Landrace (two-way crossbred) sows was better inbreeding coefficient for the Czech Duroc breed (approxi - than that of purebred Landrace sows when both were mated mately 3.6%), while for the Czech Landrace and Czech with Duroc sires (Huang Y-H et al., 2002). This illustrated the Large White breeds, it has not exceeded 3%. Furthermore, advantage to incorporate the Yorkshire breed into the pro- the inbreeding degree of Landrace in Taiwan was lower than duction system. the Czech Landrace. The rate of inbreeding (∆ F) is an effec- The assessment of GD is especially important in highly tive criterion for measuring population health, and Nicholas specialized livestock breeds because assisted reproduc- has recommended a ∆ F rate < 0.5%, whereas the Food and tion techniques, such as artificial insemination and embryo Agriculture Organization of the United Nations (FAO) transfer technologies, can potentially rapidly reduce the GD has suggested a ∆ F rate < 1% as a goal (Nicholas, 1989; of a population (Vasconcellos et al., 2003). Parameters de- FAO, 2000). All breeds meet the FAOs goal in the different rived from the probability of gene origin analysis indicated subpopulations, but only the Landrace breed conforms to the an increased trend of GD loss with time in all three breeds in rate indicated by Nicholas. The FAO suggested that N for a this study. A relatively small number of major ancestors are breed should be maintained above 50 in order to withstand needed to explain the gene origin appears in the post-1998 the effects of inbreeding (FAO, 2000), while a size of 500 is subpopulation, which shows that the occurrence of FMD led essential to sustain the genetic variability and evolutionary to the decrease of GD. In general, the number of ancestors potential of the population for several generations (Frankham explaining the entire gene pool is still much greater than the et al., 2002). These authors defined at least 500 animals of results presented in the previous literature, showing that the Ne are needed to maintain GD in the population for several present Taiwanese purebred stocks maintain a fairly good GD generations. From the perspective of current and past effec- (Melka and Schenkel., 2010; Tang et al., 2013; Krupa et al., tive population sizes, efforts should be made for all the breeds 2015). analyzed in this study to enhance the N in order to construct To further explore the causes of GD loss, the f /f ratio e e a more varied population, although they are presently sim- represents the reduction of GD due to the unequal contribu- ilar to the results of previously published literature, and all tion of founders. The f /f ratio can be used to quantify only ge e of them meet the minimum requirement for N (50 animals) the influence of genetic drift on the amount of GD. In general, (Melka and Schenkel., 2010; Tang et al., 2013; Krupa et al., if all the founders were to contribute equally, the total number 2015). As the biotechnology has advanced, it allow using ge- of founders would be the same as f . However, the f is usu- e e nome information for the assessment and management of ally lower than f, indicating there are unequal contributions GD. Genomics data used for many alternative measures of of founders due to selection, namely that the breeders have inbreeding and genomic relationships. These measures are ap- preferentially chosen certain animals as parents (Melka and plied for managing GD in genome best contribution (GOC) Schenkel, 2010). The lower these two ratios are, the strong selection schemes (Meuwissen et al., 2020). By using genomic the effect of unequal contributions of founders or random information to measure different inbreeding coefficients and genetic drift is. The results of the probability of gene origin genomic relationships, such as drift or homozygosity-based in the reference populations show the f /f ratio was similar in inbreeding coefficients, it is possible to understand the true these three breeds at different periods, which means they may situation of allelic diversity (AD), and then choose the best be under the same degree of selection intensity in Taiwan. GOC management plan. The similar results can be observed Over the past years, knowledge of the production system has in previous study that different inbreeding coefficient value allowed the adjustment of the selection intensity for growth obtained by using genomic information and pedigree data and carcass traits (such as average daily gain and back fat) (Grossi et al., 2017). These results all prove that the measure- and reproductive traits (such as the longevity and litter size) ment derived only from the pedigree are not comprehensive in these three breeds. However, the Duroc breed presents the enough, and many details in it need to be more clearly under- lowest f /f ratio in both subpopulations, indicating the sig- ge e stood and applied through the use of genomic information. nificant effect of random genetic drift. Compared to previous Therefore, genomic-based analyses should be performed in studies of Canadian and Czech populations, the influence of our future studies to elucidate the real causes of GD loss and random genetic drift in Taiwan is milder (Melka and Schenkel, to identify strategies for GD management. The Ne of Duroc 2010; Krupa et al., 2015). In this study, the impact of random and Landrace increased with time, but Yorkshire decreased genetic drift was substantial for all breeds. The highest value with time, even more so in the post-1998 subpopulation. The of overall GD lost was observed for the Duroc and Yorkshire N of Yorkshire breed has been reduced to a critical value breeds in the post-1998 subpopulation. This situation appears (50 animals), which means that the ∆ F rate of the current to have occurred because the previous results indicated in the population has reached 1%. Though the Duroc breed still Czech population were similarly due to the number of an- has the highest mean inbreeding coefficient, the Yorkshire imals, along with the proportion of imported animals, was breed has shortened the distance. According to these results, reduced during the period of analysis (Krupa et al., 2015). 8 Wu et al. Before the occurrence of FMD, Duroc GD loss was mainly should pay more attention to the adjustment of breeding due to random genetic drift, while Landrace and Yorkshire operation methods for Yorkshire breeds, with the goal of GD losses were due to the unequal contributions of founders. develop a Taiwanese Yorkshire population that is adapted However, after the occurrence of FMD, the main cause of GD to Taiwan’s environment, cautiously use foreign breeding loss of the three breeds was random genetic drift. Although stocks in the production system. Maximize the effect of for- Duroc still has the highest degree of influence of random ge - eign breeding stock, which thereby reducing the rate of ∆ F netic drift (78.44%), the influence of random genetic drift in and increasing GD in population. Landrace has been significantly increased (from 26.26% to 57.99%), which indicates that the dramatic decline of this Acknowledgments population has had a greater impact than other two breeds on its GD. The main reason for such a high proportion of This work was financially supported by the Council of GD loss caused by random genetic drift (and other reasons Agriculture, Executive Yuan, Taiwan (Grant No.: 99 AS- such as the bottleneck effect) is the population shrinkage and 04.07-AD-04) in Taiwan. the declining proportion of introduced animals in Taiwan after the outbreak of FMD. Our study compared the dif- ferent subpopulations, which contrasted the historical and Conflict of interest statement current status of genetic variability to clarify the impact We certify that there is non conflict of interest involving any of FMD. Previous studies had indicated that GD loss due financial organization regarding the material discussed in the to random genetic drift led to an increase in homozygosity manuscript. and fixation of alleles ( Fernández et al., 2005; Toro et al., 2006). The loss of diversity can be caused by random genetic drift or unequal contributions of the founders. This study LITERATURE CITED confirmed that the selection of breeding stocks for specific Barker, J. S. F. 2001. Conservation and management of genetic diver- traits by breeders has a certain impact on the loss of popu- sity: a domestic animal perspective. Can. J. For. Res. 31:588–595. lation GD in Taiwan. However, after the outbreak of FMD doi:10.1139/x00-180. in Taiwan, the population size suddenly decline and the pro- Bijma, P., and J. A. Woolliams. 1999. Prediction of genetic contributions portion of imported animals decreased, resulting in purebred and generation intervals in populations with overlapping genera- populations mainly derived from breeding stocks with similar tions under selection. Genetics 151:1197–1210. doi:10.1093/ge- genetic backgrounds, which lead to an increase in the degree netics/151.3.1197. of random genetic drift. Therefore, to solve such problems, Caballero A. and Toro M. A. 2000. Interrelations between effec- breeding strategies should be adjusted, such as introducing tive population size and other pedigree tools for the manage- new genetic material and reducing inbreeding (Haile-Mariam ment of conserved populations. Genet. Res. Camb. 75:331–343. et al., 2007) to reduce the effects of drift (Lacy, 1987) to doi:10.1017/S0016672399004449 Council of Agriculture. 1997. Yearly Report of Taiwan's Agriculture further improve the level of GD. After long-term efforts, (1997). https://www.coa.gov.tw/ws.php?id=4563 (Traditional Chi- Taiwan finally was removed from the FMD epidemic area in nese version). June 2020, which provided another opportunities for pork Falconer, D.S. and T. F. C. Mackay 1996. Introduction to quantitative exports. In the future, Taiwan breeders should refer to the genetics. 4th ed. Longman Group Ltd., Essex, UK. results of previous and this study to formulate introduction FAO. 2000. Secondary guidelines for development of national farm an- and selection/breeding strategies to effectively reduce the im- imal genetic resources management plans: Management of small pact of random genetic drift, and increase the GD of the entire populations at risk. UN Food and Agric. Org. Rome, Italy. purebred stocks, for providing a stable supply basis in the Fernández, J., B. Villanueva, R. Pong-Wong, and M. A. Toro. 2005. Ef- export market. ficiency of the use of pedigree and molecular marker information in conservation programs. Genetics 170:1313–1321. doi:10.1534/ genetics.104.037325. CONCLUSION Frankham, R., D. A. Briscoe, and J. D. Ballou. 2002. Introduction to conservation genetics. University Press, Cambridge, UK. In summary, pedigree quality of Duroc breed is better than the Grossi, D. A., M. Jafarikia, L. F. Brito, M. E. Buzanskas, M. Sargolzaei, other breeds which might be related to that Duroc is used as and F. S. Schenkel. 2017. Genetic diversity, extent of linkage dis- sire and the influence exceed the other two maternal breeds. equilibrium and persistence of gametic phase in Canadian pigs. Therefore, for the Landrace and Yorkshire breeds, breeders BMC Genet. 18:6. doi:10.1186/s12863-017-0473-y. in Taiwan need to record their pedigree more carefully to Gutiérrez, J. P., and F. Goyache. 2005. A note on ENDOG: a computer obtain more accurate evaluation results. The results of GI program for analysing pedigree information. J. Anim. Breed. Genet. analysis indicated that Taiwanese breeders should adjust 122:172–176. doi:10.1111/j.1439-0388.2005.00512.x. their management that mating the new generation in younger Haile-Mariam, M., P. J. Bowman, and M. E. Goddard. 2007. A prac- tical approach for minimizing inbreeding and maximizing genetic age according to the result of progeny test to shorten the diversity in dairy cattle. Genet. Sel. Evol. 39:269–289. doi:10.1051/ generation interval. Genetic variability measurements show gse:2007009. that after the FMD outbreak, random genetic drift played Honda, T., T. Nomura, Y. Yamaguchi, and F. Mukai. 2004. Monitoring an important role in the loss of GD in Taiwanese swine of genetic diversity in the Japanese Black cattle population by the breeds. Breeders should reduce the inbreeding rate, intro- use of pedigree information. J. Anim. Breed. Genet. 121:242–252. duce genetically unrelated individuals and formulate better doi:10.1111/j.1439-0388.2004.00452.x. mating strategy (such as manage the breeding stock mating Huang, Y. -H., S. W. Roan, and Y. -P. Lee. 2002. Reproductive perfor- in younger breeding age) to increase the effective population mance of Landrace and that of Landrace × Yorkshire sows bred by size. The ∆ F rate of the Yorkshire breeds has increased sig- Duroc sires. J. Chin. Soc. Anim. Sci. 31:19–29. doi:10.1016/S0093- nificantly. Therefore, in addition to introduction, breeders 691X(00)00364-2. Analysis of three swine breeds from Taiwan 9 Krupa, E., E. Zakova, and Z. Krupová. 2015. Evaluation of inbreeding Mackay, editors. Evolution and animal breeding. CAB Int., Wal- and genetic variability of five pig breeds in Czech Republic. Asian- lingford, UK. p. 201–209. Australas. J. Anim. Sci. 28:25–36. doi:10.5713/ajas.14.0251. Notter, D. R. 1999. The importance of genetic variability in live- Lacy, R. C. 1987. Loss of genetic diversity from managed populations. stock populations of the future. J. Anim. Sci. 77:61–69. Interacting effects of drift, mutation, immigration, selection and pop- doi:10.2527/1999.77161x. ulation subdivision. Conserv. Biol. 1:143–158. doi:10.1111/j.1523- Purchas, R. W., W. C. Smith, and G. Pearson. 1990. A comparison of 1739.1987.tb00023.x. the Duroc, Hampshire, Landrace, and Large White as terminal sire Lacy, R. C. 1989. Analysis of founder representation in pedigrees: breeds of crossbred pigs slaughtered at 85 kg liveweight. N. Z. J. founder equivalents and founder genome equivalents. Zoo Biol. Agric. Res. 33:97–104. doi:10.1080/00288233.1990.10430666. 8:111–123. doi:10.1002/zoo.1430080203. Reist-Marti, S. B., H. Simianer, J. Gibson, O. Hanotte, and J. E. O. Rege. Lacy, R. C. 1995. Clarification of genetic terms and their use in the 2003. Weitzman’s approach and conservation of breed diversity: management of captive populations. Zoo Biol. 14:565–578. an application to African cattle breeds. Cons. Biol. 17:1299–1311. doi:10.1002/zoo.1430140609. Sung, Y. Y. 1993. Pure breed registration of swine in Taiwan. In: “Pork MacCluer, J. W., A. J. Boyce, and B. Dyke. 1983. Inbreeding and ped- Industry Conference on Breeding Strategies”, Taiwan Livestock igree structure in Standardbred horse. Int. J. Hered. 74:394–399. Research Institute; p. F5–Fl2. doi:10.1093/oxfordjournals.jhered.a109824. Tang, G. Q., J. Xue, M. J. Lian, R. F. Yang, T. F. Liu, and Z. Y. Zeng. Maignel, L., D. Boichard, and E. Verrier. 1996. Genetic variability of 2013. Inbreeding and genetic variability in three imported swine French dairy breeds estimated from pedigree information. Interbull breeds in China using pedigree data. Asian Australas. J. Anim. Sci. Bulletin 14:4–54. 26:755–765. doi:10.5713/ajas.2012.12645. Melka, M. G., and F. Schenkel. 2010. Analysis of genetic variability in Toro, M. A., J. Fernández and A. Caballero. 2006. Scientific basis for four Canadian swine breeds using pedigree data. Can. J. Anim. Sci. policies in conservation of farm animal genetic resources. Proceed- 90:331–340. doi:10.4141/CJAS10002. ings of the 8th World Congress on Genetics Applied to Livestock Meuwissen, T. H. E., and Z. Luo. 1992. Computing inbreeding Production. Communication no. 33-05. Belo Horizonte, Brazil. coefficients in large populations. Genet. Sel. Evol. 24:305–313. Vasconcellos, L. P. M. K., T. Daniella, A. P. Pereira, L. L. Coutinho, and doi:10.1186/1297-9686-24-4-305. L. C. Regitano. 2003. Genetic characterization of Aberdeen Angus Meuwissen, T. H. E., A. K. Sonesson, G. Gebregiwergis, and J. A. cattle using molecular markers. Genet. Mol. Biol. 26:1415–4757. Woolliams. 2020. Management of genetic diversity in the era of doi:10.1590/S1415-47572003000200005 genomics. Front. Genet. 11:880. doi:10.3389/fgene.2020.00880. Woolliams, J. 2007. Genetic contributions and inbreeding. In: K. Nei, M. 1973. Analysis of gene diversity in subdivided populations. Oldenbroek, editor. Utilization and conservation of farm animal Prod. Natl. Acad. Sci. 70:3321–3323. doi:10.1073/pnas.70.12.3321 genetic resources Wageningen Academic Publishers, Wageningen, Nei, M., T. Maruyama, and R. Chakraborty. 1975. The bottleneck the Netherlands; p. 147–165. effect and genetic variability in populations. Evolution 29:1–10. Wright, S. 1922. Coefficients of inbreeding and relationship. Am. Nat- doi:10.2307/2407137. uralist 56:330–338. doi:10.1086/279872. Nicholas, F.W. 1989. Incorporation of new reproductive technology Wright, S. 1931. Evolution in mendelian populations. Genetics 16:97– in genetic improvement programmes. In: W. G. Hill and T. F. C. 159. doi:10.1093/genetics/16.2.97. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Translational Animal Science Oxford University Press

Pedigree-based analyses of changes in genetic variability in three major swine breeds in Taiwan after a disease outbreak

Download PDF
9 pages

Loading next page...
 
/lp/oxford-university-press/pedigree-based-analyses-of-changes-in-genetic-variability-in-three-8fQ75XwjrC

References (66)

Copyright
© The Author(s) 2022. Published by Oxford University Press on behalf of the American Society of Animal Science.
eISSN
2573-2102
DOI
10.1093/tas/txac043
Publisher site
See Article on Publisher Site

Abstract

Translational Animal Science, 2022, 6, 1–9 https://doi.org/10.1093/tas/txac043 Advance access publication 13 April 2022 Animal Genetics and Genomics Pedigree-based analyses of changes in genetic variability in three major swine breeds in Taiwan after a disease outbreak † ‡ ‖ † †,1 Ruei-Syuan Wu, Hsu-Chang Wang, Chan Liang Su, Pei-Hwa Wang, and En-Chung Lin Department of Animal Science and Technology, National Taiwan University, 10617 Taipei, Taiwan National Animal Industry Foundation, 10070 Taipei, Taiwan Dairy Association of Taiwan, R. O. C., 10644 Taipei, Taiwan Corresponding author: eclin01@ntu.edu.tw ABSTRACT Pedigree analysis was performed in three major Taiwanese swine breeds to evaluate the genetic variability in the current population and determine the main reason for genetic diversity (GD) loss after the occurrence of foot-and-mouth disease (FMD) in Taiwan. The pedigree files of the Duroc, Landrace, and Yorkshire breeds, containing 60,237, 87,177, and 34,373 records, respectively, were analyzed. We divided the population into two subpopulations (pre-1998 and post-1998) to determine the role of FMD in GD loss. Pedigree completeness and related indicators were analyzed to evaluate the pedigree quality, and several parameters were used to measure the levels of GD and further used to determine the major cause of GD loss. The pedigree completeness indexes for the different breeds were higher than 0.60, and the trend was enhanced after the FMD outbreak. The estimated proportion of random genetic drift in GD loss increased in all breeds over time (from 62.64% to 78.44% in Duroc; from 26.26% to 57.99% in Landrace; and from 47.97% to 55.00% in Yorkshire, respectively). The effective population size of Duroc and Landrace were increased by the time (Duroc: from 61.73 to 84.75; Landrace: from 108.70 to 113.64); however, it shows opposite trend in Yorkshire population (decline from 86.21 to 50.00). In summary, the occurrence of FMD led to the major loss of GD loss by random genetic drift. Therefore, for the recovery of GD, breeders in Taiwan should increase the effective population size with newly imported genetic materials and adjust the breeding strategy to reduce the inbreeding rate. Key words: genetic diversity, pedigree, random genetic drift, swine INTRODUCTION of genetic resources for future improvement and application. For instance, other countries have built up registry databases Genetic variability is used to represent the different genetic to collect comprehensive information for evaluating the ge- characteristics within or between breeds. In livestock, breeders netic variability of their national purebred swine populations typically use within-breed genetic variability as an indicator (e.g., USA: National Swine Registry and Canada: Canadian for allowing sustained genetic improvement for economi- Centre for Swine Improvement). cally important traits to enhance performance (Reist-Marti A purebred swine registry system was also established et al., 2003), and it is helpful to make breeding objectives in Taiwan in 1975, beginning with a primary focus on re- to meet particular market (Notter, 1999). However, due to cording superior imported livestock and continuing to collect the modification of production systems and breeding strategy, records of their mating, farrowing, progeny, and performance genetic diversity (GD) has been lost in swine breeds (FAO, testing to construct a complete pedigree with performance in- 2000; Barker, 2001). The GD, or the expected heterozygosity, formation. This information is useful for breeders (and some is a basic criterion for evaluating the genetic variability (Nei, producers) to select or cull of their breeding stocks. At the 1973). The reduction of GD is associated with various events, same time, tracking the selection strategy based on pedigree such as inbreeding, which is usually associated with a negative and performance information has also strengthened the func- influence on fitness-related traits ( Fernández et al., 2005) and tionality of the registry system (Sung, 1993). The registration limit the response of selection (Falconer and Mackay, 1996). of purebred herds has become routine in Taiwan today, and In closed populations, inbreeding leads to a higher rate of loss until 2017, there had been more than 200,000 registered pigs of alleles (Woolliams, 2007) due to unequal founder contri- (Fig. 1). However, there still has not been an in-depth study of bution. In contrast, even without inbreeding, the frequency the within-breed genetic variability of the Taiwanese purebred of alleles in a certain generation may randomly increase, swine population. Moreover, in 1997, there was an outbreak decrease, or even disappear in the subsequent generation. of foot-and-mouth disease (FMD) in Taiwan, which resulted Genetic drift can eliminate alleles, and the disappearance of in the culling of around 2.73 million pigs (25.5% of the entire such genetic variation can only be restored by mutation or number of heads raised) that year. Noticeably, this cull in- migration (Lacy, 1989). Therefore, understanding the genetic cluded about 470,000 purebred animals (31.9% of the pure- variability within a breed is necessary for confirming the value bred herd) (COA, 1997). Although breeders have continued Received November 10, 2021 Accepted March 31, 2022. © The Author(s) 2022. Published by Oxford University Press on behalf of the American Society of Animal Science. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. 2 Wu et al. to import new genetics since 1998, the total number of pure- Characterization of the Pedigree Information bred animals in the last two decades are only one-fifth of that There are four imported breeds (Duroc, Hampshire, Landrace, before 1998. Rapid and persistent declines in population size and Yorkshire) that have been used by breeders in the are important causes of inbreeding and loss of genetic vari- crossbreeding systems in Taiwan. Duroc and Hampshire are ability (Nei et al., 1975). Therefore, this study analyzes the the primary paternal breeds, and Landrace and Yorkshire are variation in the GD of the swine populations before and after mainly used to produce the commercial parental sow (PS) 1998 to understand the impact of FMD occurrence on the for future commercial herd production. Due to Hampshire- GD of purebred stocks in Taiwan. This study uses pedigree crosses have a higher incidence of pale soft exudative (PSE) data to present the current GD of the three major pig breeds compared to Duroc-crosses (Purchas et al., 1990), which in in Taiwan and identify the reasons for the loss of GD. turn leads to major economic losses for breeders, breeders in Taiwan gradually select Duroc breed as terminal sire breed over the Hampshire breed. Therefore, the registered number MATERIALS AND METHODS of Hampshire swine is significantly fewer than the other three breeds historically (Fig. 2). Thus, in this study, the analyses Animal Care and Use Committee approval was not obtained of genetic variability will focus on the Duroc, Landrace, and for this study because the study did not utilize any live animals. Figure 1. Number of pigs born and farm per year. Figure 2. Number of pigs born within different swine breeds per year. Analysis of three swine breeds from Taiwan 3 Yorkshire breeds. Pedigree records were obtained from the Pedigree Completeness The pedigree completeness local database maintained by the National Animal Industry levels of the two reference populations were evaluated by cal- Foundation and included all pedigree records available for culating (1) the number of fully traced generations, (2) the each breed up to 2017. To understand the influence of FMD maximum number of generations traced, and (3) the equiv- disease on the genetic variation of purebred pigs, the entire alent number of complete generations for each animal in the population was divided into two parts: the subpopulation pre- pedigree data. The first is defined as the number of genera - 1998, which includes pigs born between 1971 and 1997, and tions that separates the offspring from the furthest generation the subpopulation post-1998, which contains pigs born from where their ancestors of the individual are known. The second 1998 to 2017. For comparison of GD, some population with is the number of generations between an individual and its GD management was often selected as a reference point. This is furthest ancestor. Finally, the equivalent complete generation called the reference population, which can be some population is computed as the sum over all known ancestors of the terms in the past or the current population (Meuwissen et al., 2020). computed as the sum of (1/2) , where n is the number of gen- Previous literature has pointed out that average generation in- erations separating the individual from each known ancestor terval (GI) is used for reference population selection, which due (Maignel et al., 1996). Additionally, ENDOG offers a fourth to this reference population would comprises the last gener- indicator (4), the pedigree completeness index (PCI), which ation of data evaluated in each breed (Melka and Schenkel, describes the completeness of each ancestor in the pedigree of 2010). Similarly, present study also used average GI as the the parental generation (MacCluer et al., 1983). basis for the selection of the reference population. Since the Generation Interval The GI was computed for the four average GI of each breed in Taiwan is about 2.5 years, piglets possible selection pathways (sire–son, sire–daughter, dam– born in 1995–1997 and 2015–2017 were defined as reference son, and dam–daughter) as the average age of the parents at population for the pre-1998 and post-1998 subpopulation, re- the birth of the offspring used to replace them. spectively. The completeness of the pedigrees and the genetic variability were found by using pedigree information to calcu- The Parameters for Genetic Variability Evaluation late the associated parameters, such as pedigree completeness Inbreeding and effective population size The in- index, inbreeding coefficient and the parameters derived from dividual inbreeding coefficient (F), which is defined as the probability of gene origin, etc. The number of animals analyzed probability that an individual has two identical genes by de- in the entire pedigree and the reference population of each scent (Wright, 1922), was calculated according to the algo- breed is given in Table 1. Pedigree files collected for the Duroc, rithm described by Meuwissen and Luo (1992). The increase Landrace, and Yorkshire breeds consisted of 60,237, 87,177, in inbreeding (∆F) was calculated for each generation using and 34,373 records, respectively, for subsequent analyses. the classical formula, ∆ F = F F /1 − F , where F and F t— t-1 t-1 t t-1 are the average inbreeding at the tth generation. The effective Pedigree Quality Determination and Genetic population size (Ne) represents the relationships between the Variability Evaluation number of males and females that contribute genetically to the The ENDOG program (version 4.8) was used to identify the population. Due to the increase of inbreeding, this parameter pedigree quality and genetic variability of these three pure- was estimated in this study (Wright, 1931) based on ∆ F, as bred breeds (Gutiérrez and Goyache, 2005). Ne = 1/(2△F), for each generation having F > F , to roughly t t-1 Table 1. Basic description of pedigree datasets, and the parameters of pedigree completeness and inbreeding analysis for all breeds. Breed Duroc Landrace Yorkshire subpopulation Pre-1998 Post-1998 Pre-1998 Post-1998 Pre-1998 Post-1998 No. of animals in whole population 32,858 27,379 51,263 35,914 22,964 11,409 No. of animals in reference population 4,534 2,871 8,766 3,402 3,688 1,083 No. of Inbred animals 3,367 2,623 3,657 2,278 1,531 666 Inbred animals, % 74.26 91.36 41.71 66.96 41.51 61.49 Pedigree Completeness Index, % 0.83 0.86 0.64 0.73 0.61 0.69 Maximum generations traced 17 17 15 16 16 15 Mean equivalent generations 3.50 3.46 2.43 2.63 2.36 2.23 Complete generations 7 7 5 5 5 5 % known ancestors in: 1st generation 92.9 96.7 92.0 91.3 90.7 88.1 2nd generation 85.6 91.0 72.5 82.2 70.7 77.4 4th generation 73.5 74.8 43.6 54.8 34.4 49.3 Mean F, % 4.24 3.62 1.37 1.76 1.86 2.77 ∆ F, % 0.81 0.59 0.45 0.43 0.58 1.00 N 61.73 84.75 108.70 113.64 86.21 50.00 Inbred animals, the animals with F > 0; Mean F, average inbreeding coefficient of all animals in reference population. ∆ F, the increase in inbreeding of all animals in reference population; N , effective population size. e 4 Wu et al. characterize the effects of remote and close inbreeding. Ne 1—GD* represents the loss of GD due to unequal is defined as the number of breeding animals that would contributions by the founders to the population (Caballero increase inbreeding if they contributed equally to the next and Toro, 2000). Therefore, the difference between GD* and generation (Gutiérrez and Goyache, 2005). GD estimates the loss of diversity by genetic drift (Caballero and Toro, 2000; Honda et al., 2004). The probability of gene origin. Total number of founders (f) was defined as ancestors with unknown parents. RESULTS The effective number of founders (f ) means the number of The number of purebred pigs began to grow significantly in equally contributing founders that would be expected to gen- 1980 and remained stable before the outbreak of FMD in erate the same amount of GD as in the studied population Taiwan in 1997 (Fig. 1). However, in addition to the FMD (Lacy, 1989), was calculated using the following formula: outbreak in 1997, entry to the World Trade Organization   (WTO) in 2002 and a worldwide rise in grain prices in 2008 −1 caused a rapid decrease in the total number of pigs and   f = q breeders. Compared with the historical peak, the number of i=1 pigs born and the number of breeding farms have decreased by 66% and 76%, respectively, in the last 3 years. The inter- where q is the genetic contribution of the ith founder to the national situation and the influence of breeding conditions reference population and f is the total number of founders. The have caused Taiwan’s pig breeding industry to become con- f is usually lower than f, indicating the unequal contributions servative and gradually formed a large-scale, centralized, and of founders due to selection, which is the major mating op- corporate feeding operation. eration in the purebred industry. However, f alone may not be an useful indicator for assessing GD because the genetic Pedigree Completeness The PCIs of the reference contributions of the founders would converge after some populations of different breeds are shown in Table 1. Over number of generations (Bijma and Woolliams, 1999), and time, the PCI of the three breeds has increased. Comparison hence, the f would remain constant in the resulting popula- between the breeds indicated that the PCI of Duroc is the tion over time. The founder genome equivalent was defined as highest, while the completeness of the Yorkshire pedigree is the number of equally contributing founders with no random relatively low. Similarly, the maximum generations traced and loss of founder alleles that would be expected to give the same mean equivalent generations of the Duroc breed are also higher level of GD observed in the population under study (Lacy, than the other two breeds. The analysis of known ancestors in 1989), and it was computed as f = 1/(2 × f ), where f is the ge g g different generations shows that in the first generation, except average co-ancestry for the population considered (Caballero for the Yorkshire post-1998 subpopulation, more than 90% and Toro, 2000). The amount of GD in the reference popula- of the ancestors were known in other populations. But by the tion was calculated as (Lacy, 1995): fourth generation, only Duroc is higher than 70%, Landrace has fallen below 70%, and the value is less than 50% for the Yorkshire breed. In summary, the completeness of the GD = 1 − 2fge Duroc pedigree is better than that of the other two breeds. The requirements for pedigree integrity after the occurrence The loss of GD can be due to both genetic drift and unequal of FMD are also significantly higher than they were before the founder contribution; the value expressed as 1 − GD indicates occurrence of FMD. the amount of GD lost in the population since the founder generation due to genetic drift and the unequal founder con- Generation Interval Computed average GI and divided tribution. The amount of GD in the reference population into four selection paths in all breeds are shown in Table 2. accounting for the loss of diversity due to unequal founder As mentioned above, the average GI of all the breeds is over contribution (GD*) was calculated as (Lacy, 1995): 2 years, so the breeding stocks born in the most recent 3 years of each subpopulation are set as the reference popu- lation. The results indicate that the GI of the population be- GD = 1 − 2f fore the occurrence of FMD is shorter than that after the occurrence of FMD in all breeds. This might be due to the Table 2. Average generation intervals in evaluated breeds. Breed Duroc Landrace Yorkshire subpopulation Pre-1998 Post-1998 Pre-1998 Post-1998 Pre-1998 Post-1998 Average generation interval 2.17 2.43 2.30 2.63 2.28 2.60 GI for sire–son path 2.20 2.66 2.20 2.96 2.21 2.78 GI for sire–daughter path 2.19 2.43 2.17 2.58 2.19 2.56 GI for dam–son path 2.13 2.33 2.31 2.62 2.34 2.52 GI for dam–daughter path 2.16 2.41 2.43 2.64 2.38 2.61 GI, generation interval. Analysis of three swine breeds from Taiwan 5 reduction of the scale of production, which reduced the de- post-1998 subpopulation had the lowest N (50 animals), mand for breeding stocks, and lead to the breeding age of while the Landrace had the highest N in both subpopulations. breeding stocks is older. The lowest value of average GI was observed for Duroc breed (2.17 and 2.43 years) in pre-1998 The probability of gene origin. Table 3 summarizes all and post-1998 subpopulation, respectively. In contrast, the of the calculated parameters for the gene origin analysis of the longest GI are all obtained Landrace breed (2.30 and 2.63 reference population. Among the pre-1998 subpopulation, years) in different period. When comparing the different se- the highest number of founders were detected (f) for Landrace lection paths within each breed in pre-1998 subpopulation, (2,921), followed by Yorkshire (1,494), and Duroc (1,378). we can observe that the longest GI was the sire to son path However, in the post-1998 subpopulation, the Landrace and the shortest GI was found for the dam to son path in breed still had the highest number of founders, but the Duroc. On the other hand, there are the longest GI for the number for the Yorkshire breed was reduced to fewer than dam to daughter path and the shortest path for the sire to the Duroc breed. A similar trend also appeared for the effec- daughter path of the two maternal line breed. However, the tive number of founders (f ) and the equivalent number of longest path was all detected for the sire to son path in each founders’ genomes (f ). ge breed in post-1998 subpopulation. The highest difference The highest values of f and f were reached in the e ge among selection paths within one breed was observed in subpopulation of the Landrace breed before 1998 (457 and Landrace in different subpopulation (0.26 and 0.38, respec- 337) and the subpopulation after 1998 (169 and 71). The tively). The most balanced GI was detected for Duroc and lowest effective number of founders and founder genome Yorkshire (0.07 in pre-1998 subpopulation and 0.26 in post- equivalents were found in the Duroc and Yorkshire breeds 1998 subpopulation) when comparing the individual paths (174 and 65 founders; 80 and 36 founders, respectively) in of selection. different subpopulations. After the occurrence of FMD, the f, f , and f of all breeds decreased. In comparing different e ge The Parameters for Genetic Variability Evaluation subpopulations, the ratio of f /f shows a similar trend in Inbreeding and effective population size The sum- the Landrace and Yorkshire breeds; however, it increased in mary statistics for each breed are listed in Table 1. The results Duroc after the FMD outbreak. The highest value occurred showed that the proportion of inbred animals of all breeds in the Yorkshire breed (0.25) and the lowest value in Duroc has increased over time. In the Duroc breed, the proportion (0.13) in the pre-1998 subpopulation. In the post-1998 of inbred animals is significantly higher than in the other two subpopulation, Duroc and Yorkshire breeds had equal values breeds. However, although the proportion of inbreeding in the (0.16), while the Landrace had the lowest value (0.12). The analyzed breeds is high the value of the mean inbreeding co- ratio of f /f decreased in all the breeds in this study over ge e efficient (Mean F) is moderate, ranging from 4.24% (Duroc) time. The lowest values calculated were for the Duroc breed to 1.37% (Landrace) in the pre-1998 subpopulation and in different subpopulations (0.37 in pre-subpopulation and from 3.62% (Duroc) to 1.76% (Landrace) in the post-1998 0.22 in post-subpopulation, respectively). The accumulated subpopulation. With the occurrence of FMD, the average marginal contribution of 100 major ancestors is illustrated in increase of inbreeding (∆F) of each breed presented a different Fig. 3a–c for the different breeds. Independent of breed, more trend. The Duroc breed was reduced, and the Landrace breed ancestors are required to explain the gene origin in the pre- remained almost the same. However, ∆ F showed a dramatic 1998 subpopulation than in the post-1998 subpopulation. increase in the Yorkshire breed (from 0.58% to 1%). This One hundred ancestors can only explain about 50% of result can also be seen in the effective population size (N ). the gene pool in pre-1998 subpopulations of the Landrace Among the subpopulations pre- and post-1998, the Yorkshire and Yorkshire breeds. However, only half this amount is Table 3. Parameters derived from the probability of gene origin in the different reference populations in each breed. Breed Duroc Landrace Yorkshire subpopulation Pre-1998 Post-1998 Pre-1998 Post-1998 Pre-1998 Post-1998 Total number of founders, f 1378 1049 2921 1393 1494 515 Effective number of founders, f 174 167 457 169 369 80 Founder genome equivalent, f 65 36 337 71 192 36 ge f /f ratio 0.13 0.16 0.16 0.12 0.25 0.16 f /f ratio 0.37 0.22 0.74 0.42 0.52 0.45 ge e Number of ancestors to explain: 50% of gene pool 56 35 150 42 105 22 75% of gene pool 169 103 415 133 295 74 100% of gene pool 1143 680 2598 1042 1255 350 GD 0.992 0.986 0.999 0.993 0.997 0.986 1 − GD (GD loss) 0.008 0.014 0.001 0.007 0.003 0.014 Proportion of unequal contributions of the founders in GD loss 37.36 21.56 73.74 42.01 52.03 45.00 Proportion of random genetic drift in GD loss 62.64 78.44 26.26 57.99 47.97 55.00 GD, genetic diversity; The equation of GD calculation is GD = 1 − . 2f ge 6 Wu et al. Figure 3. Accumulated marginal contribution of 100 major ancestors in the Taiwan swine breeds. (a) Duroc breed; (b) Landrace breed; (c) Yorkshire breed. needed for this purpose in the Duroc breed. In the post-1998 some recent pedigree analyses of pigs. The completeness of subpopulation, fewer than 50 ancestors are enough to explain our analyzed Duroc pedigrees was similar to that published 50% of the gene pool in all evaluated breeds. Comparing the by Melka and Schenkel in four Canadian swine breeds and breeds indicated that Landrace contains the largest amount of more complete than the results reported by Tang et al. in three ancestors to explain its gene pool. imported swine breeds in China (Melka and Schenkel, 2010; Tang et al., 2013) However, the Landrace and Yorkshire PCIs were worse than the previous studies. These differences in DISCUSSION the quality of pedigree parameters between breeds found in From 1980 onward, the number of pigs and breeding farms our study might be caused by the dynamics and intensity of increased because Taiwan began to export purebred herds individual breeds used in commercial breeding programs. In to Southeast Asian countries, such as the Philippines and addition, some differences may also be based on the impact Thailand, and was a net exporter of prime pork meat to of imports or the depth of pedigree knowledge of imported Japan, with the offal left for domestic consumption (Sung, animals. These parameters ultimately affect the quality of the 1993). However, the outbreak of FMD led to consequences pedigree and thus the estimated inbreeding and GD. such as the inability to export pigs and pork, the development In general, the generation intervals of males are slightly of Taiwan’s purebred breeding pig industry stalled, and the shorter than that of females (Melka and Schenkel, 2010). population became closed. Therefore, this study uses pedigree Similar results can be found in previous reports (Melka and information and different parameters (such as F, N , f and Schenkel, 2010; Tang et al., 2013; Krupa et al., 2015). In this e e f ) to evaluate the GD of the population pre- and post-FMD study, the pre-1998 subpopulation also showed the same ge to understand the impact on the GD of the sharp decline in trend, except for the Duroc breed. However, in the post-1998 the number of purebred stocks and subsequent population subpopulation, each breed showed the opposite result, the closure. The accuracy of parameters such as inbreeding co- generation interval of males is longer than females. The gener- efficient, effective population size, and GD depends on the ation interval of both males and females increases over time. quality of the population pedigree. This study shows that This may be because after the occurrence of FMD, Taiwan’s the PCI of each breed can reach more than 60%, and the breeding pig industry shrank and could not be exported, de- percentage of known ancestors in the fourth generation mand for breeding stock has declined, so the mating strategy was higher than 70% in the Taiwan Duroc breed indicating has shifted to the purebred population was centralized derived a reasonable estimation of other pedigree parameters. from certain animals, which leading to older breeding age of Moreover, the correlated parameters of pedigree integrity males and females in the herd. This is because breeders have of the subpopulation in the breeds in this study before and no confidence in using new breeding animals, therefore would after the occurrence of FMD are very similar, so subsequent rather the current breeding stocks keep in production than comparisons can be made. Furthermore, the pedigree quality allow new generation to join or even replace them. This is of the Taiwanese pig population is moderate compared to evident in the individual selection paths, especially in the sire Analysis of three swine breeds from Taiwan 7 to son path of each breed. This situation might be improved the Yorkshire breeders have recently switched to the strategy through on-farm testing, which has gradually matured in of constructing the population by import instead of breeding recent years in Taiwan. By performing performance testing, native breed (the percentage of unknown parents in the regis- calculating estimated breeding values and establishing the se- tered Yorkshire population is equal to 13%) and centralizing lection index for various traits of purebred progeny, breeders by using this imported livestock to multiply the nucleus can identify the best animals and mating them in younger age. herd. The newly introduced breeding stocks are often mated It will helpful to shorten the generation distance and reduce with their progeny, causing their ∆ F rate to increase much breeding costs. faster than the other two breeds so that the usage of the The observed values of the mean inbreeding coefficient Yorkshire breed in Taiwan is very limited. However, a pre- over the breeds evaluated in this study were similar to the vious study demonstrated that the reproductive performance previous report. Krupa et al. (2015) have indicated a higher of Yorkshire × Landrace (two-way crossbred) sows was better inbreeding coefficient for the Czech Duroc breed (approxi - than that of purebred Landrace sows when both were mated mately 3.6%), while for the Czech Landrace and Czech with Duroc sires (Huang Y-H et al., 2002). This illustrated the Large White breeds, it has not exceeded 3%. Furthermore, advantage to incorporate the Yorkshire breed into the pro- the inbreeding degree of Landrace in Taiwan was lower than duction system. the Czech Landrace. The rate of inbreeding (∆ F) is an effec- The assessment of GD is especially important in highly tive criterion for measuring population health, and Nicholas specialized livestock breeds because assisted reproduc- has recommended a ∆ F rate < 0.5%, whereas the Food and tion techniques, such as artificial insemination and embryo Agriculture Organization of the United Nations (FAO) transfer technologies, can potentially rapidly reduce the GD has suggested a ∆ F rate < 1% as a goal (Nicholas, 1989; of a population (Vasconcellos et al., 2003). Parameters de- FAO, 2000). All breeds meet the FAOs goal in the different rived from the probability of gene origin analysis indicated subpopulations, but only the Landrace breed conforms to the an increased trend of GD loss with time in all three breeds in rate indicated by Nicholas. The FAO suggested that N for a this study. A relatively small number of major ancestors are breed should be maintained above 50 in order to withstand needed to explain the gene origin appears in the post-1998 the effects of inbreeding (FAO, 2000), while a size of 500 is subpopulation, which shows that the occurrence of FMD led essential to sustain the genetic variability and evolutionary to the decrease of GD. In general, the number of ancestors potential of the population for several generations (Frankham explaining the entire gene pool is still much greater than the et al., 2002). These authors defined at least 500 animals of results presented in the previous literature, showing that the Ne are needed to maintain GD in the population for several present Taiwanese purebred stocks maintain a fairly good GD generations. From the perspective of current and past effec- (Melka and Schenkel., 2010; Tang et al., 2013; Krupa et al., tive population sizes, efforts should be made for all the breeds 2015). analyzed in this study to enhance the N in order to construct To further explore the causes of GD loss, the f /f ratio e e a more varied population, although they are presently sim- represents the reduction of GD due to the unequal contribu- ilar to the results of previously published literature, and all tion of founders. The f /f ratio can be used to quantify only ge e of them meet the minimum requirement for N (50 animals) the influence of genetic drift on the amount of GD. In general, (Melka and Schenkel., 2010; Tang et al., 2013; Krupa et al., if all the founders were to contribute equally, the total number 2015). As the biotechnology has advanced, it allow using ge- of founders would be the same as f . However, the f is usu- e e nome information for the assessment and management of ally lower than f, indicating there are unequal contributions GD. Genomics data used for many alternative measures of of founders due to selection, namely that the breeders have inbreeding and genomic relationships. These measures are ap- preferentially chosen certain animals as parents (Melka and plied for managing GD in genome best contribution (GOC) Schenkel, 2010). The lower these two ratios are, the strong selection schemes (Meuwissen et al., 2020). By using genomic the effect of unequal contributions of founders or random information to measure different inbreeding coefficients and genetic drift is. The results of the probability of gene origin genomic relationships, such as drift or homozygosity-based in the reference populations show the f /f ratio was similar in inbreeding coefficients, it is possible to understand the true these three breeds at different periods, which means they may situation of allelic diversity (AD), and then choose the best be under the same degree of selection intensity in Taiwan. GOC management plan. The similar results can be observed Over the past years, knowledge of the production system has in previous study that different inbreeding coefficient value allowed the adjustment of the selection intensity for growth obtained by using genomic information and pedigree data and carcass traits (such as average daily gain and back fat) (Grossi et al., 2017). These results all prove that the measure- and reproductive traits (such as the longevity and litter size) ment derived only from the pedigree are not comprehensive in these three breeds. However, the Duroc breed presents the enough, and many details in it need to be more clearly under- lowest f /f ratio in both subpopulations, indicating the sig- ge e stood and applied through the use of genomic information. nificant effect of random genetic drift. Compared to previous Therefore, genomic-based analyses should be performed in studies of Canadian and Czech populations, the influence of our future studies to elucidate the real causes of GD loss and random genetic drift in Taiwan is milder (Melka and Schenkel, to identify strategies for GD management. The Ne of Duroc 2010; Krupa et al., 2015). In this study, the impact of random and Landrace increased with time, but Yorkshire decreased genetic drift was substantial for all breeds. The highest value with time, even more so in the post-1998 subpopulation. The of overall GD lost was observed for the Duroc and Yorkshire N of Yorkshire breed has been reduced to a critical value breeds in the post-1998 subpopulation. This situation appears (50 animals), which means that the ∆ F rate of the current to have occurred because the previous results indicated in the population has reached 1%. Though the Duroc breed still Czech population were similarly due to the number of an- has the highest mean inbreeding coefficient, the Yorkshire imals, along with the proportion of imported animals, was breed has shortened the distance. According to these results, reduced during the period of analysis (Krupa et al., 2015). 8 Wu et al. Before the occurrence of FMD, Duroc GD loss was mainly should pay more attention to the adjustment of breeding due to random genetic drift, while Landrace and Yorkshire operation methods for Yorkshire breeds, with the goal of GD losses were due to the unequal contributions of founders. develop a Taiwanese Yorkshire population that is adapted However, after the occurrence of FMD, the main cause of GD to Taiwan’s environment, cautiously use foreign breeding loss of the three breeds was random genetic drift. Although stocks in the production system. Maximize the effect of for- Duroc still has the highest degree of influence of random ge - eign breeding stock, which thereby reducing the rate of ∆ F netic drift (78.44%), the influence of random genetic drift in and increasing GD in population. Landrace has been significantly increased (from 26.26% to 57.99%), which indicates that the dramatic decline of this Acknowledgments population has had a greater impact than other two breeds on its GD. The main reason for such a high proportion of This work was financially supported by the Council of GD loss caused by random genetic drift (and other reasons Agriculture, Executive Yuan, Taiwan (Grant No.: 99 AS- such as the bottleneck effect) is the population shrinkage and 04.07-AD-04) in Taiwan. the declining proportion of introduced animals in Taiwan after the outbreak of FMD. Our study compared the dif- ferent subpopulations, which contrasted the historical and Conflict of interest statement current status of genetic variability to clarify the impact We certify that there is non conflict of interest involving any of FMD. Previous studies had indicated that GD loss due financial organization regarding the material discussed in the to random genetic drift led to an increase in homozygosity manuscript. and fixation of alleles ( Fernández et al., 2005; Toro et al., 2006). The loss of diversity can be caused by random genetic drift or unequal contributions of the founders. This study LITERATURE CITED confirmed that the selection of breeding stocks for specific Barker, J. S. F. 2001. Conservation and management of genetic diver- traits by breeders has a certain impact on the loss of popu- sity: a domestic animal perspective. Can. J. For. Res. 31:588–595. lation GD in Taiwan. However, after the outbreak of FMD doi:10.1139/x00-180. in Taiwan, the population size suddenly decline and the pro- Bijma, P., and J. A. Woolliams. 1999. Prediction of genetic contributions portion of imported animals decreased, resulting in purebred and generation intervals in populations with overlapping genera- populations mainly derived from breeding stocks with similar tions under selection. Genetics 151:1197–1210. doi:10.1093/ge- genetic backgrounds, which lead to an increase in the degree netics/151.3.1197. of random genetic drift. Therefore, to solve such problems, Caballero A. and Toro M. A. 2000. Interrelations between effec- breeding strategies should be adjusted, such as introducing tive population size and other pedigree tools for the manage- new genetic material and reducing inbreeding (Haile-Mariam ment of conserved populations. Genet. Res. Camb. 75:331–343. et al., 2007) to reduce the effects of drift (Lacy, 1987) to doi:10.1017/S0016672399004449 Council of Agriculture. 1997. Yearly Report of Taiwan's Agriculture further improve the level of GD. After long-term efforts, (1997). https://www.coa.gov.tw/ws.php?id=4563 (Traditional Chi- Taiwan finally was removed from the FMD epidemic area in nese version). June 2020, which provided another opportunities for pork Falconer, D.S. and T. F. C. Mackay 1996. Introduction to quantitative exports. In the future, Taiwan breeders should refer to the genetics. 4th ed. Longman Group Ltd., Essex, UK. results of previous and this study to formulate introduction FAO. 2000. Secondary guidelines for development of national farm an- and selection/breeding strategies to effectively reduce the im- imal genetic resources management plans: Management of small pact of random genetic drift, and increase the GD of the entire populations at risk. UN Food and Agric. Org. Rome, Italy. purebred stocks, for providing a stable supply basis in the Fernández, J., B. Villanueva, R. Pong-Wong, and M. A. Toro. 2005. Ef- export market. ficiency of the use of pedigree and molecular marker information in conservation programs. Genetics 170:1313–1321. doi:10.1534/ genetics.104.037325. CONCLUSION Frankham, R., D. A. Briscoe, and J. D. Ballou. 2002. Introduction to conservation genetics. University Press, Cambridge, UK. In summary, pedigree quality of Duroc breed is better than the Grossi, D. A., M. Jafarikia, L. F. Brito, M. E. Buzanskas, M. Sargolzaei, other breeds which might be related to that Duroc is used as and F. S. Schenkel. 2017. Genetic diversity, extent of linkage dis- sire and the influence exceed the other two maternal breeds. equilibrium and persistence of gametic phase in Canadian pigs. Therefore, for the Landrace and Yorkshire breeds, breeders BMC Genet. 18:6. doi:10.1186/s12863-017-0473-y. in Taiwan need to record their pedigree more carefully to Gutiérrez, J. P., and F. Goyache. 2005. A note on ENDOG: a computer obtain more accurate evaluation results. The results of GI program for analysing pedigree information. J. Anim. Breed. Genet. analysis indicated that Taiwanese breeders should adjust 122:172–176. doi:10.1111/j.1439-0388.2005.00512.x. their management that mating the new generation in younger Haile-Mariam, M., P. J. Bowman, and M. E. Goddard. 2007. A prac- tical approach for minimizing inbreeding and maximizing genetic age according to the result of progeny test to shorten the diversity in dairy cattle. Genet. Sel. Evol. 39:269–289. doi:10.1051/ generation interval. Genetic variability measurements show gse:2007009. that after the FMD outbreak, random genetic drift played Honda, T., T. Nomura, Y. Yamaguchi, and F. Mukai. 2004. Monitoring an important role in the loss of GD in Taiwanese swine of genetic diversity in the Japanese Black cattle population by the breeds. Breeders should reduce the inbreeding rate, intro- use of pedigree information. J. Anim. Breed. Genet. 121:242–252. duce genetically unrelated individuals and formulate better doi:10.1111/j.1439-0388.2004.00452.x. mating strategy (such as manage the breeding stock mating Huang, Y. -H., S. W. Roan, and Y. -P. Lee. 2002. Reproductive perfor- in younger breeding age) to increase the effective population mance of Landrace and that of Landrace × Yorkshire sows bred by size. The ∆ F rate of the Yorkshire breeds has increased sig- Duroc sires. J. Chin. Soc. Anim. Sci. 31:19–29. doi:10.1016/S0093- nificantly. Therefore, in addition to introduction, breeders 691X(00)00364-2. Analysis of three swine breeds from Taiwan 9 Krupa, E., E. Zakova, and Z. Krupová. 2015. Evaluation of inbreeding Mackay, editors. Evolution and animal breeding. CAB Int., Wal- and genetic variability of five pig breeds in Czech Republic. Asian- lingford, UK. p. 201–209. Australas. J. Anim. Sci. 28:25–36. doi:10.5713/ajas.14.0251. Notter, D. R. 1999. The importance of genetic variability in live- Lacy, R. C. 1987. Loss of genetic diversity from managed populations. stock populations of the future. J. Anim. Sci. 77:61–69. Interacting effects of drift, mutation, immigration, selection and pop- doi:10.2527/1999.77161x. ulation subdivision. Conserv. Biol. 1:143–158. doi:10.1111/j.1523- Purchas, R. W., W. C. Smith, and G. Pearson. 1990. A comparison of 1739.1987.tb00023.x. the Duroc, Hampshire, Landrace, and Large White as terminal sire Lacy, R. C. 1989. Analysis of founder representation in pedigrees: breeds of crossbred pigs slaughtered at 85 kg liveweight. N. Z. J. founder equivalents and founder genome equivalents. Zoo Biol. Agric. Res. 33:97–104. doi:10.1080/00288233.1990.10430666. 8:111–123. doi:10.1002/zoo.1430080203. Reist-Marti, S. B., H. Simianer, J. Gibson, O. Hanotte, and J. E. O. Rege. Lacy, R. C. 1995. Clarification of genetic terms and their use in the 2003. Weitzman’s approach and conservation of breed diversity: management of captive populations. Zoo Biol. 14:565–578. an application to African cattle breeds. Cons. Biol. 17:1299–1311. doi:10.1002/zoo.1430140609. Sung, Y. Y. 1993. Pure breed registration of swine in Taiwan. In: “Pork MacCluer, J. W., A. J. Boyce, and B. Dyke. 1983. Inbreeding and ped- Industry Conference on Breeding Strategies”, Taiwan Livestock igree structure in Standardbred horse. Int. J. Hered. 74:394–399. Research Institute; p. F5–Fl2. doi:10.1093/oxfordjournals.jhered.a109824. Tang, G. Q., J. Xue, M. J. Lian, R. F. Yang, T. F. Liu, and Z. Y. Zeng. Maignel, L., D. Boichard, and E. Verrier. 1996. Genetic variability of 2013. Inbreeding and genetic variability in three imported swine French dairy breeds estimated from pedigree information. Interbull breeds in China using pedigree data. Asian Australas. J. Anim. Sci. Bulletin 14:4–54. 26:755–765. doi:10.5713/ajas.2012.12645. Melka, M. G., and F. Schenkel. 2010. Analysis of genetic variability in Toro, M. A., J. Fernández and A. Caballero. 2006. Scientific basis for four Canadian swine breeds using pedigree data. Can. J. Anim. Sci. policies in conservation of farm animal genetic resources. Proceed- 90:331–340. doi:10.4141/CJAS10002. ings of the 8th World Congress on Genetics Applied to Livestock Meuwissen, T. H. E., and Z. Luo. 1992. Computing inbreeding Production. Communication no. 33-05. Belo Horizonte, Brazil. coefficients in large populations. Genet. Sel. Evol. 24:305–313. Vasconcellos, L. P. M. K., T. Daniella, A. P. Pereira, L. L. Coutinho, and doi:10.1186/1297-9686-24-4-305. L. C. Regitano. 2003. Genetic characterization of Aberdeen Angus Meuwissen, T. H. E., A. K. Sonesson, G. Gebregiwergis, and J. A. cattle using molecular markers. Genet. Mol. Biol. 26:1415–4757. Woolliams. 2020. Management of genetic diversity in the era of doi:10.1590/S1415-47572003000200005 genomics. Front. Genet. 11:880. doi:10.3389/fgene.2020.00880. Woolliams, J. 2007. Genetic contributions and inbreeding. In: K. Nei, M. 1973. Analysis of gene diversity in subdivided populations. Oldenbroek, editor. Utilization and conservation of farm animal Prod. Natl. Acad. Sci. 70:3321–3323. doi:10.1073/pnas.70.12.3321 genetic resources Wageningen Academic Publishers, Wageningen, Nei, M., T. Maruyama, and R. Chakraborty. 1975. The bottleneck the Netherlands; p. 147–165. effect and genetic variability in populations. Evolution 29:1–10. Wright, S. 1922. Coefficients of inbreeding and relationship. Am. Nat- doi:10.2307/2407137. uralist 56:330–338. doi:10.1086/279872. Nicholas, F.W. 1989. Incorporation of new reproductive technology Wright, S. 1931. Evolution in mendelian populations. Genetics 16:97– in genetic improvement programmes. In: W. G. Hill and T. F. C. 159. doi:10.1093/genetics/16.2.97.

Journal

Translational Animal Science – Oxford University Press

Published: Apr 13, 2022

Keywords: genetic diversity; pedigree; random genetic drift; swine

There are no references for this article.