Evaluation of the genetic diversity and population structure of five indigenous and one introduced Chinese goose breeds using microsatellite markers

Transcription

1 Evaluation of the genetic diversity and population structure of five indigenous and one introduced Chinese goose breeds using microsatellite markers Jinjun Li, Qingyuan Yuan, Junda Shen, Zhengrong Tao, Guoqing Li, Yong Tian, Deqian Wang, Li Chen, and Lizhi Lu 1 Institute of Animal Husbandry and Veterinary Medicine, Zhejiang Academy of Agricultural Sciences, Hangzhou , China. Received 28 March 2011, accepted 14 August Li, J., Yuan, Q., Shen, J., Tao, Z., Li, G., Tian, Y., Wang, D., Chen, L. and Lu, L Evaluation of the genetic diversity and population structure of five indigenous and one introduced Chinese goose breeds using microsatellite markers. Can. J. Anim. Sci. 92: The aim of this study was to determine the genetic diversity and evolutionary relationships among five indigenous Chinese goose breeds and one introduced goose breed using 29 microsatellite markers. A total of 334 distinct alleles were observed across the six breeds, and 45 of the 334 alleles (13.5%) were unique to only one breed. The indigenous geese showed higher diversity in terms of the observed number of alleles per locus ( ) and observed heterozygosity ( ) compared with the introduced breed (3.97 and 0.29, respectively). The pairwise genetic differentiation (F ST ) between the six goose breeds ranged from 0.04 between Panshi Grey goose (PS) and Yongkang Grey goose to 0.47 between PS and Landes goose; similarly, Nei s genetic distance varied between 0.25 and However, the F ST between the indigenous Chinese goose breeds was very small. In addition, genetic distance estimate, phylogenic, and cluster analyses of the genetic relationships and population structure revealed that some indigenous goose breeds had hybridized more frequently, resulting in a loss of genetic distinctiveness. Key words: Goose, genetic diversity, population structure, microsatellite marker Li, J., Yuan, Q., Shen, J., Tao, Z., Li, G., Tian, Y., Wang, D., Chen, L. et Lu, L Évaluation de la diversite ge nétique et de la distribution de la population de cinq races indige` nes et d une race exotique d oie chinoise au moyen de marqueurs microsatellitaires. Can. J. Anim. Sci. 92: L e tude devait pre ciser la diversite ge nétique et l e volution de cinq races indige` nes d oie chinoise et d une race exotique graˆce a` 29 marqueurs microsatellitaires. En tout, 334 alle` les distincts ont été observe s chez les six races, 45 d entre eux (13,5 %) étant uniques à l une ou l autre race. Les oies indige` nes se caracte risent par une plus grande diversite quant au nombre d allèles par locus (de 4,48 a` 5,90) et à l hétérozygotie (de 0,46 a` 0,53), comparativement à la race exotique (3,97 et 0,29, respectivement). La diffe renciation ge ne tique par paire (F ST ) des six races varie de 0,04 entre l oie grise Panshi (PS) et l oie grise Yongkang à 0,47 entre la PS et l oie Landes; de meˆme, l e cart ge ne tique de Nei varie de 0,25 a` 0,75. Toutefois, la F ST demeure minime entre les races indige` nes. De plus, l estimation de la distance ge ne tique, la phyloge nie et l analyse par grappes des liens ge ne tiques et de la distribution de la population re vèlent que certaines races indige` nes se sont hybride es plus souvent, ce qui a atte nue leur diffe renciation ge ne tique. Mots clés: Oie, diversite ge nétique, distribution de la population, marqueurs microsatellitaires China has a long history of goose (Anser cygnoides) breeding; various breeds have been developed by natural and artificial selection (Chen et al. 2004). In eastern China, there are five indigenous goose breeds that have evolved under different biogeographical and sociocultural conditions. The names of these breeds have been deduced either solely from their original breeding region (e.g., Taihu geese from the Taihu area of the Yangtse River delta) or combined with their feather colour (e.g., Jiangshan White geese and Zhedong White geese have white feathers, while the feathers of Yongkang Grey geese and Panshi Grey geese are grey). These local breeds, which are on the Chinese national protection list, are valuable Chinese national genetic resources due 1 Corresponding author ( lulizhibox@163.com). to their strong adaptability to extensive management, high reproduction rate, and good disease resistance and meat quality (Xu and Chen 2004). In the past three decades, meat traits have been the primary breeding objective. The populations of some goose breeds with specific features are either declining or their breed characteristics are being diluted under the present production system, because of genetic admixture among some breeds (Li et al. 2007; Tang et al. 2009). For example, due to indiscriminate crossbreeding over a long period of time, the performance of Panshi Grey goose has Abbreviations: H O, observed heterozygosity; H E, unbiased expected heterozygosity; HWE, HardyWeinberg equilibrium; JS, Jiangshan White; LD, Landes goose; PIC, polymorphic information content; PS, Panshi Grey; TH, Taihu goose; YK, Yongkang Grey goose; ZD, Zhedong White Can. J. Anim. Sci. (2012) 92: doi: /cjas

2 418 CANADIAN JOURNAL OF ANIMAL SCIENCE declined significantly resulting in a reduction in population size. Although Yongkang Grey geese and Panshi Grey geese were derived from two different regions, they do not have significantly different phenotypic characteristics (Fang 2002). Therefore, both the breeds Yongkang Grey goose and Panshi Grey goose are at risk of losing their genetic structure and diversity, even becoming genetically homogeneous. Under these circumstances, it is necessary to evaluate the genetic diversity among and within these breeds to facilitate the optimal utilization of goose genetic resources and permit efficient genetic improvement for production and conservation needs. Molecular techniques have provided many kinds of markers for the study of genetic variation. Among these markers, microsatellites are particularly advantageous due to their large number of polymorphisms, abundance, co-dominant inheritance, analytical simplicity, and transferability (Mohsen and Ghodrat 2007). In recent years, microsatellite-based studies have been used for the genetic characterization and differentiation of some Chinese geese and to establish the genetic relationships among them; however, most research has focused on local breeds in northern and southern China (Liu et al. 2006; Li et al. 2007; Chen et al. 2009). Little information is available concerning the genetic diversity, structure, and degree of admixture of the local breeds in eastern China. The present study was undertaken to assess the genetic diversity, structure, and degree of admixture in six goose breeds (Panshi Grey goose, Yongkang Grey goose, Zhedong White goose, Taihu goose, Jiangshan White goose and Landes goose) using 29 microsatellite markers, and to compare five indigenous breeds with one introduced goose breed (Landes goose). MATERIALS AND METHODS Sample Collection and DNA Extraction This study was conducted in accordance with the Guide to the Care and Use of Experimental Animals and Ethics of Animal Investigation laid down by the Canadian Council on Animal Care. Table 1. Genetic variation at the 29 microsatellite loci in six goose breeds Mean heterozygosity z In total, 180 unrelated individuals of five indigenous goose breeds (Taihu goose, TH; Jiangshan White goose, JS; Zhedong White goose, ZD; Yongkang Grey goose, YK; Panshi Grey goose, PS) and one introduced breed (Landes goose, LD) were examined. Thirty individuals per population (15 males and 15 females) were randomly collected from six different native farms located in eastern China (Supplementary Table 1). All blood samples were taken from the wing vein of geese and stored at 208C. Genomic DNA was extracted from whole blood by the method of Tadano et al. (2007). Microsatellite Markers and Genotyping Twenty-nine microsatellite markers recommended by the MERPDC for goose diversity studies were used (Molecular Ecology Resources Primer Development Consortium 2010). All forward primers were 5?-labelled with a fluorescent dye [FAM or HEX, Invitrogen (Shanghai) Co., Ltd.]. Information specific to these loci is shown in Table 1. PCR was performed in a 15-mL volume containing 1.25 U of Taq DNA polymerase (Takara), 1PCR buffer, 1.5 mm MgCl 2, 200 mm each dntp, 0.5 mm each primer, and 50 ng of goose DNA. All amplifications were carried out using a GENEAMP PCR 9700 thermocycler (Applied Biosystems) with the following cycling parameters: 5 min at 958C, followed by 30 cycles of 40 s at 948C, 30 s at the primer-specific annealing temperature (Table 1), and 30 s at 728C, with a final 10-min elongation step at 728C. The genotypes for each marker were determined using an ABI 377 DNA Sequencer (Applied Biosystems) with the internal size standard GeneScan TM TAMARA 350 (Applied Biosystems). Statistical Analysis For each goose breed, the following statistics were calculated using CERVUS version 2.0 (Marshall et al. 1998): total number of alleles and mean number of alleles per locus, observed heterozygosity (H O ), unbiased expected heterozygosity (H E ), and polymorphic information content (PIC). ARLEQUIN version 3.1 (Excoffier et al. 2005) was used to estimate the HardyWeinberg Breed name y TNA MNA H O H E PIC F IS HWE (Nm) NA PS * 8 (0) 3 YK (0) 14 TH (1) 7 JS (0) 5 ZD * 7 (0) 4 LD * 8 (3) 12 z TNA, total number of alleles; MNA, mean number of alleles per locus; H O, observed heterozygosity; H E, expected heterozygosity; PIC, polymorphic information content; F IS, estimate of inbreeding coefficient; HWE, number of loci not in HardyWeinberg equilibrium (PB0.05); Nm, number of monomorphic loci; NA, number of alleles found in only one breed; *PB0.05 y PS, Panshi Grey goose; YK, Yongkang Grey goose; TH, Taihu goose; JS, Jiangshan White goose; ZD, Zhedong White goose; LD, Landes goose.

3 LI ET AL. * GENETIC DIVERSITY OF CHINESE GOOSE BREEDS 419 equilibrium (HWE) at each locus-line combination by a test analogous to Fisher s exact test based on the Markov chain method (Markov chain length, ; dememorization steps, ). F-statistics (F IS,F IT and F ST ) for each locus, more particularly, pairwise F ST between populations (Weir and Cockerham 1984) and the estimate of average inbreeding coefficient (F IS ), were also calculated using FSTAT version (Goudet and Keller 2002). The significance level for pairwise F ST was determined from the permutation test using the sequential Bonferroni procedure (Rice 1989). Nei s (1978) standard genetic distance among the populations was calculated by POPGENE version 1.31 (Yeh et al. 1997). A pairwise matrix of the genetic distance (D A ) was then used to obtain a neighbor-joining (NJ) tree (Saitou and Nei 1987), which was visualized using TreeView version (Page 1996). The population structure and degree of admixture were evaluated based on a Bayesian clustering analysis using STRUCTURE 2.1 (Pritchard et al. 2000). This method uses multilocus genotypes to assign individuals to populations, estimate individual admixture proportions, and infer the number of parental populations (K) for a given number (K) of clusters. For all analyses, the Markov chain Monte Carlo (MCMC) parameters were set to a burn-in period of 10 5 with 10 5 iterations. The optimum K, indicating the number of true clusters in the data, was determined from 10 replicate runs for each value of K (26) using the method described by Evanno et al. (2005) and the ad hoc statistic DK, based on the rate of change in the log probability of the data between successive K values. Parameters of the method of Evanno et al. (2005) were calculated using a web-based program ( biology.ucla.edu/structureharvester/) for collating results generated by the program STRUCTURE (Earl and vonholdt 2011). A graphical display of the structure results was generated using distruct software (Rosenberg 2004). RESULTS AND DISCUSSION Variation at Microsatellite Markers The range of allele sizes (ASR), number of alleles (N a ), number of unique alleles (Nua), H E, and F ST of the 29 microsatellite markers used to genotype the six goose breeds are presented in Supplementary Table 1. A total of 334 distinct alleles were detected at the 29 microsatellite loci in 180 birds. The average N a per locus was 11.5, with a range from 6 (ZAAS038) to 21 (ZAAS182). Barker (1994) suggested that the microsatellite loci used in genetic distance studies should have more than four alleles in order to reduce the distance estimate standard error. This indicates that the loci in our study were sufficiently polymorphic for evaluating genetic diversity. Forty-five of the 334 alleles (13.5%) were unique to only one breed (unique alleles). Unique alleles were observed in 16 loci of all 29 microsatellites (55.2%). The greatest Nua was detected in ZAAS182, which had seven unique alleles across all breeds. Although several alleles were unique in the goose breeds, they were unlikely to be useful as breed markers because most of them were distributed with a low frequency; that is, 32 of 45 unique alleles (71.1%) had a frequency below 10%. Agha et al. (2008) proposed several reasons for the existence of unique alleles, including multi-origin of the breeds, little subsequent genetic exchange between them, or genetic drift. The H E per locus ranged from (ZAAS181) to (ZAAS142). The F ST per locus ranged from (ZAAS182) to (ZAAS035). In terms of the H E, higher values were observed at ZAAS142, ZAAS030, and ZAAS169. As for the allele number, ZAAS182, ZAAS181, and ZAAS144 were highly polymorphic. In regard to the F ST value, which indicates genetic variation between breeds, ZAAS035, ZAAS020, and ZAAS141 had higher values than the others. ZAAS030, ZAAS061, ZAAS142, and ZAAS169 showed a high H E and N a. This suggested that the above mentioned markers would be effective for this kind of study. Genetic Diversity within Breeds The genetic diversity within each breed is summarized in Table 1. The highest total number of alleles was observed in YK (171) and the lowest in LD (115). These two breeds also had the highest and lowest mean number of alleles per locus per breed, respectively. In the present study, a moderate level of genetic variability was revealed among five indigenous Chinese geese. The highest average Ho and He values were seen in YK (H O and H E 0.481), with an excess of heterozygotes (F IS 0.081) followed by TH (H O and H E 0.510, F IS 0.029) and JS (H O and H E 0.466, F IS 0.017), whereas the lowest value was observed in PS geese (H O and H E 0.495). The higher genetic diversity observed in YK can be explained by the mixing of populations from different geographical locations and natural selection favouring heterozygosity. The lower heterozygosity values observed in PS indicate some loss of genetic diversity. This could be attributed to its relatively small population size compared with YK, which is in agreement with its endangered status (Fang 2002). Across all breeds, the PIC Table 2. Pairwise F ST (below the diagonal) and Nei s genetic distance (above the diagonal) among the six goose breeds Line name PS YK TH JS ZD LD PS YK TH 0.063* JS 0.067* 0.065* 0.076* ZD 0.077* 0.068* 0.122* 0.071* LD 0.472* 0.454* 0.444* 0.443* 0.472* *F ST values are significantly different from 0 (PB0.05) after the Bonferroni correction.

4 420 CANADIAN JOURNAL OF ANIMAL SCIENCE Fig. 1. NJ trees of five indigenous Chinese goose breeds and one introduced goose breed based on the D A genetic distance. The numbers at the nodes represent the percentage of group occurrence in 1000 bootstrap replicates. The tree is unrooted. value for each breed ranged from in LD to in YK, and the five Chinese indigenous breeds had higher PIC values and heterozygosity than the introduced goose breed. Nua ranged from 3 (PS) to 14 (YK). In all 174 HWE tests (29 loci in six breeds), three cases could not be tested because they were monomorphic in the breed. Of the remaining 171 tests, significant deviations from HWE at the 5% level were observed in 43 cases. Of these deviations, 41.9% (18/43) of the cases had excess heterozygosity. Conversely, 58.1% (25/43) of the cases had a heterozygosity deficit. All breeds also significantly deviated from HWE (P B0.05), probably because of inbreeding, small population size, mutations and migrations. The number of loci that deviated from HWE ranged from 4 (TH) to 10 (YK). Genetic Differentiation and Relationships Among the Breeds The F ST per locus ranged from (ZAAS154) to (ZAAS035). An overall F ST value of across all breeds suggests that 75.8% of the total genetic variation was from genetic differentiation within each breed and 24.2% of the genetic variation existed among breed differentiation (Supplementary Table 2). This is slightly less than the previously reported value of 28.1% for the total genetic variance among other indigenous Chinese goose breeds (Chen et al. 2009). The pairwise Nei s D A and F ST are shown in Table 2. F ST among the six goose breeds ranged from between PS and YK to between PS and LD; similarly, Nei s D A varied between and The pairwise F ST between the five indigenous Chinese goose breeds was very small, reflecting strong genetic similarity among them, while they were different from the introduced breed (PB0.05). The low degree of genetic differentiation in five indigenous geese may have been due to the following: (1) in China, geese are reared mostly by small farms for meat production without systematic and strict selection for other economical traits; (2) the small effective population size might act against random drift, which promotes population differentiation; and (3) the close geographical distance (e.g., PS and YK) among some populations might enable frequent genetic exchanges. In contrast, large geographical barriers or national boundaries could prevent genetic exchange between indigenous and introduced goose breeds in China. Fig. 2. Structural analysis of five indigenous Chinese goose breeds and one introduced goose breed. Each individual animal is represented by a single vertical line divided into K colours, where K is the number of clusters assumed and the length of the coloured segment represents the individual s estimated proportion of membership to a particular cluster.

5 LI ET AL. * GENETIC DIVERSITY OF CHINESE GOOSE BREEDS 421 An NJ tree was constructed on the basis of the genetic distances with relatively high bootstrap values (Fig. 1). As expected, it showed two main separate clusters composed of LD and five Chinese indigenous breeds. The indigenous breeds clustered further into two genetic groups; one group had only JS, while the other group contained the other indigenous breeds. YK and PS were closest with TH deviating from the pair, still with a low bootstrap value (43%). In China, many breeders believe that the PS goose breed originated from the YK goose breed. Based on our phylogenic analysis, pairwise F ST and Nei s genetic distance, the PS and YK breeds are genetically very close. In phylogenetic analyses, low bootstrap values imply that the classifications observed on the dendrogram are not well supported because of admixing among the populations and thus do not represent distinct evolutionary units. Due to this reason, the clustering algorithm implemented in STRUCTURE (Pritchard et al. 2000) was designed to overcome the limitations inherent in phylogenetic tree models and has been applied to infer the genetic structure in several species (Álvarez et al. 2004; Bodzsar et al. 2009; Collins et al. 2011). STRUCTURE analysis defined the degree of admixture for each individual animal. In this study, we used the methodology of Evanno et al. (2005) to estimate the plateaus of log likelihood, suggestive of an appropriate K for STRUCTURE analysis. We found the maximum value of DK at K3 (Supplementary Fig. 1). At the lowest K value (K2), the introduced goose breed LD split from others until the highest value (K 6). At K 3, all of the indigenous populations showed an admixture pattern between the two distinct clusters. At K4, an additional fraction of the third ancestry cluster was detected for all of the indigenous populations, but this result was erratic. At higher K values, the two main clusters persisted with inconsistent partitioning into further clusters that were not biologically meaningful. Our current results indicate that three indigenous goose breeds (PS, YK, and ZD) have been hybridized more frequently, leading to a loss of genetic distinctiveness. In conclusion, this study provides a preliminary estimate of overall genetic diversity and population structure of six Chinese goose breeds. Results exhibited abundant genetic variation and obvious genetic structure, which is important for future breeding. However, some indigenous goose breeds are at risk of becoming genetically homogenous unless effective and appropriate breeding management practices are implemented. To maintain the present genetic diversity and structure of these breeds, genetic exchanges between breeds must be controlled on every conservation farm. ACKNOWLEDGEMENTS This work was supported by the Science and Technology Department of Zhejiang province (2007C12058; 2009C12079) and the earmarked fund for Modern Agro-industry Technology Research System, China (CARS-43-02). Agha, S. H., Pilla, F., Galal, S., Shaat, I., D Andrea, M., Reale, S., Abdelsalam, A. Z. A. and Li, M. H Genetic diversity in Egyptian and Italian goat breeds measured with microsatellite polymorphism. J. Anim. Breed. Genet. 125: A lvarez, I., Royo, L. J., Ferna ndez, I., Gutie rrez, J. P., Gómez, E. and Goyache, F Genetic relationships and admixture among sheep breeds from Northern Spain assessed using microsatellites. J. Anim. Sci. 82: Barker, J. S. F A global protocol for determining genetic distance among domestic livestock breeds. Proc. 5th World Congr. Genet. Appl. Livest. Prod. Guelph, Canada 21: Bodzsar, N., Eding, H., Revay, T., Hidas, A. and Weigend, S Genetic diversity of Hungarian indigenous chicken breeds based on microsatellite markers. Anim. Genet. 40: Chen, G. H., Wang, K. H., Wang, J. Y., Ding, C. and Yang, N Poultry genetic resources in China. Shanghai Scientific and Technical Publishers, Shanghai, China. Chen, K. W., Song, W. T., Xu, W. J., Zhu, W. Q., Li, H. F. and Tang, Q. P Assessment of genetic diversity and genetic differentiation of Chinese native goose using microsatellite markers. Acta Vet. Zootech. Sina. 40: Collins, O., Alfred, L. R., Tom, W., Evans, O., Raballah, Nicholas, O., Oguge, Walter, G. Z. O., Jura, W. O., Olivier, H. and Nicholas, G Genetic structure of hartebeest populations straddling a transition zone between morphotypes. J. Basic Appl. Sci. Res. 1: Earl, D. A. and vonholdt, B. M STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv. Genet Resour. (in press). Evanno, G., Regnaut, S. and Goudet, J Detecting the number of clusters of individuals using the software STRUC- TURE: a simulation study. Mol. Ecol. 14: Excoffier, L., Laval, G. and Schneider, S Arlequin ver. 3.0: An integrated software package for population genetics data analysis. Evol. Bioinform. Online. 1: Fang, M. H Status of the livestock production in the city of Yueqing, Zhejiang Anim. Husb. Vet. 2: Goudet, J. and Keller, L The correlation between inbreeding and fitness: does allele size matter? Trends Ecol. Evol. 17: Li, H. F., Chen, K. W., Yang, N., Song, W. T. and Tang, Q. P Evaluation of genetic diversity of Chinese native geese revealed by microsatellite markers. World s Poult. Sci. J. 63: Liu, S., Li, P., Song, Y., Li, S. Z., Wei, C. B. and Yang, H. M Analysis of genetic variations in different goose breeds using microsatellite markers. Hereditas 28: Marshall, T. C., Slate, J., Kruuk, L. E. B. and Pemberton, J. M Statistical confidence for likelihood-based paternity inference in natural populations. Mol. Ecol. 7: Mohsen, G. and Ghodrat, R. M Use of microsatellite markers in poultry research. Int. J. Poult. Sci. 6: Molecular Ecology Resources Primer Development Consortium Permanent genetic resources added to the molecular ecology resources database 1 February Mol. Ecol. Resourc. 10:

6 422 CANADIAN JOURNAL OF ANIMAL SCIENCE Nei, M Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89: Page, R. D. M TREEVIEW: An application to display phylogenetic trees on personal computers. Comp. Appl. Biosci. 12: Pritchard, J. K., Stephens, M. and Donnelly, P Inference of population structure using multilocus genotype data. Genetics 155: Rice, W. R Analyzing tables of statistical tests. Evolution 43: Rosenberg, N. A Distruct: a program for the graphical display of population structure. Mol. Ecol. Notes 4: Saitou, N. and Nei, M The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4: Tadano, R., Sekino, M., Nishibori, M. and Tsudzuki, M Microsatellite marker analysis for the genetic relationships Supplementary Table 1. Information on the breeds used in this study among Japanese long-tailed chicken breeds. Poult. Sci. 86: Tang, Q. P., Zhang, S. J., Guo, J., Chen, K. W., Lu, H. L. and Su, J. D Microsatellite DNA typing for assessment of genetic variability in Taihu goose: a major breed of China. J. Anim. Vet. Adv. 8: Weir, B. S. and Cockerham, C. C Estimating F-statistics for the analysis of population structure. Evolution 38: Xu, G. F. and Chen, K. W Photograph album of Chinese indigenous poultry breeds. Chinese Agriculture Publisher. Beijing, China. Yeh, F. C., Yang, R., Boyle, C., Timothy, B. J., Ye, Z. H. and Mao, J. Y. X POPGENE, the user-friendly shareware for population genetic analysis. Molecular Biology and Biotechnology Centre, University of Alberta, Edmonton, AB. Goose breeds Acronym Specific features Origin/population situation Panshi Grey goose PS Medium meat-purpose, gray plumage Yueqing City/declining. Jiangshan White goose JS Medium meat-purpose, white plumage Jiangshan City/increasing. Yongkang Grey goose YK Large meat-purpose, gray plumage Yongkang County/declining. Zhedong White goose ZD Medium meat-purpose, white plumage Xiangshan County/increasing. Taihu goose TH Small egg-purpose, white plumage Huzhou City/increasing. Landes goose LD Fat liver purpose, gray plumage Changxin County/increasing. Supplementary Table 2. Information on the 29 microsatellite loci used in this study z Locus Ta (8C) Na (Nua) ASR (bp) H E H O PIC F IS F IT F ST ZAAS (0) * 0.327* 0.177* ZAAS (0) * ZAAS (2) * ZAAS (3) * * ZAAS (0) * 0.293* 0.269* ZAAS (0) * * ZAAS (1) * 0.327* ZAAS (0) * 0.119* 0.211* ZAAS (3) * 0.182* ZAAS (3) * * ZAAS (3) * ZAAS (2) * ZAAS (0) * 0.425* 0.294* ZAAS (3) * ZAAS (2) * 0.200* 0.288* ZAAS (0) * 0.349* ZAAS (0) * * ZAAS (1) * 0.200* ZAAS (2) * * ZAAS (0) * 0.480* ZAAS (3) * 0.485* 0.185* ZAAS (3) * ZAAS (0) * * ZAAS (0) * 0.201* 0.068* ZAAS (0) * ZAAS (0) * 0.315* 0.263* ZAAS (1) * 0.219* ZAAS (6) * ZAAS (7) * 0.214* 0.095* All loci 334 (45) z N a, number of alleles per locus; Nua, number of unique alleles across six goose breeds; ASR, detected range of allele size; H E, unbiased expected heterozygosity; F ST, fixation index; PIC, polymorphism information content; *PB0.05.

7 LI ET AL. * GENETIC DIVERSITY OF CHINESE GOOSE BREEDS 423 Supplementary Fig. 1. Semi log plot of DK as a function of K, the number of clusters.