Subscribe Now Subscribe Today
Research Article
 

Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan



R.M. Al-Atiyat, M.J. Tabbaa, N.M. Salameh, K.A. Tarawneh, L. Al-Shmayla and H.J. Al-Tamimie
 
Facebook Twitter Digg Reddit Linkedin StumbleUpon E-mail
ABSTRACT

Genetic diversity of fat-tailed sheep of Jordan was investigated using microsatellite markers (MS). Six ovine and bovine MS located on chromosomes 2 and 6 of sheep genome were employed for genotyping three flocks in southern region of Jordan. The genetic diversity was evaluated by measuring expected (He) and observed (Ho) heterozygosity, Polymorphic Information Content (PIC) and the number of alleles per locus. The latter indicator showed specific allele frequency profiles in which the level of genetic differentiation between flocks observed as differences in allele frequencies and allele sizes. Therefore, allele frequency profiles were utilized to explain the level of genetic differentiation between flocks. Overall Ho was 0.67 for all flocks, whereas He was 0.70. The average He for each flock was 0.667, 0.652 and 0.716, while PIC was 0.639, 0.643 and 0.714 for flocks 1, 2 and 3, respectively. On the other hand, genetic distances undoubtedly revealed the expected degree of differentiation among the three flocks. The finding showed closeness of flocks 1 and 2 to each other and Flock 3 to Flock 2 than to Flock 1. Despite the three flocks were small in size number, had effective selective mating and gene flow, they matched with the Hardy-Weinberg Equilibrium, strongly proving the usefulness of MS as a genetic tool in population genetics analyses of sheep. From a genetic conservation point of view, it is recommended to maintain genetic diversity of small flocks when genetic migration of good genetic resources is absence. The present study provides new information about the population genetics parameters of fat-tailed sheep in southern Jordan.

Services
Related Articles in ASCI
Search in Google Scholar
View Citation
Report Citation

 
  How to cite this article:

R.M. Al-Atiyat, M.J. Tabbaa, N.M. Salameh, K.A. Tarawneh, L. Al-Shmayla and H.J. Al-Tamimie, 2012. Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan. Asian Journal of Animal and Veterinary Advances, 7: 376-389.

DOI: 10.3923/ajava.2012.376.389

URL: https://scialert.net/abstract/?doi=ajava.2012.376.389
 
Received: November 07, 2011; Accepted: January 13, 2012; Published: February 17, 2012



INTRODUCTION

Small ruminants such as sheep and goat are important for the livelihood of farmers in the world. In particular, the Mediterranean area is an effective transformation area for thousands of marginal hectares into high quality animal protein mostly by sheep (Boyazoglu and Flamant, 1990). Sheep in Jordan is a fat-tailed breed known as the Awassi sheep (Hailat, 2005) which possesses great adaptability to tropical environmental conditions. Galal et al. (2008) stated that Awassi sheep is often used as a triple purpose sheep, better for high milk production, in most of the countries of the Middle East.

On the other hand, Jordan has experienced diverse roles of sheep in poverty alleviation and as a result, farmers increasingly turn to sheep as their main source of life and income for its unique products of Jameed, fermented dried yoghurt (Al-Omari et al., 2008) and meat (Quasem et al., 2009). Unfortunately, fat-tailed sheep has recently faced major threats to their genetic diversity which resulted from systemic, regional and global economic crises as well as drought. A major threat was the sharp reduction in number as a result of persistent drought since 2007. Most of the sheep population is found in north region of Jordan (60%) while (26%) is found in center region and the rest (14%) is found in the south region (MOA, 2010) where drought has mostly been hit.

It is widely accepted that domesticated animals descended from a single ancestor, originating in Asia. It is also reported that all world's sheep breeds originated from central Asia since stone ages (Crawford, 1984). At the present time, the Mediterranean domesticated sheep breeds are the most closely related to their ancestors, which were longtime ago the animals brought from Fertile Crescent (Moiseeva et al., 1996) which includes Jordan. However, there is no published data on the fat-tailed sheep biodiversity or its ancestors in Jordan. The use of advanced molecular DNA technology has provided wide opportunities to analyze genetic variability at the DNA level in animal breeds of Jordan (Al-Atiyat, 2009). Such technology is microsatellites DNA markers that are widely used in livestock species since they are polymorphic and are randomly distributed in the organism’s genome (Karaca et al., 1999; Ahmadi et al., 2007; Sulaiman et al., 2011; Ismoyowati and Purwantini, 2011). These markers have also been successfully used to study the biodiversity and genetic relationship between and within sheep breeds and/or populations (Romanov and Weigend, 2001; Rosenberg et al., 2001). They provide reliable information on allele profiles and allele frequencies for a single DNA sample that can be extracted from blood or tissue.

Limited information is available on the molecular genetic variation of Awassi sheep in Jordan (Galal et al., 2008). Therefore, the present study aimed to analyze the genetic diversity of fat-tailed sheep in the southern region of Jordan and to utilize the possible findings for future management conservation of genetic resources of fat-tailed sheep in Jordan.

MATERIALS AND METHODS

Sheep flocks: Three flocks of Awassi sheep were studied in southern region of Jordan in Maan governorate. Capital departments of Maan governorate include the city of Maan, Showbak, Husseiniya and Petra (Alhroot and Al-Alak, 2009) (Fig. 1). The three flocks were selected from Husseiniya and Petra and named as Flocks 1, 2 and 3, respectively. Flock 1 was 30 km away from Flock 2, whereas Flock 3 was 50 km away from Flock 2 within an area of oldest and most known historical places of Petra, Showbak Castle and Dana nature reserve (Fig. 1). The three flocks were managed by their own owners and have not been in connection to exchange animals into their flock from each other for the past 50 years.

Sampling and DNA extraction: Tissue sample were collected from ewes and rams of each flock. The total number of sheep samples was 140 samples (Table 1) taken as 0.5 cm tissue samples from ears using an ear puncher. Samples were placed in Eppendorf microcentrifuge tubes and stored at -18°C until analyzed. DNA extraction was performed using a commercially available kit/protocol of E.Z.N.A® MicroElute Genomic DNA extraction Kit (OMEGA Bio-Teck Corporation, 2009).Subsequently, DNA concentrations were estimated by Nano-DNA spectrophotometer (AlphaSpec®) in which the quality of DNA was evaluated using the ratio of A260/A280.

Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan
Fig. 1: Map showing regions of the three sampled sheep flocks

Table 1: Number of animals sampled collected from the three flocks
Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan

Table 2: Microsatellite markers used for genotyping the three flocks
Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan
Ch: Chromosome no. in sheep genome; F: Forward primer; R: Reverse primer

DNA genotyping: Six ovine and bovine MS (Table 2), located on chromosomes 2 and 6, were employed for genotyping experiments using Silver Sequence™ DNA System of Promega® company (Promega Corporation, 1998). Selection of the markers was based upon their close linkage to each other. On the other hand, their primers were selected for ease of use in PCR reaction with special regard to the annealing temperature and MgCl2 concentration in particular. Primer sequences were taken from the Australian Sheep Gene Mapping website (Maddox et al., 2001) and synthesized by BioEngland® (address) (Table 2).

PCR reaction utilized 10 μL volumes of DNA and reagents for genotyping. DNA samples were liquated into a 48 well PCR plate. Thermal cycling was performed on an MJ Research PTC-100 thermal-cycler. The Amplified PCR products were resolved on a 5% polyacrylamide gel electrophoresis using a Sequi-Gen GT gel rig for Silver staining of Promega® company. Sequencing ladders were prepared using fmol® DNA Cycle Sequencing system (Promega Corporation, 1998) and 3 μL of each of the four reactions loaded onto the gel, so that the size of the microsatellite alleles could be determined. When the electrophoresis run was completed, the gel was recovered and developed. Then, the gel was dried and viewed by APC Film Development method. The film was developed as a photo picture to be ready for scoring the genotypes. Allele sizes were scored by visual comparison with the sequencing ladder; pGEM®-3Zf (-) Vector.

Genetic analysis: Population genetics of the three studied flocks was investigated. Allele frequency and polymorphism under Hardy-Weinberg equilibrium (HWE) were analyzed using Cervus software package (Marshall et al., 1998). The first parameter was observed heterozygosity (Ho) which was calculated as:

Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan

and the second was expected heterozygosity (He) at each locus calculated as (Nei, 1987):

Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan

where, N is the number of individuals in the sample, Nij is the number of observed heterozygotes, n is the number of alleles and pi is the frequency of the ith allele.

Furthermore, third parameter was polymorphic information content (PIC) which was estimated to describe the amount of polymorphism at a single locus using the following formula by CERVUS software (Botstein et al., 1980):

Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan

where: pi is the population frequency of the i-th allele and pj is the frequency of the jth allele.

On the other hand genetic distances matrix between populations were measured using Genetic Data Analysis (GDA) software (Weir, 1996),which utilizes the most widely used measure of genetic distances proposed by Nei (1972) as:

Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan

where, Gi, Gj and Gij are frequencies of the ith and jth allele respectively drawn in populations i and j (Nei, 1972). Using genetic distance matrix, the neighbor-joining tree was formulated using MEGA software program, version 5 (Tamura et al., 2011).

RESULTS AND DISCUSSION

DNA quantity and quality: The DNA concentration of the samples was determined for optimal MS genotyping and prior to the PCR amplification step in the genotyping process. The results of DNA concentration showed wide range of 50 to 400 ng 1 μL-1 (Fig. 2). This difference in concentration was due mainly to size of sheep ear tissue and to the amount and time of Proteinase-K enzyme that also used in the extraction process and the incubation period. In general, the best performance with PCR amplification of MS markers using commercially available kits occurred within a fairly narrow range of input DNA amount typically on the order of 5 to 2.5 ng per 1 μL (Sambrook et al., 1989; Rane and Barve, 2011). Therefore, based on DNA concentration data, every sample was diluted to a concentration of 10 ng per 1 μL as a workable sample for PCR reactions. This was in one hand regarding to quantification of DNA extracted. On the other hand, quality of DNA extracted was also evaluated using the ratio at A260/A280 (Fig. 2). As a general rule, the quality of DNA is established by the ratio of absorbance at 260 nm to the absorbance of 280 nm. For highly purified DNA, this ratio is about 1.8 (Sambrook et al., 1989). Figure 4 shows that all samples’ absorbance ratios were averaged 1.83 and ranged from 1.6 to 2.2.

Genetic variation parameters
Allele frequency:
Estimates of allelic frequency of every locus were calculated to describe the genetic variation within and between flocks.

Table 3: No. of alleles per locus, expected heterozygosity (He), observed heterozygosity (Ho) at the marker loci, polymorphic information content (PIC) and allele sizes for 6 MS loci in the three flocks under Hardy-Weinberg equilibrium (HWE)
Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan
NS: Not significant at p<0.05

Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan
Fig. 2: DNA quantification of all samples using Nano-DNA spectrophotometer

Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan
Fig. 3: Allele frequency of the six studied marker for the studied sheep population

The described alleles and their frequencies in the six studied loci for each flock, are presented in Fig. 3, while the average number of allele per locus, Ho, He and PIC are presented in Table 3.

The average number of alleles per loci for all population was 6.17, whereas the number of alleles per each locus ranged from 5 to 8 (Table 3). The average allele number for each flock was 5.5 4.7 and 5.5 for Flock 1, Flock 2 and Flock 3, respectively (Table 4). This indicates a high number of MS alleles in studied sheep flocks. MS loci varied in number of described alleles from 7 (OARHH55) in Flock 1 to 3 (OARAE101) in Flock 3. It is worthy to make full description of allele frequency profile in which allele frequency and size profile are presented (Fig. 3). In particular, there is a high number of MS alleles as well as a wide range of allele sizes for each locus in the studied flocks. Similar observation was reported by Arranz et al. (1998) who estimated MS variation of Spanish Merino sheep. Furthermore, a specific profile of allele frequencies at each locus was observed, reflecting specific distributions of allele frequencies and sizes (Fig. 3). Such profiles are mostly reflecting a normal distribution and thus could be used to predict the level of the genetic variation and genetic differentiation within and between populations. Similar observations of population genetic variations using allele specific profiles were reported for sheep populations by Iovenko (2002) and Peter et al. (2007). Furthermore, a similar pattern of allele frequency profiles in sheep has been reported earlier by Arranz et al. (2001). On the other hand, Meadows et al. (2008) reported that analysis of genetic diversity within five sheep breeds showed merino contained the highest genetic diversity as average number of alleles observed per locus was 8.13 and He was 0.70, whereas Macarthur Merinos contained the lowest amount of diversity, with an average number of alleles of 3.03 and He of 0.4.

Furthermore, the MS loci are different in the allele frequencies in three flocks, considerably from locus to locus and from flock to flock (Fig. 4). Again, in the three flocks, the profile of allele frequencies appears to be similar to binomial distribution (for example INRA040 and ILSTS30 for the three flocks). Despite the significant variation of allele frequencies from locus to locus within and between flocks, there are some obvious similarities. For instance allele 222 (ORAHH30) is by far the most common in all three flocks and in each one (Fig. 4), followed by 152 (ILSTS30). In general, the majority of alleles were found in all three populations, except the alleles with low frequencies. However, the comparisons between three flocks showed that some alleles are not commonly shared between them. For example, allele 231 of INRA40 was not found in Flock 2 but found in Flock 1 and Flock 3.

Table 4: No. of alleles per locus, expected heterozygosity (He), observed heterozygosity (Ho) at the marker loci, olymorphic information content (PIC) and allele sizes for 6 MS loci in each sheep flock
Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan

Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan
Fig. 4: Allele frequency of the six studied markers for of the three flocks

Also, allele 118 of ORAHH55 was only found in Flock 1 (Fig. 4). Two possible conclusions can be drawn. First, locus-specific profiles of allele frequencies, at least in some loci, were inherited from the common ancestor. Second there might be some forces that are maintaining such similarity.

On the other hand, it is notable that the most frequent alleles in one population were quite often frequent in the others. Some exceptions are there where the most frequent alleles in a flock were not most frequent in others like allele 126 of ORAHH55 in Flock1. In some cases however it is vice versa. Generally, alleles of extreme size in all flocks and loci occur at very low frequencies. Similar observations were reported for human populations by Zapata et al. (2001) and for sheep breeds by Peter et al. (2007). This may indicate that the alleles did not get an advantage either due to random or selective forces. There are little variations in their allele frequencies profile between flocks. This probably reflects that random drift in Flock 1 and Flock 2 and gene flow into Flock 3 and may contribute to this genetic variation. Allelic frequencies and sizes of alleles for Flock 3 were found to be highly variable from those in Flock 1 and Flock 2.

Heterozygosity and polymorphic information content (PIC): The commonly used measures of genetic diversity at a single locus, expected and observed heterozygosity and PIC, are shown in Table 3 for all flocks and in Table 4 for each flock. Overall Ho was 0.67 for all flocks. In most of the cases, Ho was lower than He and resulted overall in slightly higher average He of 0.7 for all flocks (Table 3, 4). The average He for each flock was 0.667, 0.652 and 0.716 for Flocks 1, 2 and 3, respectively (Table 4). These results showed high He in all flocks and for most of studied loci except OARHH30 in which Ho was 0.47 and He is 0.55. In regards, each MS locus, OARHH30 showed lowest in Flock 1 and Flock 3, whereas INRA40 showed the lowest Ho and He in Flock 2. While, the highest values of both Ho and He were for BM143 locus. In a similar study, based on three microsatellite loci, the heterozygoties of Awassi, Kivircik, Akkaraman breeds of Turkey as well as two of their crossbreeds were high and ranged from 0.667 to 0.782 (Soysal et al., 2005). The average expected heterozygosity was 0.72 and the average PIC was 0.67 in Afshari sheep breed of Iran (Qanbari et al., 2007). Arora et al. (2011) reported that both Ho and He averaged 0.665 and 0.786, respectively and ranged from 6.40 to 7.92 in six Indian sheep breeds. The results also show that PIC values were high for all studied loci. It seems that high values of either Ho or He, were highly relevant to higher values of PIC.

The PIC average was 0.65 for overall flocks (Table 4). They were 0.639, 0.643 and 0.714 for Flock 1, Flock 2 Flock 3, respectively (Table 4). Similar range of results was reported by Arora et al. (2011) in which PIC varied from 0.543 to 0.929 and averaged 0.775. In the current study, the lowest value (0.483) was for ILSTS30 in Flock 2 and the highest value (0.737) of PIC was for BM143 in Flock 3 (Table 4). In general, estimates of He and Ho and PIC at the 6 studied loci were high and close to each other. These results were expected for microsatellite loci, which demonstrated high polymorphism in all species studied so far. In addition, it reflects the well-known relationship between high number of alleles per MS locus and as a consequence high heterozygosity and high PIC. These results closely match those of Uzun et al. (2006) who stated that the mean number of alleles per locus ranged from 9.3 to 10.4, except for Hemsin sheep (7.8) and for Spanish Churra sheep (9.2). They also reported that the gene diversities (He) varied from 0.694 in Hemsin to 0.738 in all studied Turkish sheep. The overall level of microsatellite loci variation observed in this study is within the recently reported range observed for different worldwide sheep breeds (Zahedi-Zahra et al., 2007; Alvarez et al., 2009; Ligda et al., 2009; Zhong et al., 2010; Arora et al., 2011).

The three studied flocks with small sheep numbers revealed high level of genetic variation seen from high number of alleles per locus, He, Ho and PIC. These results were similar to those reported in other studies of microsatellite markers in sheep populations such as Bancroft et al. (1995) and Coltman et al. (2003). The most genetically variable flock was found to be Flock 3, which had more alleles per locus, higher He, Ho and PIC. This might be a result of gene flow into this flock, whereas the other two were using more of a self-replacing strategy. This observation was explained as when genetic material, in form of male gametes mainly, migrates from one flock to another (Alvarez et al., 2004) statement is vague. However, the high level of genetic variation in both Flock 1 and Flock 2 could be due to several possible factors such as management and breeding practices under which selection took place. The latter is more likely to explain the high genetic variation that could be a result of selection in these flocks as explained by Woolaston and Piper (1996). This explanation may receive relative attention whether variation can be explained by selection. Different types of selection and their effects were extensively studied and then reported in this regard. For example balancing selection as an example might act directly or indirectly on the trait by maintaining variation at loci. Direct selection can maintain variation by inducing heterozygote advantage (Barton and Slatkin, 1986). The major type of selection in the studied flocks was selecting sires from within-family dam of good reputation. The consequences of family selection on the viability of small populations both in the short and in the long term was studied using simulation model by Theodorou and Couvet (2003). They suggested that family selection could be proposed for use in management programs of small flocks since it increases genetic variability and short-term viability. Generally, the conditions assumed in this simulation (small size, well managed and family selection) were typically the same conditioned applied for the three studied sheep flocks. On the other hand, Kijas et al. (2009) revealed that the genetic structure of sheep in the world was a result of selection. Therefore, it is plausible to suggest that selection of within-family was the cause of high genetic variation within the sheep flocks. From genetic conservation point of view, this management could be recommended to maintain biodiversity of small flocks in case of genetic migration of good genetic resources is absence.

Finally, deviations from HWE in this study were estimated utilizing genotypic data in the three flocks (Table 4) and using Bonferroni correction test. There was no significant deviation from HWE (Table 4) for all studied loci in three flocks.

Table 5: Pairwise genetic distances between the three flocks
Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan

Image for - Analysis of Genetic Variation of Fat Tailed-sheep in Southern Region of Jordan
Fig. 5: Genetic relationship among the three sheep flocks according to the neighbour-joining phenogram with its location

Corresponding observations were found for no consistent HWE deviations across five Turkish sheep breeds (Uzun et al., 2006) and for any of the MS loci analyzed. In contrast, Zahedi-Zahra et al. (2007) reported that twelve loci were at Hardy-Weinberg equilibrium (p<0.005) in Iranian sheep population. However, analysis of deviations from HWE for each locus was then executed in each population revealing that only one MS locus, INRA40, in Flock 3 was significantly deviated from HWE (Chi-square = 8.23, p<0.004). The significant deviations from HWE are usually observed when there was an excess of heterozygotes or deficiency of heterozygotes (Wahlund effect) (Guo and Thompson, 1992). In this case of INRA40, Flock3 showed heterozygote excess in the locus but without significant deviation caused by heterozygote deficit in this flock. A conclusion can be drawn that the small population size, limited number of used sires, which leads to nonrandom mating and selection pressures were sufficient factors to significantly shift the genetic structure of Flock3 from the Hardy-Weinberg proportions in this locus. It is worthy to note that Hardy-Weinberg principle states that single-locus genotypic frequencies after one generation of random mating can be represented by a binomial (two alleles) or multinomial (with multiple alleles) function of allelic frequency (Hedrick, 2000). Overall, the studied sheep flocks were in HW equilibrium, even though they were small in size and non-randomly mating took place of a few sires.

Genetic distances: Genetic distances were estimated as pairwise genetic distances using the GDA® software (Weir, 1996) from the allele differences between two flocks as 0.026 between Flock 1 and Flock 2, 0.035 between Flock 1 and Flock 3 and finally 0.033 between Flock 2 and Flock 3. This was based on common alleles per locus that should be used to estimate genetic distances, as suggested by Kalinowski (2002). For genetic distances, it is unity ? when they have no common alleles and zero if the alleles are identical. As expected, the pairwise genetic distance between Flock 1 and Flocks 2 was relatively short; 0.026 (Table 5). The distances between these flocks were described by phylogenetic tree which is shown in Fig. 5 as neighbor-Joining phenogram (Tamura et al., 2011). Figure 5 revealed a considerable degree of differentiation between flocks. The finding firstly supports closeness of Flock 1 and Flock 2 to each other. It also shows that Flock 3 is slightly closer to the Flock 2 than to Flock 1. These results are in agreement with known history of the three flocks in regards to their location and thus possible gene flow and their common ancestors.

CONCLUSION

The three studied flocks with small sheep numbers revealed a high level of genetic variation. The genetic variation was revealed by measuring He and Ho, PIC and the number of alleles per locus. The latter showed specific allele frequency profiles which can be used for different genetic and evolutionary studies, in particular for genetic conservation of livestock in hand of their owners. Furthermore, allele frequency profiles were useful tools to explain the level of genetic differentiation between populations using the differences in allele frequencies and allele sizes. In general, estimates of He and Ho and PIC at the six studied loci were high and close to each other. These results reflected the well known relationship between high number of alleles per MS locus and as a consequence high heterozygosity and PIC values. In addition, the results also showed whenever PIC values were high for studied loci, the values of either Ho or He was high. These results were expected for microsatellite loci, which demonstrated high polymorphism in all species studied so far.

Despite the three flocks were small in number and both selective mating and possible gene flow were effective, they were in HWE. Therefore, this is a proof of MS usefulness as a genetic tool in genetic population analysis of sheep. The results clarified that for a population of small size, policy can be very effective for preserving significant level of genetic variation. Genetic distances also undoubtedly revealed the expected degree of differentiation in the three flocks. From genetic conservation point of view, it is recommended to maintain biodiversity of small flocks in case of genetic migration of good genetic resources is absence. Overall, the present study sheds new light on the molecular and population genetics of fat-tailed sheep in Jordan.

ACKNOWLEDGMENT

This project was supported by King Saud University, Deanship of Scientific Research, College of Food and Agriculture Sciences, Research Center.

REFERENCES

1:  Ahmadi, A.K., G. Rahimi, A. Vafaei and H. Sayyazadeh, 2007. Microsatellite analysis of genetic diversity in pekin (Anas platyrhynchos) and muscovy (Cairina moschata) duck populations. Int. J. Poult. Sci., 6: 378-382.
CrossRef  |  Direct Link  |  

2:  Al-Omari, A., J.M. Quasem and A.S. Mazahreh, 2008. Microbiological analysis of solar and freeze-dried jameed produced from cow and sheep milk with the addition of carrageenan mix to the jameed paste. Pak. J. Nutr., 7: 726-729.
CrossRef  |  Direct Link  |  

3:  Al-Atiyat, R.M., 2009. DNA polymorphism of indigenous chickens in Jordan. Aian J. Anim. Vet. Adv., 4: 237-244.
CrossRef  |  Direct Link  |  

4:  Alhroot, A.H.H. and B.A.M. Al-Alak, 2009. Tourism roots and destination marketing bridging civilizations, cultural heritage and tourism theories. Asian J. Market., 3: 108-116.
CrossRef  |  Direct Link  |  

5:  Alvarez, I., A. Traore, H.H. Tamboura, A. Kabore and L.J. Royo et al., 2009. Microsatellite analysis characterizes burkina faso as a genetic contact zone between sahelian and djallonke sheep. Anim. Biotechnol., 20: 47-57.
CrossRef  |  Direct Link  |  

6:  Alvarez, I., L.J. Royo, I. Fernandez, J.P. Gutierrez, E. Gomez and F. Goyache, 2004. Genetic relationships and admixture among sheep breeds from Northern Spain assessed using microsatellites. J. Anim. Sci., 82: 2246-2252.
PubMed  |  Direct Link  |  

7:  Arora, R.J., S. Bhatia, B.P. Mishra, A. Jain and B. Prakash, 2011. Diversity analysis of sheep breeds from Southern peninsular and Eastern regions of India. Trop. Anim. Health Prod., 43: 401-408.
CrossRef  |  

8:  Arranz, J.J., Y. Bayon and F. San Primitivo, 1998. Genetic relationships among Spanish sheep using microsatellites. Anim. Genet., 29: 435-440.
CrossRef  |  PubMed  |  Direct Link  |  

9:  Arranz, J.J., Y. Bayon and F.S. Primitivo, 2001. Genetic variation at microsatellite loci in Spanish sheep. Small Ruminant Res., 39: 3-10.
CrossRef  |  PubMed  |  

10:  Bancroft, D.R., J.M. Pemberton and P. King, 1995. Extensive protein and microsatellite variability in an isolated, cyclic ungulate population. Heredity, 74: 326-336.
PubMed  |  

11:  Barton, N.H. and M. Slatkin, 1986. A quasi-equilibrium theory of the distribution of rare alleles in a subdivided population. Heredity, 56: 409-415.
CrossRef  |  

12:  Botstein, D., R.L. White, M. Skolnick and R.W. Davis, 1980. Construction of a genetic linkage map in man using restriction fragment length polymorphisms. Am. J. Hum. Genet., 32: 314-331.
PubMed  |  Direct Link  |  

13:  Boyazoglu, J. and J.C. Flamant, 1990. Mediterranean Systems of Animal Production. In: The World of Pastoralism: Herding Systems in Comparative Perspective, Galaty, J.G. and D.L. Johnson (Eds.). Guilford Press, New York, USA., pp: 353

14:  Coltman, D.W., J.G. Pilkington and J.M. Pemberton, 2003. Fine-scale genetic structure in a free-living ungulate population. Mol. Ecol., 12: 733-742.
CrossRef  |  

15:  Crawford, R.D., 1984. Domestic Fowl. In: Evolution of Domesticated Animals, Mason, I.L. (Ed.). Longman, London, UK., pp: 298-311

16:  Galal, S., O. Gursoy and I. Shaat, 2008. Awassi sheep as a genetic resource and efforts for their genetic improvement-A review. Small Rumin. Res., 79: 99-108.
CrossRef  |  Direct Link  |  

17:  Guo, S.W. and E.A. Thompson, 1992. Performing the exact test of Hardy-Weinberg proportion for multiple alleles. Biometrics, 48: 361-372.
PubMed  |  Direct Link  |  

18:  Hailat, N., 2005. Small Ruminant Breeds of Jordan. In: Characterization of Small Ruminant Breeds in West Asia and North Africa, Iniguez, L. (Ed.). Vol. 1. ICARDA, Aleppo, Syria, pp: 30-61

19:  Hedrick, P.W., 2000. Genetics of Populations. 2nd Edn., Jones and Bartlett Publishers International, London, UK., ISBN-13: 978-0763710767, Pages: 553

20:  Kalinowski, S.T., 2002. How many alleles per locus should be used to estimate genetic distances? Heredity, 88: 62-65.
CrossRef  |  PubMed  |  Direct Link  |  

21:  Karaca, M., F.G. Karaca, C. Patel and M.G. Emara, 1999. Preliminary analysis of microsatellite loci in commercial broiler chickens. Proceedings of the 7th Conference on Plant and Animal Genome, January 17-21, 1999, San Diego, CA., USA., pp: 155-

22:  Kijas, J.W., D. Townley, B.P. Dalrymple, M.P. Heaton and J.F. Maddox et al., 2009. A genome wide survey of SNP variation reveals the genetic structure of sheep breeds. PLoS ONE, Vol. 4,
CrossRef  |  Direct Link  |  

23:  Ligda, C., J. Altarayrah and A. Georgoudis, 2009. Genetic analysis of Greek sheep breeds using microsatellite markers for setting conservation priorities. Small Rumin. Res., 83: 42-48.
Direct Link  |  

24:  Iovenko, V.N., 2002. Genetic diversity of protein markers in sheep population from Ukraine. Genetika, 38: 1669-1676.
PubMed  |  

25:  MOA, 2010. Livestock annual statistics in Jordan. Ministry of Agriculture, Jordan.

26:  Maddox, J.F., K.P. Davies, A.M. Crawford, D.J. Hulme and E.P. Cribiu et al., 2001. An enhanced linkage map of the sheep genome comprising more than 1000 loci. Genome Res., 11: 1275-1289.
Direct Link  |  

27:  Marshall, T.C., J. Slate, L.E.B. Kruuk and J.M. Pemberton, 1998. Statistical confidence for likelihood-based paternity inference in natural populations. Mol. Ecol., 7: 639-655.
CrossRef  |  Direct Link  |  

28:  Meadows, J.R.S., E.K.F. Chan and J.W. Kijas, 2008. Linkage disequilibrium compared between five populations of domestic sheep. BMC Genet., Vol. 9,
CrossRef  |  Direct Link  |  

29:  Moiseeva, I.G., I.G. Chzhan, I.A. Zakharov and A.A. Nikiforov, 1996. Comparative analysis of morphological traits in mediterranean and Chinese chicken breeds. The problem of the origin of the domestic chicken. Genetika, 32: 1553-1561.
PubMed  |  

30:  Nei, M., 1972. Genetic distance between populations. Am. Naturalist, 106: 283-292.
Direct Link  |  

31:  Nei, M., 1987. Molecular Evolutionary Genetics. 1st Edn., Columbia University Press, New York, USA

32:  OMEGA Bio-Teck Corporation, 2009. Micro-elite genomic DNA kit. OMEGA Bio-Teck Corporation, British Columbia, Canada.

33:  Peter, C., M. Bruford, T. Perez, S. Dalamitra, G. Hewitt, G. Erhardt and ECONOGENE Consortium, 2007. Genetic diversity and subdivision of 57 European and Middle-Eastern sheep breeds. Anim. Genet., 38: 37-44.
CrossRef  |  PubMed  |  

34:  Promega Corporation, 1998. The Silver Sequence DNA sequencing system of Promega. Woods Hollow Road, Madison, USA.

35:  Romanov, M.N. and S. Weigend, 2001. Analysis of genetic relationships between various populations of domestic and jungle fowl using microsatellite markers. Poult. Sci., 80: 1057-1063.
CrossRef  |  Direct Link  |  

36:  Rosenberg, N.A., T. Burke, K. Elo, M.W. Feldman and P.J. Freidlin et al., 2001. Empirical evaluation of genetic clustering methods using multilocus genotypes from 20 chicken breeds. Genetics, 159: 699-713.
PubMed  |  Direct Link  |  

37:  Sambrook, J., E.F. Fritsch and T.A. Maniatis, 1989. Molecular Cloning: A Laboratory Manual. 2nd Edn., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, USA., ISBN-13: 9780879695774, Pages: 397
Direct Link  |  

38:  Soysal, M.I., E. Koban, E. Ozkan, V. Altunok, Z. Bulut, M. Nizamlioglu and I. Togan, 2005. Evolutionary relationship among three native, two crossbreed sheep breeds of Turkey: Preliminary results. Rev. Med. Vet., 156: 289-293.
Direct Link  |  

39:  Tamura, K., D. Peterson, N. Peterson, G. Stecher, M. Nei and S. Kumar, 2011. MEGA5: Molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance and maximum parsimony methods. Mol. Biol. Evol., 28: 2731-2739.
CrossRef  |  PubMed  |  Direct Link  |  

40:  Theodorou, K. and D. Couvet, 2003. Familial versus mass selection in small populations. Genet. Sel. Evol., 35: 425-444.
PubMed  |  Direct Link  |  

41:  Uzun, M., B. Gutierrez-Gil, J.J. Arranz, F. San Primitivo, M. Saatci, M. Kaya and Y. Bayon, 2006. Genetic relationships among Turkish sheep. Genet. Sel. Evol., 38: 513-524.
CrossRef  |  Direct Link  |  

42:  Weir, B.S., 1996. Genetic Data Analysis 2: Methods for Discrete Population Genetic Data. 2nd Edn., Sinauer Associates Inc., Sunderland, MA., USA., ISBN-13: 978-0878939022, Pages: 445

43:  Woolaston, R.R. and L.R. Piper, 1996. Selection of merino sheep for resistance to Haemonchus contortus: Genetic variation. J. Anim. Sci., 62: 451-460.
CrossRef  |  Direct Link  |  

44:  Zapata, C., C. Carollo and S. Rodriguez, 2001. Sampling variance and distribution of the D measure of overall gametic disequilibrium between multiallelic loci. Ann. Hum. Genet., 65: 395-406.
CrossRef  |  PubMed  |  Direct Link  |  

45:  Zhong, T., J. Han, J. Guo, Q. Zhao and B. Fu et al., 2010. Genetic diversity of Chinese indigenous sheep breeds inferred from microsatellite markers. Small Rumin. Res., 90: 88-94.

46:  Ismoyowati and D. Purwantini, 2011. Genetic Variability of bali and alabio ducks on the basis of phenotypic and microsatellites. Asian J. Poult. Sci., 5: 107-115.
CrossRef  |  

47:  Qanbari, S., R. Osfoori and M.P.E. Nasab, 2007. A preliminary study of marker data applicability in gene introgression program for afshari sheep breed. Biotechnology, 6: 513-519.
CrossRef  |  Direct Link  |  

48:  Quasem, J.M., A.S. Mazahreh and A.F. Al-Shawabkeh, 2009. Nutritive value of seven varieties of meat products (Sausage) produced in Jordan. Pak. J. Nutr., 8: 332-334.
CrossRef  |  Direct Link  |  

49:  Rane, P.P. and S.S. Barve, 2011. Standardization and optimization of mtDNA isolation and molecular genetic analysis of D-loop region in animal natural fibres. Int. J. Zool. Res., 7: 190-200.
CrossRef  |  Direct Link  |  

50:  Sulaiman, Y., C. Wu and C. Zhao, 2011. Phylogeny of 19 indigenous sheep populations in northwestern china inferred from mitochondrial DNA control region. Asian J. Anim. Vet. Adv., 6: 71-79.
CrossRef  |  Direct Link  |  

51:  Zahedi-Zahra, S. Esmaeelkhanian and R.V. Torshizi, 2007. Microsatellite variation in one breed of Iranian sheep with 12 markers. Pak. J. Biol. Sci., 10: 4455-4460.
CrossRef  |  PubMed  |  Direct Link  |  

©  2021 Science Alert. All Rights Reserved