Genetic Diversity and Population Structure of Selected Ethiopian Indigenous Cattle Breeds Using Microsatellite Markers

Background In Ethiopia, livestock contributes 45% of agricultural GDP. Despite the economic role played by the sector, there have been little efforts to genetically improve the indigenous cattle. Morphological characterization of selected Ethiopian indigenous cattle has been made for (Bonga, Jimma, and Kerayu) cattle types. But, the selected indigenous cattle were not characterized at molecular level (genetic diversity information). Hence, this work was initiated to detect and determine the genetic diversity and population structure of selected Ethiopian indigenous cattle ecotypes using microsatellite markers. Results Different alleles were identified (131) and thirty-three of these alleles were unique to specific ecotypes. All loci used were informative with PIC values ranging from 0.5 (TGLA126) to 0.84 (ETH10) with a mean of 0.70 per locus. The Shannon information index ranged from (I = 1.02) ILST006 to (I = 1.63) ETH10 with an average of 1.28 revealing there is genetic diversity. Moreover, analysis of molecular variance (AMOVA) revealed 84% genetic variation within a population and 13% variation among populations. The value of F-statistics (Fst) (0.129 = 13%) indicated that there was moderate genetic differentiation among ecotypes. The (UPGMA) revealed, Bonga and Jimma clustered together while Kerayu cattle were relatively distinct, Principal coordinates analysis (PCOA) and structure analysis grouped the individual into different clusters confirming the presence of ecotype admixture due to geographical origins and uncontrolled mating. Conclusion In general, this study has successfully characterized the genetic diversity and population structure of Bonga, Jimma, and Kerayu cattle ecotypes using high polymorphic/informative microsatellite markers. According to this study, Kerayu cattle have high AR and PA when compared to Bonga and Jimma cattle populations. So, the Kerayu population is more diverse than others and it is the hotspot for genetic diversity study. The generated information is very relevant for breeder and genetic conservation.


Introduction
In several parts of developing countries including Ethiopia, livestock production is the major and achieving good living standards. From Africa, Ethiopia is ranked frst and among the top 10 countries in the world in major farm animal populations [1].
Te livestock sector in our country has advantages for agricultural production, and domestic farm animals, in particular, are very signifcant, accounting for 30-40% of the food and agricultural sectors' economic value. In Ethiopia, livestock contributes 45 percent of agricultural GDP, 18 percent of overall GDP, and 19 percent of export earnings [2]. Tere are diverse numbers of indigenous cattle breeds in Ethiopia. According to the Ethiopian Biodiversity Institute [3] report, there are more than 28 native cattle breeds/ ecotypes that have been recognized to exist in the country. Cattle diversity in our country is due to the prevalence of both Bos taurus and Bos indicus cattle in the country. Ethiopia is also well known for its varied climatic and topographic environments. Tis has contributed to several distinct cattle breeds that have evolved [4].
Despite the global economy's commitment and diversity, there is a rapid depletion of farm animal genetic capital [5].
Among the factors, cross breeding, inbreeding, breed replacement, starvation, drought, and confict are the factors which afects genetic diversity of indigenous cattle [6].
As a result, understanding farm animal genetic diversity is needed to contribute to meeting current production needs in diferent settings, enable continued genetic development, or promote rapid adaptation to evolving environments (harsh environments), as well as to be used for breed selection based on their genetic contents for the dairy industry and beef industry [2].
Terefore, molecular characterization helps for efective conservation and long-term use strategies. Some phenotypic and genotypic characterization research has been done on Ethiopian indigenous cattle. However, so far identifcation and characterization of Ethiopian livestock resources are not exhaustive. Tere are, however, certain indigenous breeds that are known to exist at various levels of threat, interbreeding with other indigenous breeds and changes in production practices, for example, Sheko cattle. In addition, the cattle breeds Begait, Irob, Ogaden, Afar, and Borana are all threatened in some way [2,3]. As a result, the identifcation of indigenous cattle breeds, as well as their distinct characteristics, should be prioritized. Tat should be a prerequisite for developing conservation and sustainable utilization programs to assist indigenous breeds to compete in the future with limited production resources such as land, feed, labor, and capital, knowing the genetic diversity of Ethiopian indigenous cattle is for improvement and conservation purposes. As an example, particularly in terms of disease-tolerant cattle population in the country (for example, Sheko), cattle have been reported as tolerant to trypanosomosis. Not all ancestral Ethiopian cattle have been analyzed at the molecular level. Some attempts have been made in the country to classify and recognize cattle genetic resources. However, those attempts are not enough to give the actual cattle genetic diversity of the country's indigenous (local) cattle breeds [3].
Characterization attempts were mainly focused on farms, phenotypic features of genetic properties, and their products such as meat and milk, among other things [3]. Some molecular characterization of selected indigenous cattle breeds has been done, and it is not sufcient to say all Ethiopian indigenous cattle have been characterized at a molecular level. So far, the genetic diversity of Ethiopian indigenous cattle such as Horro, Guraghe, Arsi and Abigar, Zebu, Boran, Ambo, Adwa, Ogaden, Zebu Sanga, Fogera, Sanga, Raya-Azebo, Danakil, and African taurine Sheko have been analyzed using RAPD, microsatellite (SSR), and single nucleotide polymorphism (SNP) [6][7][8]. Even though cattle ecotypes like Kerayu are claimed for resistance to heat and tolerant to drought, they are not yet characterized. Similarly, the Bonga and Jimma ecotypes have variable coat colors, which help the ecotype to adapt to the very hostile environment and heat stress, and the Jimma ecotype is resistant to disease and high milk production when compared to both ecotypes, respectively. Morphological characterization has been done for Bonga, Jimma, and Kerayu cattle. However, molecular characterization has not been done for the three selected indigenous cattle. Terefore, the main aim of this study was to characterize the genetic variability of these three Ethiopian indigenous cattle ecotypes (Bonga, Kerayu, and Jimma) cattle using microsatellite markers.

Objective
(i) To identify the extent of genetic diversity among three indigenous cattle breeds (ii) To determine the population structure of the selected indigenous cattle breeds

Study
Area. Generally, the study was conducted in three areas that have diferent agroecological setups ( Figure S1). Te frst study area was Jimma Zone, Southwestern Ethiopia. Te area is situated approximately between 360 10′E longitude and 70 40′N latitude at an elevation ranging from 880 to 3360 meters above sea level. Te zone has an agroecological setting of highlands (15%), midlands (67%), and lowlands (18%) [9]. Te second study area was in the Fentale district located in the East Shoa zone of Oromia, the southern part of the northern Rift Valley of Ethiopia. Te area falls within an altitude range of 800-1100 masl. However, there are high peaks on the Fentale mountain from which the district derives its name, reaching up to 2007 masl [10]. It is found at a distance of about 200 km east of the capital city of Ethiopia, Addis Ababa, on the way to Harar. It is afected by recurrent droughts due to disrupted rainfall patterns. Te total land area of the district is 1170 km 2 . Te third study area was Southern Nations Nationalities and Peoples Regional State, Kafa administrative Zone of Chena woreda. It is situated in the southwestern part of Ethiopia (7°34′N latitude and 37°6′E Longitude) and with an altitude range of 1851-1900 mean above sea level.

Jimma-Type
Cattle. Zenga type (Zebu), mainly distributed in the Jimma zone with medium horns and big body frames. Daily milk yield per day is expected to be about 1.92 liters. Among the Jimma zone, the Dedo district is the highest in milk production and lactation length (3-8) months [11]. Tis may be due to management and genetic makeup used for draft power, milk, and meat. Coat color is red, black, and mixed white red and resistant to heat and disease.

Kerayu
Cattle. Sanga type breed and distributed in the Kerayu area of Eastern Shoa, mainly for milk, meat, saving, and dowry, with an average body weight of 300.4 kg and 249.9 kg male and female, respectively [9]. Tey are well adapted to the hot environmental situation with a straight profle, long thin legs, and long horns plain, patchy, and spotty.

Bonga Cattle.
Bonga cattle are a Zebu type found in the northern, western, and northwest parts of the Kafa zone. It originated from around Horro Guduru of Wollega and is used for draft power, milk production, and meat production. Te average lactation length of the Bonga cattle ecotype is about 8.5 months and the daily milk yield per cow is about 1.98 liters [10]. Te coat color of the Bonga cattle ecotype is red, black, light red, grayish, patchy, and spotted. Tey have downward, upward, mix, and forward horn orientations.

Blood Samples.
A total of 72 genotype were collected from (Bonga, Jimma, and Kerayu) and 24 from each.
Genotype from the same administrative zone was considered as one population with assumptions that they were more likely shared within zone than among zones through animal exchange for breeding.

Sampling Method.
Purposive sampling method (judgmental based or researcher based) was applied during breed selection and simple random sampling (lottery method) was used to select an individual animal. Te selection of administrative places (zones, districts, and kebeles) was conducted based on previous phenotypic characterization information of the cattle ecotypes. A list of animals that have the mentioned phenotypic characteristics was found from the selected kebeles. Tis list was used as a sampling frame for the study. Individual animals were selected using simple random sampling from the sampling frame in all the study areas [12,13].

Sample
Collection. Blood samples were collected from 24 unrelated animals of each cattle breed using 4 ml EDTAcoated vacutainer tubes. It is recommendable to study diversity within breed 20-30 range [12,13] samples. From the three study areas, about 72 blood samples were collected. Ten, after gently mixing, collected blood samples were placed in an Ice box and transported to National Agricultural Biotechnology Research Center, Holeta, Ethiopia, and stored at −200°C until DNA extraction.
3.6. DNA Extraction. DNA extraction from blood samples was conducted according to the standard salting-out protocol [14]. A 500 μl blood sample was transferred into a 2 ml Eppendorf tube; then 800 μl of lysis bufer was added to each tube (repeated until a white pellet formed).

Determination of DNA Concentration and Quality.
Te extracted genomic DNA concentration was checked by Nano drop (Nano Drop ® ND-8000). DNA quality was checked using gel electrophoresis by loading 5 μl sample DNA on a 1% agarose gel at 100v for one hour. Te gel was stained with gel red and visualized under UV light gel documentation system ( Figure S2).

Polymerase Chain Reaction (PCR).
A total of 16 bovinespecifc microsatellite markers were used for cattle genetic characterization (Table 1) [15]. Polymerase chain reaction (PCR) was performed by touch down method with two steps. Te 1st step was initial denaturation at 950C for 3 minutes. Ten, it was followed by 20 cycles of denaturation of 950C for 20 sec, annealing begins at 790C and ends at 52.40C for 45 sec, and extension at 720C for 1 minute. Te annealing temperature was decreased by 10C until it reached 52.40 C. At the second cycle, denaturation of 940C for 20sec, with 10 cycles, 52.40C for 45 sec, and 720C for 1 minute was applied. Te fnal extension 720C for 10 minutes was applied in all reactions. Te fnal volume of the reactions was 10 μl. Te polymerase chain reaction component was done in a total of 10 μl, which included 5 μl DreamTaq PCR master mix 2X, 10 μM forward primer (0.25 μl), 10 μM reverse primer (0.25 μl), 20 ng templates DNA (0.5 μl), and nucleasefree water (4 μl) and control reactions with no DNA template has been prepared to check for DNA contamination and primer dimers. At the end of the reaction, the PCR products were stored at +40C.

Gel Electrophoresis.
To assess amplifcation, 6 μl of the PCR product was loaded on 2% agarose gel prepared by dissolving 2 g of agarose in 100 ml 1XTAE bufer, staining with gel red. Electrophoresis was carried out at 80 V for 3 : 00 hrs. After completion of electrophoresis, the gel pictures were taken under UV Tran's illuminator by Biodoc analysis with a digital cannon camera (Figures 1 and 2).

Data Scoring and Statistical Analysis.
Te clear and visible amplifed bands of 72 genotype of selected Ethiopian indigenous cattle using microsatellite markers were scored using the PyElph 1.4 software [16].
Genetic variability was measured by estimating observed (Ho) and expected (He) heterozygosities [17]. Te Polymorphic Information Content (PIC), the unbiased Fstatistics [18], and the Analysis of Molecular Variance (AMOVA) were determined using the GenAlex software version 6.5 [19] and Power Marker. Pairwise FST (proportion of genetic variability due to population substructuring) values among pairs of populations were computed for all populations using GenAlex software version 6.5. Using the POPGEN version 1.31 software package [20] and the observed heterozygosity was done according to the algorithm of Levene [21]. Arlequin 3.5 was used to estimate basic frequencybased population genetic parameters such as gene diversity, the total number of alleles (No), Ne, breed private alleles (PAs), allele sizes, and allele ranges (in base pairs) [22]. Principal Coordinate Analysis (PCoA) was used to infer genetic similarities using the covariance matrix of Nei's genetic distance (DA) and unbiased genetic distances (DS) measurements. Dendrograms were created from pair-wise matrices of DA using the agglomerative hierarchical clustering unweighted pair group with arithmetic mean (UPGMA) method and DARwin vars to visualize evolutionary relationships among breeds. Te Dendro UPGMA online application [23] was used to design trees, which were then displayed in Tree View [24]. Bootstraps of 1000 replicates were used to establish Genetics Research confdence statements about the breed groups and to test the clusters' dependability. HP-Rare 1.1software were used to calculate the rarifed allelic richness (Ar) and private rarifed allelic richness (Arp) [25].
With independent allele frequencies and an admixture model (burn of 50000, followed by 100000 MCMC iterations with 10 replicate runs for each), the population structure analysis was carried out (1-10 K). Its most appropriate K value was identifed based on the computed K value, and k = 3 was discovered to be the most likely number of clusters to partition the 72 genotypes into three ( Figure 3) using the STRUCTURE harvester program [26]. Te CLUMPAK tool, developed by Kopelman et al. [27], was used to determine the best alignment from the STRUCTURE data, and the CLUMPAK result revealed genetic mixing and no clear geographic origin-based population structuring.  (Table 1).

Genetic Diversity of Population.
Comparatively the mean number of alleles observed in this study ranged from Na  (Table 3).

Analysis of Molecular Variance and Gene Flow.
Population variance could be classifed based on AMOVA among individuals within populations variability (84%) and (13%) variation among populations/among individuals and 3% within the individual. Te analysis also confrmed the presence of considerable gene fow (1.69) among subpopulations (Table 4).

Genetic Distance and Genetic Diferentiation between
Populations. Te genetic diferentiation between populations ranged from 0.100 between (Bonga and Jimma) to 0.120 (Kerayu and Bonga). Te highest GD (0.120) was between Kerayu and Bonga populations, and the lowest genetic diferentiation was observed between Jimma and Bonga populations (Table 5). Tis might be due to geographical locations and types of population/ecotype used. Te highest (0.46) genetic distance was observed between Kerayu and Bonga cattle breeds. However, the lowest (0.407) genetic distance was determined between Jimma and Bonga cattle breeds.

Cluster Analysis of Genotype in Tree Ethiopian
Indigenous Cattle. Te unweighted neighbor-joining cluster analysis (UPGMA) categorized the 72 genotypes into three major clusters (Cl-I, Cl-II, and Cl-III). From Figure 2, three types of clustering, dark-red, blue, and green color clusters showed Bonga cattle, Jimma, and Kerayu cattle ecotypes, respectively. Cl-I (36 percent), Cl-II (48.2 percent), and Cl-III (16 percent) of the overall population make up the three clusters, respectively. Te frst cluster, which contained 24 genotypes from all populations excluding Bonga and Jimma cattle genotypes, was the major cluster, whereas the second cluster contained 24 genotypes except the Bonga genotype. Te third cluster contains 12 genotypes. Te frst cluster consists of 33% genotype from Kerayu, 1.4% genotype from Bonga, and 1.4% genotype from Jimma cattle, respectively.
Cluster-II consists of 33% and 15.2% from Jimma and Bonga, respectively. Cluster three consists of only genotype from Bonga (16%). Genotypes from Bonga are mainly found in all three clusters, especially found in cluster-II (Jimma) (Figure 4). Figure 5 indicates that there is high gene fow between the two breeds (Bonga and Jimma). Population grouping was also carried out based on the UPGMA method to determine the relationship among the three selected indigenous cattle groups. According to the analysis, the populations are divided into two major clusters. Kerayu (I) is categorized under cluster one and both Jimma and Bonga (II) are categorized under the second cluster. Clustering patterns indicated that populations from geographically adjoining regions like Bonga and Jimma are subclustered together ( Figure 5).

Principal Coordinate Analysis (PCOA) and Population
Structure. Principal coordinate analysis was also used to look at the genetic relatedness of 72 genotypes (PCOA). Te pattern of genotype distributions on a two-dimensional plot showed separate clustering of populations based on the geographic locations and revealed a high pattern of grouping. Te PCoA analysis displayed in Figure 3 below confrms and was complementary to the result of the NJ cluster analysis shown in Figure 4.
With independent allele frequencies and an admixture model (burn of 50000, followed by 100000 MCMC iterations with 10 replicate runs for each), the population structure analysis was carried out (1-10 K). In the method (K � m| L″(K)|/s[L(K)] published by Evano et al. [28], the suitable number of clusters was discovered using K values that refected the proportion of change in the logarithmic probability Pr(X|K) of data between K values (28) (Figure 6). Its most appropriate K value was identifed based on the computed K value and k � 3 were discovered to be the most likely number of clusters to partition the 72 genotypes into three (Figure 7).

Genetic Diversity.
According to the classifcation of Botstein et al. [29], the highly informative markers have PIC values >0.50, the reasonably informative markers have PIC values between 0.25 and 0.50, and the slightly informative markers have PIC values <0. 25. In this study, all sixteen microsatellite loci used to profle genetic diversity of 72 genotype were found to be highly polymorphic with PIC >0.5. For evaluating genetic diferences between animal breeds, all 16 microsatellite markers exceeded the FAO's suggested minimum threshold of fve alleles per locus [13,15].Öner et al. [30] reported average mean PIC value was 0.87. Te average means PIC value (0.70) of the present study indicated the markers used in this study were highly informative and the availability of high allelic variation in the marker loci and their distribution within the population's genome. Te result of the present study was also used for genetic diversity analysis. In general, the PIC of this study (0.70) was higher than that of the previous one Df � Degree of freedom, SS, the sum of a square, Ms, means of a square, NM � gene fow.    reported by Gororo et al. [15] (0.664). Demir and Balcioglu [31] and also Hussain et al. [32] reported a PIC value of 0.82 using microsatellite markers, which was greater than the PIC value of this study. Te variation in PIC value could be due to a higher number of samples/breed types and an increased number of markers and implying the high discriminating ability of the markers. Gororo et al. [15] found 119 alleles in total, with an average of 7.4 alleles per locus and 34 private alleles. Hussain et al. [32] also reported a total of 476 alleles with a 22.33 average mean of alleles per locus using 21 microsatellite markers. Jakaria et al. [32] reported 46 alleles in four microsatellite markers with an average mean of 11.5 per locus. Other studies byÖzşensoy et al. [33] reported 269 alleles in 20 markers with a 13.45 average mean of alleles per locus.Öner et al. [30] reported a total of 545 alleles in 22 microsatellite markers with an average mean of 23.14 alleles per locus, and this observed number of alleles diference might be due to the number of genotypes, number of the marker, and breed number used.

Genetic Diversity Analysis along with Populations.
Previously, both Dadi et al. [8] and Gororo et al. [15] reported similar observed and expected heterozygosity across the populations. Agung et al. [34] also detected 0.66, 0.68, Ho and He, respectively. But, in this study, the level of observed and expected heterozygosity across the study populations was diferent, the Ho and He of this study were lower than the study reported by previous workers [8,35].
Demir and Balcioglu [31] reported that the Ho and He mean values were 0.63 and 0.74, respectively, due to the   [8] study and others. Tis diference might be due to factors like Null alleles, assortative mating, the Wahlund efect, selection against heterozygotes, inbreeding, or a combination of all of these reasons that can all explain this state [15,35].
In this study, the mean number of alleles for each breed (Bonga, Jimma, and Kerayu) was 4.94, 5.063, and 5.56, respectively. Demir and Balcioglu [31] reported that the mean number of alleles for four breed types such as Turkish Grey steppe (7.95), Eastern Anatolian Red (7.15), Anatolian Black (8.45), and HF (7.1). Agung et al. [34] also reported the mean number of alleles for 11 diferent cattle breeds was 6.28. Te variation of mean alleles in this study and the previous study might be due to the increased number of microsatellite markers and the number of breed types used for the study. Gororo et al. [15] reported a mean number of alleles (Na) of 5.16, which was similar to this study, microsatellite markers based on characterization of selected Ethiopian cattle like Bonga, Jimma, and Kerayu and Na detected were 5.18.
Te other measure of gene diversity is the Shannon information index (I), if Shannon's information index value is close to one or above, it indicates that there is variation in the tested populations and that the markers are suitable for studying diversity [36]. Te value obtained in this study ranged from I � 0.86 to I � 1.63 with an average mean of 1.28. Tis implies that selected Ethiopian indigenous cattle have genetic diversity.
According to Zerabruk et al. [35], allelic richness reported in north Ethiopia of cattle ranged from 5.67 to 6.27 with a mean of 6.23 AR per breed. But, in this study, AR ranged from 3.88 to 5.56 with a mean of 4.5 AR per breed. Tis showed that there is genetic diversity among populations.
Agung et al. [34] reported a 3.81 efective number of alleles across 11 diferent breeds using 12 microsatellite markers; however, in this study, 3.23 efective numbers of alleles were detected using 16 microsatellite markers from 3 diferent Ethiopian indigenous cattle breeds. Hussain et al. [32] reported 6.7 efective numbers of alleles. Tis showed that the types of breeds and the number of breeds used might be the cause for the variation in an efective number of alleles.
To quantify the genetic variability, in Ethiopian indigenous cattle breeds and 16 microsatellite markers were used resulting in a Fis value (0.966), showing that the selected Ethiopian indigenous cattle have undergone inbreeding. Jakaria et al. [37] reported a Fis value of 0.07 using fve microsatellite markers, and it was lower than the Fis value reported in this study. FST values could indicate small (0-0.05), medium (0.05-0.15), high (0.15-0.25), and very high (FST >0.25) genetic diferentiation between breeds [38].
Te estimated FST value for 3 diferent cattle breeds was higher than Zimbabwean cattle breeds (0.084) [15]; this fnding was comparable with those reported by Sharma et al. [39], 13.3% of FST value in Indian cattle using STR markers and high FST value of this study indicated that 16 microsatellite markers used for 3 breeds/ecotypes were signifcantly high and are useful indicators of markers that could be powerful tools for genetic diferentiation of diferent breeds. But, lower than Pakistan cattle breeds (0.1456) Rahal et al. [40], and Indonesian cattle breeds (0.243) [39]. Jakaria et al. [37] also revealed a higher number of FST values (0.246) than this study. El-Sayed et al. [41,42] detected a higher FST (0.236) value from two Egyptian cattle breeds. Tis FST variation might be due to gene fow and exchange of breeding animals. Te FST is higher when the populations are isolated between them.

Analysis of Molecular
Variance. In addition, AMOVA shown in genetic variation among breeds was 13%, and 84% within the population, 3% within an individual was observed. Signifcant genetic diferentiation was observed among all 3 diferent Ethiopian indigenous cattle (Bonga, Jimma, and Kerayu) estimated by FST � 0.129. Dadi et al. [8] reported 1.3% genetic variation among populations and 98.7% genetic variation within populations. Jakaria et al. [37] also reported genetic variation within a population (70.6%) was higher than among populations (29.4%). Te genetic variation within the population was higher than among populations this implies that there might be due to interpopulation gene fow, sexual recombination, and mutations.

Cluster Analysis, PCOA, and Population Structure.
In clustering, a dendrogram of cluster analysis based on NJ algorism using UPGMA categorized the three ecotypes into three clusters based on the geographical locations (I, II, and III) with diferent subgroups. Dadi et al. [8] obtained two main clusters using 30 microsatellite markers with 10 indigenous, one HF, and diferent subgroups formed under two main subgroups. Cervini et al. [43] showed two diferent clusters of dendrogram-based UPGMA/NJ using 12 microsatellite markers in 11 diferent cattle breeds. A study by Edea et al. [6] also revealed two main clusters in six Ethiopian indigenous and one Korean cattle breed using SNP markers.
Te diference might be due to the number of microsatellites/type of microsatellite used, some ecotypes/breeds, and sample numbers. Te clustering model showed that there was a relationship between the patterns of genetic diversity and the geographical origins of the collection. Populations collected from Bonga and Jimma had a strong relationship. Bonga and Jimma cattle were close to each other and the existence of gene fow among the neighboring populations seems possible.
Moreover, this result is also clearly refected in population structure showing weak admixture of a gene across populations. It revealed the existence of substructuring (K = 3) in three populations of Ethiopia. Previously, Jakaria et al. [37] also reported population structure with k = 3 in four cattle populations and there were genetic admixtures. Tis could be likely due to the presence of gene fow between the ecotypes because of the movement of cattle and uncontrolled mating/exchange of breeding animals, migration from one region to another. Grouping of PCoA corresponds with the clustering dendrogram based, which showed conformity result obtained from UPGMA analysis Agung et al. [44].

Conclusion
In this study, the genetic diversity and population structure of the selected Ethiopian indigenous cattle were covered using highly polymorphic microsatellite markers. Te study confrmed the presence of genetic diversity in selected Ethiopian indigenous cattle breeds, indicating the possibility of improvement through breeding and the importance of maintaining diversity by applying appropriate conservation strategies. Tis study also indicated genetic variation (84%) accounted for within populations, likely due to gene fow, sexual recombination, and mutation. Te populations showed moderate genetic diferentiation, due to high gene fow. Te highest genetic diversity indices were recorded for Kerayu cattle populations, suggesting that this area is a hotspot for genetic diversity studies, sources of important alleles for breeding purposes, and conservation strategies must be applied.
In general, cluster analysis, PCoA, and population structure analysis exhibited moderate grouping of samples. Studying the genetic basis of locally adapted indigenous cattle populations is critical for developing appropriate breeding strategies and programs aimed at improving and conserving their genetic diversity. From this study, the generated information is valuable for the national animal breeding program and conservation purposes.