Limits...
Extent of Linkage Disequilibrium and Effective Population Size in Four South African Sanga Cattle Breeds.

Makina SO, Taylor JF, van Marle-Köster E, Muchadeyi FC, Makgahlela ML, MacNeil MD, Maiwashe A - Front Genet (2015)

Bottom Line: We found that moderate LD extends up to inter-marker distances of 40-60 kb in Angus (0.21) and Holstein (0.21) and up to 100 kb in Afrikaner (0.20).Analysis of effective population size based on the extent of LD, revealed Ne = 95 (Nguni), Ne = 87 (Drakensberger), Ne = 77 (Bonsmara), and Ne = 41 (Afrikaner).Results of this study form the basis for implementation of genomic selection programs in the Sanga breeds of South Africa.

View Article: PubMed Central - PubMed

Affiliation: Agricultural Research Council-Animal Production Institute Pretoria, South Africa ; Department of Animal and Wildlife Sciences, University of Pretoria Pretoria, South Africa.

ABSTRACT
Knowledge on the extent of linkage disequilibrium (LD) in livestock populations is essential to determine the minimum distance between markers required for effective coverage when conducting genome-wide association studies (GWAS). This study evaluated the extent of LD, persistence of allelic phase and effective population size (Ne) for four Sanga cattle breeds in South Africa including the Afrikaner (n = 44), Nguni (n = 54), Drakensberger (n = 47), and Bonsmara breeds (n = 46), using Angus (n = 31) and Holstein (n = 29) as reference populations. We found that moderate LD extends up to inter-marker distances of 40-60 kb in Angus (0.21) and Holstein (0.21) and up to 100 kb in Afrikaner (0.20). This suggests that genomic selection and association studies performed within these breeds using an average inter-marker r (2)≥ 0.20 would require about 30,000-50,000 SNPs. However, r (2)≥ 0.20 extended only up to 10-20 kb in the Nguni and Drakensberger and 20-40 kb in the Bonsmara indicating that 75,000 to 150,000 SNPs would be necessary for GWAS in these breeds. Correlation between alleles at contiguous loci indicated that phase was not strongly preserved between breeds. This suggests the need for breed-specific reference populations in which a much greater density of markers should be scored to identify breed specific haplotypes which may then be imputed into multi-breed commercial populations. Analysis of effective population size based on the extent of LD, revealed Ne = 95 (Nguni), Ne = 87 (Drakensberger), Ne = 77 (Bonsmara), and Ne = 41 (Afrikaner). Results of this study form the basis for implementation of genomic selection programs in the Sanga breeds of South Africa.

No MeSH data available.


Related in: MedlinePlus

Correlation of allele phase by physical distance represented as generations in the past.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4664654&req=5

Figure 5: Correlation of allele phase by physical distance represented as generations in the past.

Mentions: The statistic r was used to examine the extent of persistence of marker allele phase in the studied breeds. For example, if r was positive in one population and negative in another population, then different haplotypes are prevalent in the two populations and phase is not preserved between them. However, if r is positive in both populations, then similar haplotypes are common in both populations and phase tends to be preserved. A correlation of 1 between r values indicates that the marker phase persists across populations while a correlation of zero indicates that the marker phase in population one cannot predict phase in the other population (de Roos et al., 2008). In this study, the correlation between alleles at contiguous loci indicated that phase was not strongly preserved between breeds and decreased with increasing genetic distance between the breeds (Figure 5). The largest correlation was observed between the Nguni and Bonsmara pair (0.50 at 10 kb) whilst the lowest correlation was observed between Afrikaner and Drakensberger (0.29 at 10 kb). As expected, the correlations between r statistics for all B. taurus-African breed pairs were lower than those found among the South African breeds (Figure 5) with the lowest correlation observed between Afrikaner and Angus for marker pairs separated by less than 10 kb. This was supported by the estimated time of divergence of these breeds that suggested that South African cattle breeds diverged from each other approximately 131 (Nguni vs. Afrikaner) to 192 (Afrikaner vs. Drakensberger) generations ago (Supplementary Material 3). On the other hand B. taurus breeds diverged from the South African cattle breeds approximately 245 (Bonsmara vs. Angus) to 884 generations ago (Nguni and Holstein). Figure 6 shows the neighbor-joining tree based on time since breed divergence and also a neighbor-joining tree based on the Fst-values reported by Makina et al. (2014).


Extent of Linkage Disequilibrium and Effective Population Size in Four South African Sanga Cattle Breeds.

Makina SO, Taylor JF, van Marle-Köster E, Muchadeyi FC, Makgahlela ML, MacNeil MD, Maiwashe A - Front Genet (2015)

Correlation of allele phase by physical distance represented as generations in the past.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4664654&req=5

Figure 5: Correlation of allele phase by physical distance represented as generations in the past.
Mentions: The statistic r was used to examine the extent of persistence of marker allele phase in the studied breeds. For example, if r was positive in one population and negative in another population, then different haplotypes are prevalent in the two populations and phase is not preserved between them. However, if r is positive in both populations, then similar haplotypes are common in both populations and phase tends to be preserved. A correlation of 1 between r values indicates that the marker phase persists across populations while a correlation of zero indicates that the marker phase in population one cannot predict phase in the other population (de Roos et al., 2008). In this study, the correlation between alleles at contiguous loci indicated that phase was not strongly preserved between breeds and decreased with increasing genetic distance between the breeds (Figure 5). The largest correlation was observed between the Nguni and Bonsmara pair (0.50 at 10 kb) whilst the lowest correlation was observed between Afrikaner and Drakensberger (0.29 at 10 kb). As expected, the correlations between r statistics for all B. taurus-African breed pairs were lower than those found among the South African breeds (Figure 5) with the lowest correlation observed between Afrikaner and Angus for marker pairs separated by less than 10 kb. This was supported by the estimated time of divergence of these breeds that suggested that South African cattle breeds diverged from each other approximately 131 (Nguni vs. Afrikaner) to 192 (Afrikaner vs. Drakensberger) generations ago (Supplementary Material 3). On the other hand B. taurus breeds diverged from the South African cattle breeds approximately 245 (Bonsmara vs. Angus) to 884 generations ago (Nguni and Holstein). Figure 6 shows the neighbor-joining tree based on time since breed divergence and also a neighbor-joining tree based on the Fst-values reported by Makina et al. (2014).

Bottom Line: We found that moderate LD extends up to inter-marker distances of 40-60 kb in Angus (0.21) and Holstein (0.21) and up to 100 kb in Afrikaner (0.20).Analysis of effective population size based on the extent of LD, revealed Ne = 95 (Nguni), Ne = 87 (Drakensberger), Ne = 77 (Bonsmara), and Ne = 41 (Afrikaner).Results of this study form the basis for implementation of genomic selection programs in the Sanga breeds of South Africa.

View Article: PubMed Central - PubMed

Affiliation: Agricultural Research Council-Animal Production Institute Pretoria, South Africa ; Department of Animal and Wildlife Sciences, University of Pretoria Pretoria, South Africa.

ABSTRACT
Knowledge on the extent of linkage disequilibrium (LD) in livestock populations is essential to determine the minimum distance between markers required for effective coverage when conducting genome-wide association studies (GWAS). This study evaluated the extent of LD, persistence of allelic phase and effective population size (Ne) for four Sanga cattle breeds in South Africa including the Afrikaner (n = 44), Nguni (n = 54), Drakensberger (n = 47), and Bonsmara breeds (n = 46), using Angus (n = 31) and Holstein (n = 29) as reference populations. We found that moderate LD extends up to inter-marker distances of 40-60 kb in Angus (0.21) and Holstein (0.21) and up to 100 kb in Afrikaner (0.20). This suggests that genomic selection and association studies performed within these breeds using an average inter-marker r (2)≥ 0.20 would require about 30,000-50,000 SNPs. However, r (2)≥ 0.20 extended only up to 10-20 kb in the Nguni and Drakensberger and 20-40 kb in the Bonsmara indicating that 75,000 to 150,000 SNPs would be necessary for GWAS in these breeds. Correlation between alleles at contiguous loci indicated that phase was not strongly preserved between breeds. This suggests the need for breed-specific reference populations in which a much greater density of markers should be scored to identify breed specific haplotypes which may then be imputed into multi-breed commercial populations. Analysis of effective population size based on the extent of LD, revealed Ne = 95 (Nguni), Ne = 87 (Drakensberger), Ne = 77 (Bonsmara), and Ne = 41 (Afrikaner). Results of this study form the basis for implementation of genomic selection programs in the Sanga breeds of South Africa.

No MeSH data available.


Related in: MedlinePlus