Skip to main content
Premium Trial:

Request an Annual Quote

International Research Team Uses Illumina Chips to ID Sheep Breeding Markers, Map Genetic Diversity


By Justin Petrone

A team of international researchers has produced an "unprecedented" map of genetic diversity in sheep, tracing both the breeding history of the livestock species and identifying regions in the genome that have changed in response to selection for genes controlling desired traits such as coat color, body size, reproductive qualities, and the presence or lack of horns.

Using Illumina OvineSNP50 BeadChips, the International Sheep Genomics Consortium genotyped nearly 3,000 sheep from 74 different breeds. The results of the study were published in a paper in PLoS Biology earlier this month. The authors believe the results will inform the adoption of genomic selection by industry, and will provide more insight for others in the research community.

"SNP chips have propelled genome-wide association studies and genomic selection in sheep in a way that we really couldn't have envisioned five or 10 years ago," said James Kijas, lead author of the new paper and a principal research scientist at the Commonwealth Scientific and Industrial Research Organization, Australia's national science agency.

Kijas said the study was funded by participating institutions. In addition to CSIRO, participants included Utrecht University in the Netherlands; the Department of Primary Industries Victoria in Melbourne, Australia; the French Institute for Agricultural Research; Illumina; Embrapa, the Brazilian Agricultural Research Corporation; the University of Bari in Italy; the University of Sydney; and AgResearch New Zealand.

Kijas said that the group had two main findings. The first was the generally high level of genetic diversity present across sheep breeds, which was "much higher than we see in dairy cattle and a lot higher than commercial dog breeds," Kijas told BioArray News. "This finding tells us things about genetic history and the process of domesticating sheep," he said.

Based on the study results, the authors concluded that there "must have been a large and diverse population that was recruited 10,000 to 12,000 years ago to make domestic sheep," said Kijas. This has allowed sheep to maintain diversity, meaning the "genetic bottleneck which took place was not as severe during the development of sheep as for some other animal domesticates," according to the paper.

The other "important" finding was the identification of regions of the genome that appeared to have evolved more quickly over time in response to breeding efforts. By performing a genome-wide scan for the signatures of selection, the group identified 31 genomic regions that contain genes for coat pigmentation, skeletal morphology, body size, growth, and reproduction, according to the paper.

"We found a gene that causes the presence or absence of horns," said Kijas. "We also saw genes related to body morphology and reproduction."

The overall results provide "unprecedented insight into the phylogeographic structure of sheep populations and the results of centuries of breeding practices," the authors wrote.

Study Design

Using the Illumina OvineSNP50 BeadChip, the investigators analyzed samples from 2,819 animals from each continent, including six breeds each from Africa and North and South America, seven from the Middle East, eight from Asia, and the rest from Europe.

The content for the OvineSNP50 was derived from three separate sequencing experiments performed using Roche 454 and Illumina Genome Analyzer instruments, as well as Sanger sequencing. From the content generated by these experiments, 49,034 SNPs were included on the chip. Illumina launched the OvineSNP50 in 2009 (BAN 1/27/2009).

Using the arrays, the investigators were able to determine relatedness between animals and find evidence for high levels of admixture among sheep breeds.

A number of aspects of the SNP diversity indicated that high levels of gene flow have occurred between sheep populations following domestication, according to the paper. These included a high degree of conservation in linkage disequilibrium phase and haplotype sharing across short chromosomal distances. The authors did not detect a strong association between genetic diversity and physical distance from the domestication center.

The authors were able to identify genomic regions relevant in selection. The highest selection signal was detected on Chromosome 10. The highest ranked SNP was located at megabase position 29.54 near the relaxin/insulin-like family peptide receptor 2, which has been linked with the absence of horns in sheep.

Genomic Selection

The results of the study will inform genomic selection of sheep for future breeding, Kijas said, adding that some industrial entities have already adopted arrays for such a purpose.

"From a wider perspective, the SNP50 has been remarkably successful and it is currently driving lots of activities in the sheep research industries and communities," Kijas said. Genomic selection has been "adopted enthusiastically in the dairy breeding sector and it is just … at the point of adoption in the sheep industry in Australia and New Zealand," he said.

One issue in the adoption of arrays for genomic selection in sheep versus cattle is the cost of the animal. As it costs less to raise sheep, and because they mature faster than cattle, breeders have less of an imperative to adopt the new technology, whereas in the cattle industry, arrays have been widely adopted given the high cost and amount of time to raise cattle to select based on phenotype.

Donagh Berry, a quantitative geneticist at Teagasc, the Irish Agriculture and Food Development Authority, told BioArray News recently that farmers "cannot spend a lot of money" on array-based genomic selection for sheep (BAN 2/21/2012).

"For an average farmer, the price of a lamb might be between €70 and €100," said Berry. "So he cannot spend €30 of that to genotype an animal," he said. "Therefore, it is a very different equation with the sheep because of the cost of the animal relative to genomic selection."

CSIRO's Kijas agreed that the unit costs of sheep and cattle are different.

"If you are a producer and you want to buy a ram to bring into your enterprise, obviously that is a lower cost than if you want to go and buy an elite dairy bull or a beef cattle bull," Kijas said. Still, he said that the price of genotyping sheep should come down as the technology becomes more widely adopted.

"The volume of SNP50 genotyping done in the sheep industry is lower than what is done in cattle," said Kijas. "We don't have the volume to drive down the cost of the technology and so we wind up paying up more for it."

According to Kijas, the International Sheep Genomics Consortium is planning follow-on studies to identify markers associated with traits that are hard to measure. "If we can make genomic predictions about those and reduce the phenotyping cost, then the payoff could be big," said Kijas.

He said that the consortium is moving to genotyping by whole-genome sequencing to evaluate genetic diversity in greater depth. It is also considering whether to proceed with the development of a high-density SNP chip of about a million markers for such studies. According to Kijas, the consortium is in discussions with Affymetrix and Illumina at the moment about whether there will be enough demand to necessitate development of such a chip.

Illumina and Affy have commercialized high-density SNP genotyping arrays for bovine research (BAN 1/12/2010).

"Whole-genome sequencing has its advantages, but even at the price of the existing technology, we cannot go out and sequence thousands of sheep," said Kijas. "There is a case for a high-density sheep SNP chip, if for $100 or less one could [look at] a million SNPs across the genome," he added.

Have topics you'd like to see covered inBioArray News? Contact the editor at jpetrone [at] genomeweb [.]com