Genomics in Animal Breeding
Genomic selection consists in using DNA profiles to help predict the genetic merit of animals, and select them on that basis. It can be used for any trait for which traditional genetic evaluations are already computed.
Genetic Erosion
We loose breeds all the time which take with them their unique genetic make-up. Losing these breeds is like losing a global insurance policy against future threats to food security.
It undermines capacity to adapt livestock populations to environmental changes, emerging diseases or changing consumer demands.
Genomic selection
According to Semex, a Canadian dairy bull breeding company and an international leader in bovine genetics (artificial insemination using genetic science) genomic selection consists in using DNA profiles to help predict the genetic merit of animals, and select them on that basis. It can be used for any trait for which traditional genetic evaluations are already computed.
The accuracy of genomic selection depends on many factors, especially the number and quality of the DNA markers used to genotype the animals, the number of proven bulls used to estimate the marker effects, and the estimation methods. The choice of validation bulls is also crucial in making sure that the accuracy of genomic evaluations is not over-predicted.
With genomic selection, information from the DNA profile of each animal is combined to any available performance data to create a new, more accurate genetic evaluation. The superiority of this evaluation is measured by its increased reliability. In order to measure the increase in reliability, a group of validation bulls are genotyped, their DNA profile is established, and the evaluations obtained from combining the DNA profile and parent average of each bull is compared to its proof. If the combined evaluation is more accurate in predicting the proof than the parent average alone, then the DNA profile is adding reliability and genomic selection is working. The higher the increase in reliability over parent average, the better the results.
Inbreeding
Genomics can help control inbreeding in several ways. First, the genotyping of animals with large marker platforms offers new approaches to study and monitor inbreeding. Second, genomic selection decreases reliance on pedigree information for selection. It makes it more likely to select animals from new lines or cow families that had not been used before. As a result, scientific studies have shown that it leads to a significant decrease in the rate of inbreeding per generation. On the other hand, the more rapid turn-over resulting from the use of unproven bulls as sires of sons or cows results in a shorter generation interval, so it will remain necessary to keep monitoring and controlling inbreeding, even when genomic selection is used.
Using unproven bulls
The use of DNA profiles allows us to better evaluate what traits a bull can pass on to his daughters even before he has any daughters. It therefore opens the door for the use of unproven bulls, which can be used at a much younger age. However, even with tens of thousands of markers, DNA profiles are still much less accurate than bull proofs. Younger bulls with high DNA profiles should therefore be used with caution.
The Genomic Parent Average (GPA) combines the parent average (PA) with the Direct Genomic Value (DGV) of each bull, offering a higher reliability compared to the PA alone. It reaches approximately 70 percent for fat and 42 percent for conformation, for values calculated in the USA. The very first Canadian results give genomic parent averages reaching 54 percent fat and 67 percent conformation. We are therefore implying a reliability level around 50- 65 percent according to various traits, with lower reliability for functional traits.
Progress
In a recent remarkable research in this field, University of Alberta scientists have successfully sequenced the genome of two influential bulls, one beef and one dairy, the first animals to have been fully sequenced in Canada.
Sequencing the genome of these two bulls enables scientists to more accurately identify the genetic markers that are responsible for economically important traits such as efficiency, yield, fatness and tenderness, say the scientists. Producers can use this information to breed healthier dairy cattle that produce more and higher quality milk as well as beef cattle that produce better quality beef.
“Consumers will benefit from more cost-effective and healthier products on store shelves,” said Stephen Moore, head of the Bovine Genomics Program at the U of A, who along with his colleague, Paul Stothard, completed the sequencing. “Understanding what genes contribute to specific cattle traits will also have spinoff applications related to other fields like human health and disease.”
Increased impact
The animals used in the project have had a high impact on the breeding and commercial sectors in dairy and beef. The dairy bull was sequenced in collaboration with Semex Alliance.
“This sequencing is significant to the dairy cattle industry because the bull’s genes are likely to make an important contribution to the genetic make-up of future generations,” said Dr. Jacques Chesnais, chief geneticist at Semex Alliance.
There are also benefits for the beef industry. Better knowledge of the genetic variation across the breeds will, through better breeding decisions, improve production efficiency, product quality and animal health, and reduce the environmental footprint of beef cattle production. The team sequenced the bulls’ genomes using a new technology – the SOLiD™ 3 System by ABI Life Technologies.
“SOLiD technology has allowed us to generate high density sequence information in a small number of runs. Combined with the high accuracy of the data, it has made the recent sequencing much more effective than was previously possible” said Prof. Moore.
Biosystems SOLiD 3
The sequencing was carried out using the new next generation sequencing.
The Applied Biosystems SOLiD 3 platform enables full genome sequencing to be undertaken at 1,000 times less cost and in a fraction of the time of the older approach that was used for many species including human, dog and cow.This is according to Prof. Moore.
The approach results in many millions of relatively short DNA sequences (50-100 bases) being generated. Due to the numbers of these it is possible with the right computing approach to reassemble these short reads into the entire 3 billion bases of DNA of the cattle genome.
The sequencing was done in the Agriculture Genomics and Proteomics Unit at the University of Alberta, while the analysis was a collaboration between scientists at Livestock Gentec at the U of A, Applied Biosystems Lifetec and a group called Penguin that uses Cloud Computing to handle large data sets such as we generated.
What we have now is a scan of polymorphisms (mutations) across the cattle genome. The current (old) technology has 50,000 of these spread more or less evenly across the cattle genome.
Future progress
A new platform with 850,000 polymorphisms will be available later this year. These platforms work very well in predicting the genetic potential of animals provided enough individuals have been genotyped to develop the prediction equations that translate the profile of polymorphisms into a breeding value. However they only tell part of the story, as there are millions of polymorphisms in the genome of cattle and they are not the only type of genetic variation that may affect animal performance. Neither do they get us further towards understanding the biology underlying the variations in, for example, performance or disease response.
It will also give us information on the results on the genome of many generations of selection that have been undertaken in cattle over many hundreds of years. For example how do dairy and beef cattle differ at the fundamental genomic level? This will enable much more accurate predictions on animal performance and health.
