Dr. Ole Meland - Vice President of Genetics, Accelerated Genetics
Genomics has burst onto the scene in recent months. I was asked to shed some light on what it is and how it is likely
to affect A.I. and dairy cattle breeding. In general,I started by sending out an email to many of my colleagues and to
producers around the country asking them to let me know what questions they have or what questions producers have had as
it relates to what affect genomics will have on dairy cattle breeding. The response was great! I received 82 questions
from 18 sources. Obviously, many had similar questions.
Let me start by saying I am confident that genomics will prove to be the greatest techonological advancement in the dairy industry in the last 40 years. As encouraging as it is now, it is just the beginning of where this technology
can take us in the future. So, let's get started.
What is Genomics?
In 1988 Genetic Visions began utilizing marker assisted selection to help choose between full brothers, to determine which young sires should enter the progeny test program. This new technology that we are calling Genomics is substantially more powerful than what has been utilized in the last 20 years.
With the completion of the bovine gene mapping, researchers in the Bovine Functional Genomics Lab (BFLG) at USDA, in conjunction with a few other research labs both domestic and foreign, were able to estimate the SNP effects at some 50,000 loci. DNA from over 5200 Holstein bulls from the CDDR (Cooperative Dairy DNA Repository, semen was donated by 7 AI organizations in the US and Canada), were used to estimate the SNP effects. ((SNP is a Single Nucleotide Polymorphisms. The SNP’s that USDA is using are limited to a single basepair substitution or deletion. The change in the DNA structure is inherited or transmitted to offspring. Because they can detect the change it becomes a genetic marker. SNP’s are one type of genetic marker. Markers provide a means to track the inheritance of a trait. Many of these substitutions or deletions have no effect on proteins produced or function as they are not in the genes themselves but in the same region of the chromosome. They do however serve as makers in the DNA that can be useful to associate them with economically important traits)).
After BFLG estimated the SNP affects of each of the 50,000 loci in the Holstein breed they are able to use this information to adjust the parent average by adding or subtracting the SNP affect at each loci and coming up with a genomic estimate for each of 26 traits or composites such as NM$, milk, fat, protein, SCS, calving ease, PL, DPR, most of the linear traits. These genomic estimates can be made at the time of birth of the animal.
The addition of the genomic estimates increases the reliability of the PTA’s from around 35 % to around 70% (varies by trait). This increase in reliability is equivalent to parent average plus about 11 daughters in 11 herds.
Are other Countries doing genomic research?
Yes, I understand New Zealand and France are using the same 50K SNP chip from Illumina as the US is using. Holland is using a different chip. Germany should have genomic estimates with in 12 months. The US project is very extensive with a data set of over 5200 Holstein bulls used in the project and many more bulls and cows have been tested to date. The strength of the genomic estimates, increase as the number of progeny tested animals are genomic tested.
Have any high reliable/second crop bulls been genomic tested and if so what affect did genomics have on the PTA estimates of these high reliable/second crop bulls?
Yes, bulls born before 1998 (predictor bulls) were used to predict the PTA’s of bulls born between 1998 and 2002 (predictee bulls). As you would hope would be the case, high reliable second crop bulls had very little change in PTA’s (+/- 15 NM$ was what I generally observed for this group of bulls).
What about first crop bulls, have they been genomic tested and if so what affect did genomics have on this group of bulls?
All Accelerated Genetics active Holstein bulls will have genomic estimates by August 20, 2008 and All Jersey bulls will be genomic tested by October 6, 2008. Some AI studs have tested most if not all their bulls, other AI studs have done very few of their active bulls.
The lower the reliability of the PTA estimates the more change you would expect. Most progeny tested bulls do not have a significant change in either direction, but a few bulls do. The genomic PTA’s will be official with the January 2009 proofs and will replace the traditional PTA’s for those bulls at that time. Until January 2009 genomic estimates are not to be used in advertizing or promotional materials.
Will all bulls be genomic tested by January 2009 and if not what will be used for their PTA’s?
Many bulls will be genomic tested by January 2009, but not all of them will be. If an animal has been genomic tested their genomic PTA’s will be substituted for the traditional PTA’s. If a young bull has been genomic tested the reliability of his PTA’s will be much higher, in the range of 70% vs 35% for non genomic tested bulls. Bulls with first crop genomic PTA estimates will likely on the average have 5-7% higher reliability. Young bulls will have to be enrolled in the NAAB cross reference program or be at least 18 months of age before his genomic PTA’s are official and appear on his pedigree. Genomic PTA’s will be official for all females in January 2009 and will appear on their pedigrees.
What is the expected change with genomic tested animals?
The average expected change is zero, but there will be considerable change up or down on any given individual bull. The change with genomic tested bulls will be less than the change seen with non genomic tested bulls. The reliability estimates for genomic tested bulls are currently around 70% depending on the trait and that is equivalent to parent average plus 11 daughters in 11 herds. As more bulls and cows are tested we expect the reliability of genomic tested young bulls to increase. Some suggest they may get as high as 85-90%.
How should I use genomic tested sires?
As of January 2009 the genomic PTA’s will be published on pedigrees of all females and on bulls that have been enrolled in the NAAB cross reference program or are at least 18 months of age. These genomic PTA’s will be official and can be used in advertisements. The reliability’s will be in the range of 70% rather than the 35% range, but these genomic PTA’s can and will change.
One needs to remember the principals we have always suggested when using young sires. Use a group of bulls and not over use any one bull.
What affect will genomic testing have on inbreeding?
Genomic testing has the potential to increase inbreeding but given the fact that AI companies can now screen for high genetic merit bulls with low inbreeding values, more outcross/low inbreeding bulls will be brought into the program. This very well could keep inbreeding in check and could even help to reduce it.
Will genomics replace progeny testing?
In time if genomic tested bulls reach the 85-90% reliability at birth this would greatly reduce or eliminate the need for progeny test. However, at the current reliability range of 70%, there is still a need to progeny test bulls to verify the genomic PTA estimates. The one thing that will be different is that the level of genetics going into progeny test herds will be substantially higher. The bottom 40-50% of young sires that would have entered progeny test programs before genomic testing will never be put through the program.
What affect will genomics have on the way AI companies buy their bulls?
No doubt genomics will affect the way AI companies source their genetics. Each organization will develop their own system. Some have suggested they will buy fewer bulls since they expect to have a higher graduation rate. Unlikely 3-4-5 full brothers will enter AI. An AI company will likely work with many more cows from more sources than ever before. It is likely more “commercial herds” can be a great source of young bulls for AI. The price per bull may well increase since the reliability of what is being bought has increased, but currently the reliability for genomic PTA’s is only 70% and that still leaves a lot of room for re-ranking once a bull has several hundred progeny with performance records.
How will genomic testing affect the cost of semen?
Clearly the semen from high ranking genomic tested young bulls is more valuable than the $3.00- $5.00 that most AI companies have charged for young sires in the past and the extremely high ranking young bulls will be in high demand.
On the flip side the genetic improvement producers will get from these young bulls as a group will pay big dividends through genetically superior offspring for not only production and type but health traits, lower insentience of calving ease and better fertility.
Should I test my better cows and heifers?
The cost to do the 50K SNP test will be in the $250.00 range. There is a lower density chip being tested (384 SNP chip). When this is available commercially they expect the cost will be in the $40.00-$50.00 range. Since these 384 SNP’s are spaced out across the genome the hope is that the accuracy will be relatively good compared to the 50K SNP chip.
At a cost of $250.00 it might only be cost effective to do some of your more elite females, to decide which ones to invest the money in flushing. When the lower cost, low density chip is available commercially (384 SNP chip for around $40.00-$50.00), it will be more cost effective to screen a higher proportion of your cows and heifers and then have the 50K SNP chip run on the females that come out the best on the 384 SNP test.
Who can test bulls?
Only the 7 AI companies that contributed DNA to the CDDR can test bulls for five years (March 2008- March 2013). The DNA in the CDDR was critical to the completion of this research project; the project could not have been done without it.
Anyone can take advantage of the results of this research project to test females by September 15, 2008.
Where can I get females tested? What do I need to do to get females tested? How long will it take and how much does it cost? Do you use blood or hair? Will data appear on official pedigrees?
Any Holstein breeder anywhere in the world can get their females tested. Holstein USA told me they should be ready to accept applications for genomic testing by around September 15, 2008. The 3 approved labs in the US will take either blood or hair samples (the 3 approved labs in the US are Geneseek, Genetic Visions and GIVF). The cost for the 50K SNP chip will be in the $250.00 range. USDA is generating the genomic PTA’s from the genotypic data they get from these 3 labs on a every other month schedule. My guess is that USDA will run the genomic PTA’s more often as demand increases. January 2009 these genomic PTA’s will be official and printed on pedigrees of all females tested. Bulls will be official and printed on pedigrees in January 2009 provided they are entered in the NAAB cross reference program or when they turn 18 months of age.
Genomic testing of Jersey animals will start October 2008 and the rules for being official will be the same as for the Holstein breed. The reliabilities of genomic PTA’s for the Jersey breed are not expected to be as high as Holstein due to the fact there are not as many Jersey bulls available for the research.
Other breeds may not have genomic estimates due to the lower number of bulls to do the research. The Brown Swiss breed is working on trying to combine data for several countries around the world, so they may well get enough accuracy by doing that.
This technology will continue to grow and improve over the next few years. It will be fun and exciting to watch as it is fully implemented into the national and international breeding programs, not only at the AI level but ultimately at the producer level.