Friday, August 26, 2016

ASA Fall Focus: Taking Technology Home to the Farm and Ranch

Here are videos from the American Simmental Association's Facebook livestream of my presentation titled "Taking Technology Home to the Farm and Ranch."




Seedstock Producers as Educators

Decker's Rants

EPDs and Environment 


Saturday, August 20, 2016

ASA Fall Focus: Information Learned from the IGS Genomics and Genetic Evaluation

Dorian Garrick
Iowa State University

Garrick started working on animal models applied to sheep and goats in 1982 (the year I was born ☺).

In the old system, each breed has their own data silo. This is combined together to have a joint pedigree and performance data.

Genomics has changed this.

EPDs are determined by the collective action of many genes. Selection increases the frequency of favorable gene effects and decreases the frequency of unfavorable gene effects. This allows producers to breed better cattle year after year. Genomics allows us to increase the accuracy of genetic prediction, especially for young animals with little or no data.

In human medicine, researchers are looking for individual DNA variants that are predictive of a person's risk for developing a disease. In beef cattle genomics, we don't use this conservative approach; we use all of the DNA variants simultaneously. Using all of the DNA variants gives much better predictions.

The hoped outcome is data to information to knowledge. The genetic prediction methods really can be a black box for producers.

Right now, the major problem is merging the data from the multiple breed associations and doing so in a repeatable and efficient manner.

The Irish Cattle Breeding Federation has negotiated a deal that they can genotype an animal for $20, because they bought 1,000,000 SNP assays at once (a bulk deal). United States breed associations need to consider more stringent DNA testing requirements, such as genotyping all parents, which would lead to bulk deals.

We can continue to improve predictions by:

  • Better marker panels - fewer better features used
  • More animals genotyped
  • More phenotypes collected (particularly for carcass, reproduction, and disease)
  • Improved quality control of all data
  • Better models and analytical methods

The purpose of collecting pedigree, performance and genomic data is to make better selection decisions.
The information systems used to input, store, and analyze data need ongoing development.

ASA Fall Focus: BOLT

Bruce Golden
Theta Solutions LLC

Historically, Simmental has been one of the leaders in the development of genetic prediction.

There has been evolution of statistical models used to predict genetic merit (EPDs). Each time the EPDs got better and better. To predict EPDs, you do two things; first you build the problem on the computer then you solve the problem.

What drove the evolution of the methods used to predict genetic merit?
Knowledge of the model? All of these models were well known by 1970.
New methods? Maybe a little.
Data? Yes, there has been the creation of genomic data and more phenotypic records.

But, the main driver has been to improve the accuracy of prediction. Striving to reduce the prediction error variance. Try to make sure that we are making better solutions and increasing the rate of genetic gain.

Improvements in computing power has also helped in the development of genetic predictions.

Another change is going to be better DNA markers that are closer to the genes and causal variants.

Computer gaming required really fast and affordable computer processors. Theta Solutions is using these improvements in computing to increase the speed of genetic prediction.

Theta Solutions is taking the data, new hardware, and new computer programs to do genetic analysis of large genomic data.

"It is not enough to take the old software and run it on this new gaming hardware. It requires rethinking and rewriting the software." Golden said.

Not only are they trying to compute EPDs more quickly, is implementing models that are more complex and specific.

Why would one use a Bayesian sampler for mixed models?

  • No accuracy approximation bias (better way to calculate EPD accuracy)
  • Can get the prediction error covariance (Instead of just comparing a bull's accuracy to zero, you can compare the accuracy differences between bulls or between groups of bulls. This will also allow a highly reliable accuracy measure for economic selection indexes.)
  • Marker selection methods
  • Prior knowledge
The IGS analysis feels they have:
  • Superior Marker Effects Model
  • Superior accuracy computation
  • New stayability approach - Random regressions
  • New breed effects model (using USDA MARC differences and year trends)
  • Carcass traits are being solved together with birth weight (birth weight is included to acount for a selected subset of the animals having carcass data)
Also has a method for including external EPD

New stayability model can now differentiate between unknown or missing data versus data saying a cow did not calve.

Theta Solutions is going to start working on final production acceptance testing in two weeks.

ASA Fall Focus: Application of Genomic Technology to Optimize Herd Replacement and Produce Elite Breeding Stock

Mahdi Saatchi
Lead Genomicist
International Genetic Solutions

Imagine a sire who is heterozygous (one A variant and one B variant) for a DNA position. At that same position a dam  is also heterozygous.
If we consider two progeny of this pair of sire and dam, they can be 0% related to 100% related at this position.

Calf 1 Calf 2 Relationship
A/A A/A 100%
A/A A/B 50%
A/A B/B 0%
A/B A/A 50%
A/B A/B 100%
A/B B/B 50%
B/B A/A 0%
B/B A/B 50%
B/B B/B 100%

If we apply this to the entire genome, we expect full siblings to share 50% of their DNA. But, just as the relationships can vary at a single locus, the relationships can vary for the entire genome. In chicken data, researchers see that the relationship between siblings ranges from 0.2 to 0.7.

Fig. 2 from Lourenco et al. 2015
By more precisely measuring the relationship between animals, genomics allows us to more precisely predict an animal's genetic merit.

Genomics allows us to improve several parts of the key equation for genetic change. Genomics allows us to have more accurate selection decisions, increase the selection intensity and decrease the generation interval.

Genomic predictions have previously been shown to be accurate for Simmental cattle (Saatchi et al. 2012).

Genomic predictions are reliably predicting yearling weight. Genomic predictions are explaining real differences in yearling weight. Based only on the genomic prediction (molecular breeding value), there is a 100 pound difference between animals in the top 25% and bottom 25% of animals based on the genomic prediction.

Genomic information can be used for more than just producing genomic-enhanced EPDs. Saatchi points out that strings of DNA variants (called haplotypes) are sometimes never observed in two copies in an animal. If these haplotypes are never seen it two copies it likely means that they carry a variant that is responsible for the loss of pregnancies.

ASA Fall Focus: Nuts and Bolts of Animal Breeding

Wade Shafer

What is the science of animal breeding? Shafer cited Wikipedia, saying, "The scientific theory of animal breeding incorporates population genetics, quantitative genetics, statistics, and recently molecular genomics and is based on the pioneering work of Sewall Wright, Jay Lush, and Charles Henderson."
He highlighted the work of Sewell Wright, Jay Lush, and Henderson, and Lenoy Hazel. Not only is animal breeding about the genetic value of animals, it is also about the economic value of those animals.

Animal breeding is where the rubber meets the road. "It is one of the most practical sciences."

Simmental has the slogan of "Visual analysis tells you what an animal appears to be, his pedigree tells you what he should be, his performance and progeny tells you what he actually is."

In 1971, Vaniman used Paul Miller, a dairy genetics at ABS, to produce the first sire summary using Boeing Airlines computers.  The foreword said that sire summaries would revolutionize beef breeding.

In 1982, ASA signed a long term contract with John Pollack and Dick Quass at Cornell University to deliver and advance genetic evaluation. This was the largest BLUP evaluation at the time. Best Linear Unbiased Prediction, BLUP, was developed in the 1950s and 1960s, but wasn't able to be deployed at a large scale till the 1980s when computer technology caught up. The contract with Cornell was exclusive, mean Cornell only did genetic evaluation for Simmental.

Simmental has also benefited from a close relationship with Montana State University and the cattle industry in the state of Montana.

Shafer pointed out stalwarts in the breed. He pointed out gathers such as Jerry Lipsey and Ropp. Steve McGuire has been the shepherd of the data.

Glimpse of the future, through the past. Shafer highlighted the career development of Dorian Garrick at Cornell. Red Angus is a partner of the Simmental Association. Red Angus worked with Colorado State University, with Bourdan and Brinks leading the way. They had a productive graduate student, especially with computers, named Bruce Golden.

In 1997, the Simmental Association and Cornell University teamed up to produce the first multi-breed genetic evaluation. Originally, this was deveoped to do a better job of predicting cattle, such as half bloods, that were being used in the grading up process.

In 2001, the National Beef Cattle Evaluation Consortium was federally funded through an earmark to support genetic evaluation.

In 2007, six breeds tried to join together to start a joint genetic prediction. This never materialized. In 2010, the Red Angus Association of America and the ASA created a joint genetic prediction. This ultimately lead to the creation of International Genetic Solutions. "An unprecedented collaboration between progressive breed associationss to enhance beef industry profitability." IGS now contains 12 breed associations, with 17 million animals, with 340,000 new records each year.

Theta Solutions is a company created by Dorian Garrick and Bruce Golden to advance the technology of genetic prediction. Theta Solutions software is called BOLT for Biometric Open Language Tools.

Lauren Hyde and Jackie Atkins

What are our selection decisions? How many bulls should I use? Should I use old bulls or young bulls? Which replacement females should I keep?

The speed of genetic change (hopefully progress) is
Directly proportional to:

  • Accuracy of selection
  • Selection intensity
  • Genetic variation

Inversely related to:

This is called the key equation. 

Accuracy of selection is the strength of the relationship between the ture breeding values and their predictions for the trait under selection. Accuracy increases when we use EPDs from BLUP. If using phenotypes, the accuracy is based on the heritability of the trait (which is much lower that using EPDs). 

If the BIF accuracy is 0.1, the change in weaning weight is 8 pounds. If the BIF accuracy is 0.25, the improvement in weaning weight is 12 pounds. 

Selection intensity is how choosy we are when we are selecting animals. Are we keeping everything or selecting at random or are we choosing the very top animals? Selection intensity is the difference between the average and the selected parents, divided by the variability (standard deviation) in the trait.

It is hard to change amount of genetic variation in a herd.

Generation interval is the average age of the parents when the progeny are born, In cattle this tends to vary between 4 to 6 years.

Accuracy vs. generation interval
Decrease in generation interval causes decrease in accuracy and vice versa.
Quick turnover of herd sires mean we have fewer progeny per sire and less accurate predictions.

Selection intensity vs risk
Selection risk is that the true genetic merit of replacements is significantly worse than expected. With fewer sires we increase the intensity but increase our risk. With more sires we decrease our risk but we also decrease our selection intensity.

Male selection can be more important than female selection. We can be more choosy when selecting bulls and we have more progeny per bull.

You are not supposed to do single trait selection. But, selecting on an index is not single trait selection. ASA recommends that producers use indexes to select for increased profit.

Wednesday, August 17, 2016

17th Annual Missouri Livestock Symposium December 2nd and 3rd

The Missouri Livestock Symposium committee is currently planning their 17th annual Symposium in Kirksville, MO for December 2nd and 3rd. The Missouri Livestock Symposium (MLS) committee works year-round to find the best speakers on timely topics that benefit producers in their respective enterprises.
The MLS began in 1998 with a simple conversation between then University of Missouri Extension Livestock Specialist Bruce Lane (retired) and local Adair County livestock producer and current MLS chair, Garry Mathes. Since those humble beginnings the MLS has grown exponentially and recently had attendees covering a majority of Missouri’s 114 counties plus 16 states, with over 2,000 in attendance.
Attendees have an opportunity to attend the largest agricultural-based trade show in the Midwest, featuring many local, state, and national agricultural businesses. Comments regularly heard include, “my favorite show,” “something for everyone in the field of agriculture,” and “we were made to feel very welcome and were well taken care of.”
The Symposium kicks off Friday afternoon December 2nd at the William Mathew Middle School. Visitors can peruse the sold out trade show before the free beef meal at 6 pm. The evening program begins in the auditorium at 7 pm when the MLS committee will recognize the 2016 Agriculture Educator of the Year and the Livestock Person of the Year. The Friday evening program concludes with a keynote speaker. This year Dr. Scott Brown, University of Missouri extension agricultural economist, will address current domestic and global farm and ranch financial pressures.
The Saturday program is unique and is what has branded the MLS for the last 16 years. Instead of focusing on one particular species of livestock, the MLS holds programs on beef cattle, horses, sheep, meat goats, forages, stock dogs, estate planning and farm management; as well as timely nutrition/food topics, saving the honeybees and backyard poultry production. The MLS features eight different educational tracks and contracts over 30 speakers from across the nation representing universities, governmental agencies, and private industry. Speakers address the timeliest information for producers to take home and improve their operations. A Governor’s Style Luncheon is provided on Saturday and is free to attendees. The meal is sponsored by Missouri’s commodity groups.
Highlights of speakers and topics for 2016 include Richard Winters, Winter’s Horsemanship and Kim Lindsey, AQHA ranching director in the horse section. The beef section will focus on the impending effects of the Veterinary Feed Directive (VFD), highlight some of the latest advances in beef cattle genomics, and feature a panel discussing capturing value when marketing your calf crop.
Additional highlights include Dr. Michael Neary, Purdue University, featured in the sheep and stock dog sections and Dr. Ron Hanson, University of Nebraska, conducting a 3-session estate planning course. Other topic highpoints include land prices, beef economy outlook and climate trends affecting agriculture.
Since the beginning, the MLS remains deeply connected to University of Missouri Extension and their mission of improving the lives of citizens through education. MLS chair, Garry Mathes and current Livestock Specialist Zac Erwin believe this will be the best program to date. Anyone seeking further information on the MLS are encouraged to visit our website at, Like us on Facebook, call the Adair County Extension Center at 660-665-9866 or MLS chair, Garry Mathes at 660-341-6625.

Decker's Take Home Message
I will be speaking at the meeting in Kirksville, and hope to see you there!