Skip to main content

Select-a-bull: Inside Ireland’s cattle-breeding database

October 8, 2018

The Irish Cattle Breeding Federation created a central database where all the key players in the cattle industry can search relevant information related to each animal. 

The following is an edited transcript of Tom Martin's interview with Ryan Martin (no relation) of the Irish Cattle Breeding Federation. Click below to hear the full interview:

Tom:                I'm talking with Martin Ryan, technical support manager at Glanbia Agribusiness in County Tipperary, Ireland. Mr. Ryan is an award-winning pedigree cattle breeder and exhibitor, a past president of the Irish Charolais Cattle Society and is a current board member of the Irish Cattle Breeding Federation. He joins us to talk about what can be accomplished when various interests work together toward a common goal. We thank you for joining us, Martin.


Martin:            Thank you very much.


Tom:                Tell us about the centralized database that's been created by the Irish Cattle Breeding Federation. What's it called?


 Martin:           ICBF is the short acronym we use for it. It's the Irish Cattle Breeding Federation. I think it goes back to 1998 when a lot of small-breed societies were all doing various genetic improvement programs and trying to do their own thing. We had a vision to create our centralized database, which would bring economies of scale, first of all, and actually improve the reliability by getting everybody to work together. And we envisaged bringing together not just the breeders, but also the processors, the livestock markets, the processing factories — right through the chain — artificial insemination (AI) organizations, milk recording, you name it. We wanted to get them all involved.


Tom:                So, this actually began in the 90s?


Martin:            In 1998, the idea took hold, and by the year 2000, we had a federation put together and created, basically, a co-op where all the interested parties took a shareholding, and that’s how they are represented today.


Tom:                And, since then, there's been quite a revolution in digital technology — must have changed things considerably.


Martin:            Absolutely, but fundamentally, I think at the start, we had at least five decades of good registration information from birth. Every calf was tagged at birth. We had quite a lot of their ancestry information, so that formed the basis of the database. What we didn’t have was the processes for pulling together all the issues like milk recording, linear scoring, weight records from animals being slaughtered, confirmation scores, fat scores, daily live weight gain — all that good information. So that's really what we set out to achieve: put all that information together in one place.


Tom:                Is that what makes the system unique, or are there more attributes that do that?


Martin:            I think that was how it set itself apart from everything else, beginning from day one. So, every time an animal moves in Ireland, it is captured in the database. If it goes to a livestock market, its weight on that day is transferred back to the database, along with the price it brought. If it's an animal that's slaughtered in a processing plant, on that day, its carcass weight, its carcass confirmation and its fat score are all transferred back into the database. Of course, we have the sires of those animals as well. So, all that information — or, indeed, on the dairy side, if it’s milk recording — all that comes back into the database, which gives us a very simple model. We have the data coming from everywhere, then you have a lot of on-farm recording programs as well, which transfers back-data into the database. It gives us a powerful tool for creating genetic evaluations.


Tom:                And how many users are there?


Martin:            We have at least 70,000 people that are working in the system between dairy and beef farmers. You could say about 80 percent of all active farmers are very involved in the database. Everybody gets their evaluations from there. We have, on breed society level, six dairy breeds and fourteen beef breeds completely involved. Every processor is involved — every milk processor, as well as beef processors. Every livestock market is involved. So, it's a very complete, comprehensive system that captures data at every level.


Tom:                There can be competitive considerations to sharing data, but I understand there's a great deal of collaboration in this system. Is competitiveness an issue?


Martin:            Oh, absolutely. It’s not the case today, but I think at the very outset there would have been a lot of concern by individual breed societies that they were putting information into a central repository and wondered, "How safe is that? Can I get it back and have it become my own again? What responsibilities do I have, and are we safe to share that information? Are we allowed to share it?”


                        So, there was a lot of concern initially, and that's why it probably took two years to become incorporated and get to where we are now. But, today, people don’t see any alternative.


Tom:                A great deal of data comes in, and then it’s analyzed and reshared back with the users?


Martin:            Yes, the system is completely open, so any individual farmer can go on and look at his own herd. He'll get reports every week, if he wishes. Every week, the database shares all the registration data and AI information that came in, so you can see how many cows are milk-recorded or how many cows were inseminated during the week, for example. So, it's completely open in that regard. You can check any animal.


                        If you're a farmer looking for stock bull, you can go into the bull search, which has a lot of parameters. You can set your parameters, even including the distance from where you live, as well as calving years, et cetera, and select a bull accordingly.


Tom:                Is any training required to use the system?


Martin:            We provide a lot of webinars both to the breed societies and to individual farmers for using the system, but it's very intuitive because people understand the terminology quite well.


Tom:                “Data reliability” is a term that might seem obvious, but what is it and why is it important?


Martin:            Well, at the very outset, I think data reliability would have been considered relatively poor because you didn’t have a lot of information. To improve that data reliability, I suppose, the first thing is we had to increase the usage. As soon as more and more people submitted data, that completed it.


                        We also introduced the system among beef farmers whereby there is some benefit to putting in the data themselves. Out of the 50,000 beef producers, over 50 percent of them — and particularly the larger ones — joined up immediately. That gave us a big volume of data going into the system to improve that reliability. Then, along came things like genomics, where we're sampling up to 35,000 cattle every week genomically, and that has improved the reliability substantially as well. On the AI side, we have over 50 billion SNPs (single nucleotide polymorphisms) in the system, which gives us enormous reliability.


Tom:                Is this a global model?


Martin:            I think it's a global model. I suppose every country doesn’t necessarily work the same. It's highly unique to have every piece of the industry sharing information together. That sharing has benefited dairy farmers to the tune of €150 million per year. It's about a hundred and seventy million dollars in today's language. In a small country, it's nothing to sneeze at.


                        On the beef side, the average beef carcass has gained at least €70 per head, so that's quite significant. On the replacement index side, things have been quite a bit slower because people focus initially on the slaughter part, but today, that is increasing as well. In the last year alone, [it] has increased by about €15 per animal.


Tom:                Is the system continuing to be developed, and what is yet to be achieved?


Martin:            I think the main part that will give us further momentum in the future is the genomics component. We use 54K SNP, which is a very high level of intensity of data. From that, we can pick up quite a lot of other issues in the genome, including genetic defects and so on. I think, down the road, we'll get to gene editing, ultimately.


Tom:                Martin Ryan is technical support manager at Glanbia Agribusiness. We thank you so much for being with us.


Martin:            Thank you very much. Pleasure.




Ryan Martin spoke at ONE: The Alltech Ideas Conference. Don't miss another game-changing idea from the beef industry. Join the world's thought-leaders and experts at ONE19. Learn more here