Latest / The Source – The Jorgensen Land & Cattle Podcast / Episode 8: Genomics 101 with Lane Giess
Transcript
- speaker-0: Mm-hmm. Over a hundred years of experience lie in ideal South Dakota. The team at Jorgensen Lanning Cattle is going to take the time to offer insight into that experience along with many other industry experts. If you're ready to learn and grow, this is The Source. Hello folks. Welcome to another episode of The Source. ⁓ fun episode for me today because I have a good friend of mine with me ⁓ that I've gotten to know over the last year. ⁓ fellow K-State graduate, which Obviously we all we all gather together, yes sir. gonna introduce y'all to Lane Geis. Lane is with the genetics team at Zoetis and he works with the team at Jorgensen Lan and Cattle on our genomic ⁓ IBE evaluation. And ⁓ I wanted to get Lane on because Lane's a a wealth of knowledge about how we get to genomic enhanced EPDs. ⁓ We won't dig into how we're using them today or differences in tests, but I actually wanted to give you guys a chance to understand how we get to GEPDs. So, Lane, first of all, ⁓ from Minnesota originally, seed stock operation. ⁓ go ahead and tell the folks a little bit about yourself, where you grew up at, and the the part of the cattle industry that that you're involved in from a production standpoint today. speaker-1: Yeah, sure. Thanks, Jamie. ⁓ so like you say, I grew up in central Minnesota. ⁓ my folks have been in the cattle business all their lives. and so we I think it was the late nineteen or excuse me, early nineteen eighties they started ⁓ Dar Lynn Cattle Company. So my mom is Lynn and my dad is Dar. And they ⁓ They were originally in the Charlet business and had some Angus for a period of time and then were really strong in the Solitaire business, but finally settled on South Devon in the late nineteen eighties. ⁓ and so from there, ⁓ have kind of built a a a offering of South Devon genetics, ⁓ which I know is a a small breed population, but works for us. And ⁓ so I think this last year we're over I think we had our thirty thirty third or thirty fourth annual production sale. and so yeah, I'm still directly involved in all that. I still help with all the mating decisions. Even though I don't live in Minnesota anymore, I try to make it back as often as I can and still help out with the sale as much as I can. But yeah, it's it's a very it's a small family owned ⁓ cow calf seed stock operation. We don't have any outside help for the most part. ⁓ and just sell about a hundred bowls a year. So that's that's kind of my background. Yeah. speaker-0: You bet, and then then tell ⁓ about about school and and how you got to to genomics and ⁓ landed at Zoetis. Yeah you have a as I as do I, you have ⁓ a laundry list of what you've done to get to where we're we're sitting at today. speaker-1: Yeah. Kinda funny how that goes. You know, all the small decisions you make when you're young kinda and ends up puts you on a path that gets you to very fun places. So I went to Kansas State for my bachelors in twenty eleven and I was I I primarily went because I wanted to go be on the livestock judging team at K State. I wanted to I wanted to be a a livestock judger and do all those things and while I was there I had some really good mentors and I I got involved with the meat stojin team, the livestock team, you know, eval team, and I think through those teams had a really good experience with the faculty that were in charge and and all of them kind of encouraged me to consider grad school. ⁓ as I mean all throughout that entire experience. They were all saying, You should really try to consider grad school. I think there's a lot of opportunity there and I thought for the longest time I wanted to be a nutritionist, ⁓ until I learned ⁓ you know, there's there's some nuances there, but ⁓ Doctor Bob Weber had a New project. He just came to K-State a couple years prior and was working on a project for feet and like soundness ⁓ that was funded by the American Symatol Association and the Red Angus Association. And so he brought ⁓ because I had livestock judgment experience, he brought Brady Jensen and myself onto this project to help develop the sort of under basic understanding of genetics. around feet and legs and and soundness and Try to help build the recommendations for the industry to like what are the ch what are the soundness traits that we care about, which ones are related, how are they related to production traits? ⁓ is there ⁓ genetic control for soundness? All those questions we we had to ask. So ⁓ Brady and I spent a better course of two years traveling around to many, many seed stock operations ⁓ across I think nine states, and got to learn a lot from the those those ⁓ from those producers but also learned a lot about data collection, learned a lot about best practices for sea stock producers. And so I think given my kind of upbringing, right, with ⁓ raising cattle and selling bulls and then transitioning into the data collection space, it I started to really gain an appreciation for the work that it takes to build genetic evaluations ⁓ and ⁓ learning the science behind it. So had's ⁓ and I I you know to your point, like K-State had some really fantastic mentors and and faculty members that ⁓ you know that they cared about getting you getting you through school, but they also cared about finding your interest and and and helping guide you in a a really positive direction, right? So like I I I I I think looking back, you know, Dave Nichols and ⁓ Weber and Terry Hauser, like those those those three probably did more for me to get me to where I am today than than I I think most would recognize. So yeah it ⁓ Key State was a p phenomenal experience w working through that whole process. speaker-0: Yeah, and I'll I'll add Casey Olson to that list because he was the one that pushed on me, you know, the Jorgens and Lanic cattle were jackrabbits, but ⁓ cut me open, I bleed purple. That's my background. And then off to CSU after you got done with the red Angus. Now there was a break in there, correct? There was a break speaker-1: Yeah. Yeah, so I I was coaching the meat start team at K State or I excuse me, I coached one year of the meat starting team at K State and wanted to do an assistantship with Terry Hauser ⁓ doing genetics of of like meat science. So we're gonna we're gonna try to look at some of the genetic control of ⁓ Beef on dairy, ⁓ this was back in twenty seventeen. So right as kind of that was starting to really take off in the industry. ⁓ but then he he moved his positions ⁓ at K State so the assistantship would have been more ⁓ extension meet science and that's not something I wanted to do. So it kind of forced me to go I shouldn't say forced, but it it pushed me to go look for jobs outside of Just to go look for jobs, right? And so I went to National Cadoman's ⁓ convention in Phoenix, I believe. and ran into Wade Schaefer and Chip Kemp with the Cimitol Association and they had a they had a job basically that they helped put together to look at how can we grow commercial cow calf producers utilization and understanding of EPDs and genetic predictions and genetic evaluations. And so started straight out of my master's program with the American Cemitol Association and it it I think the project Right, the Fiend Lakes Project where we we went out and visited a bunch of Cymatol producers and I had worked with the association staff. It was a very seamless fit. And I, you know, I think it was a that that role really helped develop me. So for I spent two years between ⁓ between my master's at K-State before going back for my PhD. ⁓ and those two years I you know got to meet a lot of SeaSock producers, members of the Cymatol Association, ⁓ had ⁓ and that role I I was able to kind of carve out some additional ⁓ responsibilities and that was r during a time where a lot of these new traits were starting to kinda come on the scene. Like hair shed was one, right? ⁓ teat nutters, while not a new trait, it was new for data collection and and so the association needed somebody to help members who had historically been, you know, just focusing on capturing ⁓ growth weight measurements, ⁓ carcass ultrasound measurements, ⁓ and and Maybe grabbing some DNA if you're a progressive producer. So helping guide those producers into here's some new traits that are gonna come ⁓ that we wanna build evaluations for, here's some best practices for how to capture those new traits. And just trying to help members better understand this kind of new era of genetic evaluation. ⁓ ten years ago now. So so I spent two years with the Cemetole Association and then had the desire to go back and get a PhD because I wanted to contribute more to the IGS evaluation and the science and and had approached Wade and the team and asked if I could grab a research project or a new trait that they wanted to develop for IGS, take that off of the RD shelf and make that my dissertation project, and went to the folks at CSU who had already had a relationship helping develop ⁓ the PAP evaluation for IGS. and see if they would take me on as a grad student. And so it kinda worked out really well where CSU got a free grad student, IGS got a basically a a project taken off the shelf that they didn't have to worry about. And I got a PhD out of it. And now that I'm looking back, I think I probably got the short end of the straw on that one. speaker-0: Yeah, you probably did. You probably did. speaker-1: Yeah. So that so yeah, started my PhD in twenty twenty in the middle of COVID. speaker-0: Yep. And ⁓ chase heifer preg, correct, that was the the PhD dissertation. Yep. speaker-1: Yep. The the project was to develop a basically an understanding of the multi breed, trying to b essentially build the heifer pregnancy evaluation for IGS. So the factors that we had to account for that other evaluations didn't was better understanding heterosis, ⁓ better under and categorizing breed differences across all these various breeds within IGS and trying to investigate what are what's the best model, how do we establish contemporary group effects, how do we ⁓ account for you know certain fixed effects like age at first calving or age of first exposure, like all those different kinds of things. So it was it was from the beginning to the end just trying to better wrap our arms around how do we build a heifer pregnancy evaluation for this multi-breed, primarily pure ⁓ excuse me, primarily seed stock ⁓ oriented production systems. So speaker-0: And then you've been with Zoetis right about a year? speaker-1: Yep, so my PhD went ⁓ I finished in twenty twenty four, ⁓ fall of twenty twenty four. And so I've been with Zoetis just over a year now. Yep, so I finished that PhD and had an opportunity to come work with ⁓ Kent Anderson and the team here at Zoetis. So it was it was a a very nice transition for sure. speaker-0: You bet. And so now you get to work with ⁓ now you get to work with the Jorgensen crew on our on a fairly routine basis and we're a pain in the neck because we like it that way. No. ⁓ so at Jorgenson's we talk about the ideal beef evaluation and ideal beef genetics a lot. ⁓ it's really one of the ⁓ Really at this point we have two anchor points. We have the mother load project, which is, you know, the the philosophy behind how we breed and the goal of making great females. And then the other anchor point to our entire program is the genomics and ⁓ understanding the predictors that that we're gonna use to make better cattle that will in turn make our commercial c customers better cattle to make them more profitable moving forward. But we really haven't ever talked about how we get to those predictors, Lane. So the first thing I want to start with is let's do DNA testing in genomics one ⁓ one for for somebody that ⁓ maybe brushed across genetics in college and that was the extent of it. So how do we get from an ear punch to these numbers that we look at? speaker-1: That's a that's a big question. speaker-0: It's a deep it's a deep topic. Take it any way you want. Yeah. Like I said w one-on-one here and we we just we want to give everybody kind of the the basics of how we get there. speaker-1: Well and I think, you know, i to I think to answer that question, maybe just take a quick step back, right? And and you know, say like so for the longest time, right, genetic evaluations were primarily limited to seed stock operations and to breed associations, right? Because those breed associations were the ones that were developing those genetic evaluations and they were fueled by seed stock data, right? So, you know, I think I think how we get to genomics, just to tell a quick just to kind of get to that point, right? Genetic evaluations primarily rely on performance data and pedigree relationships. And in a registry system, the pedigree piece is easy because we have so many connections and and we know the siren dam of every animal, and we and as seasoned producers, we go out of our way to to better understand all those those relationships. And then speaker-0: Yeah, these type of things. And I guess to stop you there for just a second, Lane. We go back to previous before genomics, we still had EPDs, but they were calculated strictly off of pedigree and performance. And it was a guess between dad does this, mom does this. This is what we should expect out of these offspring. speaker-1: Right. Right, the curse of the parental average. Like it's it's one of those things where, you know, if you know ⁓ Sire's performance ability and dam's performance ability, you expect their progeny to be directly in the middle of that, right? But we know that that's not the the case because they might inherit all of the good ⁓ alleles from Sire and all the good alleles from Dam and then they'll exceed their parents, right? And so that's the that's our goal of genetic evaluation is to find those animals that do better than their parents and and call the ones that do worse than That's that's our whole objective, right? and so ⁓ And you said something there that I think is is kind of key to to all of this, right? So the the h previously the C stock sort of producers, they would market bulls for ⁓ and and for the most part, commercial cow cow cow producers, they care a lot about revenue trades. So live calf percentage and weaning weight, right? Yep. So those are those are the things that they would select on. And they would go to their sea stock provider and they would try to say which which are the Cabernese bulls or which are the high wing wing bulls, right? And and so they would go to and out of a out of so what you described is a P Blub evaluation, pedigree blob. So blob stands for best linear unbiased prediction. Basically what it means is it's sophisticated statistical methodology for how we d arrive at genetic predictions for individual animals. And a P Blub is just using pedigree relationships and performance data. And that's it. And for young animals that have no progeny, those predictions were historically f pretty lowly accurate, right? And They're probably better than than nothing. and they're probably be still better than a ratio or an actual weight, right? But th there was still enough error there that some commercial producers would buy a Cavanese bull and they would have a wreck, right? Or, you know, it would turn out that this this bull got all the bad genes from Zire and Dan and so he you know, even though the evaluation thinks he's the Cavanese bull, he really wasn't. And so I think that's really where There you know, that that kind of challenge that the Seasock industry had was We we can't arrive at more accurate evaluations on these young animals unless we change something. And so that's where genomics I think comes into the into the scene because ⁓ late two thousands ⁓ is really when Gene Starr kinda came to to the to the industry where ⁓ there was we now had an opportunity to grab a DNA sample on an animal. evaluate its its genes and and and look at its DNA and say, is this animal desirable for cavities or not? And can we augment that with these other predictions to to better understand genetic potential in these sires? And really what it became was, you know, one thing that we learned in this business was Seaside producers really didn't want to do because it I DNA is expensive, right? But they didn't really want to do a whole lot of DNA testing until the commercial producers who bought their bowls asked for it. Until the commercial producers said, I'm not gonna buy a bowl, ⁓ I'm not gonna spend a lot of money, I'm not gonna spend X amount of dollars and and have a sizable investment in a Cavanese bowl when he just turns around and and creates a wreck. Like that that's not something I want to do. And so if I can use if I can if if you DNA test these bulls and we can reduce that risk of these of these cider that I'm buying, then that fundamentally changed the business to where more and more SeaSock producers are DNA testing. And and in my mind it's a I I think it's the new standard, right? I I think if you're gonna be a SeaSock producer you need to be DNA testing your bowls that you sell. speaker-0: Yeah. And I mean we're we're chasing a rabbit now, but I'll I'll take that path just a little further. You know, ten, twelve years ago when I was in school and getting out of school, we were talking about genomics. You know, talking two thousand twelve to two thousand fifteen. You know, we were talking about it in in class and everybody was kinda going genomics, huh? Yeah. You know, and and it really wasn't it was starting to catch and then when I was Just ranching on a seed stock deal for two years, you know, nineteen to twenty one, that was when you were at a point there to where the tipping point had hit and you were gonna have to genomic ⁓ You just Yeah. The customers at that point said, We aren't gonna buy bulls unless they're genomic tested. Through whether it was through the association, that was kinda all at the time that was available to guys, but They were gonna have to have minimum parentage verification. Right. Yeah. Absolutely. Some kind of genomic enhanced EPD. So that's really to me we're we're still on the forefront of, you know, genomics being an anchor point. speaker-1: Yeah. Well and I think too, like it it there was learnings that needed to happen with genomics like in the late two thousands and the two thousand tens because when it first came, right, it was here's the five genes that influence cavanese and you don't even need to use an E P D anymore because these are are describing all of the variation and and we learned that that's not true, right? There's there's tens of thousands, if not hundreds of thousands, of genes that affect weaning weight, right? There's probably maybe a few thousand that have a high additive effect. And so those are the ones that we care about. But those early renditions of genomic tests were were were subpar to say the to s to say the least, right? So we had to learn how to use the information. Those initial ⁓ single gene, single trait tests, you know, they they didn't quite work to the expectations of the producers that were that were paying quite a bit of money for those tests. Right. And so we had to learn how do we how do we take the existing infrastructure that we have, these pedigree performance blot evaluations. Can we add DNA information to those to add increase the accuracy of those evaluations? So that's what we did. ⁓ So the initial methods were you would take you would essentially build a molecular breeding value, which is just the accumulation of all these additive effects for specific traits. So weaning white, there might be a couple hundred to a thousand really high impact ⁓ markers that are associated with weaning white. So we would create a molecular breeding. Value and we would say this one's you know top one percent or bottom one hundredth percent you you depending on on how you represent those predictions. And we would blend those into the PBLUP evaluations, but that blending process also created quite a bit of noise. ⁓ and so Those those blended evaluations only lasted a couple years until ⁓ there was there was some competing methodologies, ⁓ one out of Iowa State called the super hybrid model. That's what IGS uses, and then there was the SSG blood methodology that which came out of Georgia. They're they're essentially equivalent in the output, but the way that they arrived to the to the output of those predictions is there's some fundamental differences. one uses the ⁓ the actual Marker effects themselves in the evaluation, the other one augments the genomic relationship matrix to better account for relationships in the animals in the evaluation. ⁓ Essentially equivalent. There's there's some nuances to those, but for the most part, any single so we we would call those evaluations single step. So rather than blending in after the fact the molecular information, we're including them from the very beginning and through the entire process of developing a genetic prediction. speaker-0: Yeah. And so basically everything we do today now is single step. speaker-1: Yep. For seed stock evaluations, everything that we do for the most part is at least across all the major seed stock ⁓ breed associations, ⁓ seed stock producers that are using those ⁓ evaluations, and most of these commercial evaluations, they're all on single step. It's it is the most effective way to use performance information and genomic information in an evaluation for sure. Yep. speaker-0: So let's let's go ahead and go back just a little bit. So I take a TSU tissue sampling unit. What a great name, right? TSU, so very convenient. I send it to Zoetis. What is the process of that sample becoming a set of numbers in a spreadsheet? speaker-1: I wish I had my SNP chip. so really any DNA sample collecting facility I mean, you know, the industry standard is these tissue collection devices, ⁓ but for the longest time it was hair cars and blood cards and ⁓ speaker-0: ⁓ my god, yeah. speaker-1: Anything that can carry DNA ⁓ we can evaluate. We our preferences are for sure those tissue collection physic units. so the the DNA is is put into those collection devices and depending on whether it's a dry desiccant or a liquid, right, the the the once that sample arrives at the lab the technicians will extract the the DNA and in those T SUs, the the ones that you you talked about, they're actually extracting from the liquid. They're not extracting directly from the tissue itself because that that That liquid will kind of ⁓ break apart the DNA and it'll be free-floating. And so they'll just pull that out of the liquid. And so once they have extracted the DNA, they'll run some chemistry and ⁓ put it onto what's called an aluminus SNP chip. and there's there's there's essentially a lot of different colors on ⁓ of Yeah, different colored boxes on that SNP chip and they'll smear the DNA ⁓ on that SNP chip and they'll run a laser through all those different boxes and depending on the color th ⁓ that comes out of ⁓ out of those different boxes it'll assign a A or B sort of genotype for those specific SNPs. ⁓ on the Zoetta SNP chip there's like eighty four thousand SNPs. ⁓ so it's there'll be essentially eighty four thousand results for eight ⁓ you'll get an A and B or A A or B B right for all those eighty four thousand ⁓ SNPs and then that information will ⁓ go into our evaluation and we'll translate those into the effects that are needed to go into the evaluation. So essentially what that means is like we were talking about weaning weight and I like to weaning weight as an example because it's highly polygenic, meaning that it's controlled by many, many genes, right? So there might be a couple thousand of those of those SNPs that have an additive effect for weaning weights. If you if we simple it down to the ten that are most influential, the first one might if you have an AA genotype, it might be plus one. If you have a BB, it might be minus one, right? And then for the second marker, it'll be if you have an AA, it'll be plus three. If you have a BB, it'll be minus one, right? So there's there's additive additive effects for all those SNPs. And cumulatively, if we add all those together, then we can arrive at a breeding value, a likely breeding value for those specific SNPs or for those specific markers that go into the evaluation. And then we can also use those and the genotypes that arrive from that information. information to better augment the relationship because we typically assume that full siblings are ⁓ half related, right? But they can be anywhere from 0% related to 100% related based on the alleles that they get from their siren down. So we can augment the relationship matrix using that that genomic information. ⁓ And really that's that's kind of the process that we use today. ⁓ So once it goes to the evaluation and it's translated to these additive effects, it goes into the generic relationship matrix and we run SSGB models and for all these different traits. Does that answer question? speaker-0: Yeah, no, we're we're getting there. But so the model has been trained by taking animals with highly proven data running their genomic profile correct. Just kinda go go there a little bit, Lane, as far as how how now we can say we're gonna test this eight month old calf. And I'm gonna add another fifteen to twenty percent of accuracy to their EPDs. speaker-1: Yeah. Yeah, that's a that's a and that's a key point to all of this, right? So in a single step step evaluation, like the these are like we're running, so you brought up a good point. The reference population is is vitally important to the success of this, right? So if you if you use animals that are distinct distantly related to the source population, the efficacy of the of the predictions might be diminished because so so like to your point, right? But how we arrive at the additive effects for those specific markers ⁓ depends on we need to have animals that are genotyped and we need have animals that have that have phenotypes or weaning weight observations. So that way, and we need to have many, many of them, and we need to have variation in their their pedigree and variation in their genomic diversity, so that way we can look at all the different combinations to make sure that we're not having So that so that way we're we're balancing it out. and so populations that are well connected to the reference population do really well for for genetic prediction. and if they're distantly so I'm gonna actually back up a second. So there's ⁓ there's two different there's two different types so there's speaker-0: Okay. speaker-1: what we talked about which is genomically enhanced EPDs, right? Right. And then we mentioned earlier there's molecular breeding values. Molecular breeding values are only using the genomic results and the additive effects to arrive at a a breeding value and then we're just using that. If you're a and so then the genomically enhanced EPDs or single step G BLUP evaluations, those are using pedigree performance data and then they're augmenting in the genomic re information. So the there's two different types of models and two different types of evaluations. For the marker effects model, if we have ⁓ so that one relies quite heavily on the reference population and the data to fuel those additive effects for all the different traits. If you're an animal that is well connected to that ⁓ reference population, those marker effects models work very, very well. If you're distantly related, right, so if you had a marker effects model specifically for Angus cattle and you brought in the Scottish Highland, right, those those predictions aren't gonna work at all because their their their genetic diversity is so drastic that they're not those predictions aren't gonna be informative for that that that animal. So that's why so there's there's some nuances to this. But ⁓ to your point, right, w we need to have really robust and quality data sources that fuel both. And you know and and this has been historically ⁓ the even the P Blub, the pedigree performance blop evaluations, those still also required on really data collection and you know there's there's a lot of when I was at Cemital there's a lot of work that our team did to look at can we build algorithms to identify if there's either false or mischaracterized or erroneous data that are entering into an evaluation and and remove it. And there was quite so we we built a system that did that to try to filter out data that were just introduced Right. I I'm sure I went off and much speaker-0: No, no, and that that's perfect, Lane. That really that really guides into the next part of this conversation I want to have because it's good to understand how we get from TSU to numbers. But there there's more to it there and and the other side of what I wanted to dive into with ya was the as you started talking about the data sets that back these evaluations and does it matter turning in phenotypes speaker-1: Yeah. speaker-0: w and you know we'll we'll try and stay out of the the flip side of this coin, but you know, I'm gonna spend eighty dollars on DNA testing. Why am I taking weights? Why am I turning all this in? It's a conversation that, you know, in the seed stock industries had every day. Right. So let's talk a little bit about, you know, the data sets that that back these evaluations, Lane, and why it matters. speaker-1: Well yeah, and so you you brought up if if if I'm spending money to do a DNA test, why do I need to be collecting performance data and all those things, right? So as a C stock producer, ⁓ I'm gonna separate this into two different classes. So as a C stock producer, right, your job is to provide the most accurate genetic predictions that you can to your commercial producers to de-risk their purchasing decision. So that's that's the priority. Number one, DNA can only be so informative, ⁓ especially on young animals. And so we still rely quite heavily on preference. And so even in the dairy industry, there's this huge effort to still incentivize and and really push for those nucleus bleeders who are responsible for all the genetic improvement in the industry. They still need to be collecting performance because to the earlier point, right. the DNA sort of how we use that information is only as accurate as the reference populations we have to to fuel everybody. So we still need performance data. We still need performance data on the best populations. And so if you're going to provide the most accurate predictions to your commercial producer, you're still you still have the onus of collecting performance data or as much as you can to help ⁓ add accuracy to those predictions. Not every sea stock producer is collecting every single trait. Correct. It's a lot of work to do that as as you and your your two are very familiar with. It takes significant effort and investing to that. ⁓ but the producers that do that, those are the ones that I think the industry recognizes as they're doing they're doing the the j the job that you that they're responsible to do. ⁓ the funds providing the best producers. So as a sea stock producer, I I think, you know You need to be collecting phones you need good D I just need you. All your bulls that you sell and preferably everything ⁓ for commercial producers, this is where things get a little bit different, right? So commercial producers do not there's no incentive for them to collect data, right? Right. They're not getting paid on how accurate their predictions are or how well they can sell their CAVs. ⁓ based on genetics because the performance data is is is used to function potentially. They are what they are at that point. And so their their job is to try to make improvements where they can. And historically, since all these genetic evaluations relied on knowing a style, knowing a game room and reporting performance That's why commercial producers have never had access to general variations unless they come perform the same unless they know as we're petting it. And that's not the same thing the gas control education producer. And so their job, so while there's a wealth of information between speaker-0: Mm-hmm. Business business. speaker-1: It's not quite as vitally important for them to to collect those data to the CD. Now on the flip side, but the SeasOC industry Sea producers, we we capture all kinds of data, but there's some key traits that we have struggled. Cal longevity is leather, right? speaker-0: Mm. speaker-1: If I'm trying to maximize my generation result, I might not be keeping females for their entire lifetime, right? So there's there's some challenges there. So in the C SAC industry, you know. I think we're as a as actually as a beef industry, as a genetics industry, we're shifting our focus from relying so heavily on performance data that's only collectively in C stack section, and we're opening it up to looking at are there potential data pipelines to fuel genetic evaluations and genomic predictions and complex those staff definitely. speaker-0: And so when you talk about trait development lane and this goes back to to data sets, how many phenotypes do you feel like you need to have to be able to accurately start to describe a trait through genomics? speaker-1: That's a yeah, that's a good question. And it it it it kinda speaker-0: And I'm sure it I'm sure it's trait specific on that. speaker-1: Yeah, it's trait specific and it it and it depends, right? So for really high heritability traits, it it Not not so many to basically understand general heritability, right? But to build a genetic evaluation, ⁓ it you know, it'll require many thousands. ⁓ it also depends on the size of the the population in the evaluation. So like for evaluations that have 10, 15, 20 million animals, you need many, many thousands of records, you don't have high heritability traits to inform those. ⁓ But for low heritage body traits, that number increases. So for traits like fertility or health traits, right? Those are those are lowly heralded traits, those you need many, many thousands. And if it's and so then there's also another nuance where if it's a continuously measured trait, ⁓ you know it it becomes a little bit easier. But if it's like a binary trait, like fertility, yes or anything, then it becomes even more challenging to evaluate with few meters to even many many more positive. So it's it's it's the job of of of a as a genetic evaluation like Genetesis, we have good job security because our job is to find data resources wherever and whenever we can. ⁓ because as these evaluations grow, ⁓ for all traits, but we need to continuously be feeling those with with new and better philosophical and high quality format. Because variation is our friend because it allows us to identify outcomes and we can't get out of genetic modifications. speaker-0: And you've got it. Yeah, and you you probably as we move through generations is there a need for more data because The DNA changes as we move generations. In my mind anyway, tell me anywhere I'm wrong here, Lane, because I'm you're exactly so so data needs to continually come to say this is staying on the right path. speaker-1: Yeah, that's and that's a key point to this. So, like to your point, the the Allele so genes will segregate, right, as as through generations. So even if we collected all the weaning weights and all the yearling weights, all the mature weights that we could today, right, on every anima, and we say, Yep, we're good, the quadgeny will start to segregate and those those gene comb those allele combinations will will start to shift and over time that will erode to where those performance data no longer inform ⁓ animals genetic potential through its genome time as reliably and as accurately as it doesn't data. And so so to your point, yes. It's not that we just need large volumes of data, it's that we need large volumes on current and modern populations and it's an annual thing. So it so it's I mean we constantly need to be fueling out of application. Yeah. speaker-0: Yeah. So Lane, I've taken a lot of your time this morning, but I would love for you to just just kind of touch on what you see in the future when it comes to genetic testing. ⁓ you know, here at Jorgenson's we we really want the opportunity to test our our customers commercial females because then we can we can better help them get where they want to go. And We're looking at adding more traits, more things that are economically relevant to our commercial customers, or it's a convenient trait that they're looking for. ⁓ all those little things. Is there anything on your horizon that you're really excited about that that's gonna be impactful to to the commercial industry? speaker-1: Yeah. Yeah, you bring up something important and I think so we talked about the the role of a sea stock producer before, right? As as one that goes to the effort of collecting performance data, building the best genetic predictions that we can and and raising nucleus populations ⁓ and and and deriving genetic improvement in the in the commercial contracts. ⁓ so if I'm a commercial producer and I'm buying from the people similar cities of of sea stock communities who are doing all of those things I'm taking care of the side side of how a few folks on. But the other piece, the the the female side, historically has largely been left speaker-0: Mm-hmm. speaker-1: ⁓ unnoticed, right? So it's it's one where you know, if if I'm spending lots of money on a bull, great, I can drive genetic improvement over time. But it'll be s it'll it'll be slowed down if I'm not if I'm not making the same level of perceivability. And so this is something where I got really excited when it seems to why this trying to provide solutions to Historically been left with the question of which replacement pathways do I keep that, right? Which one's the other. And historically it's been which ones do individual sort of ⁓ characteristics that every need, ⁓ which ones are the heaviest, which ones are a certain side group, which ones are certainly how should those are the things that have preferably done how to use research to the replacement entities. But that largely would draw you're just selecting the big ones, right? You're just driving higher energy costs. We're just driving heavier weights right of your mature FUMAS. We're not driving better fertility. We're not driving better because a lot of those are unknown. And so that's where if there can be genetic predictions on these these commercial phenas, you can provide an opportunity to identify which ones still have the look that you require. but also have the email potential. And so you're using the same so you're using the bull binding things and the customers. And so that's why I got you know, I think that's that's really where I got excited about the food shrimp and opportunity. And it's it's not just happening in Civilist, it's happening kind of stickless and syllabus. And so there's there's gonna be so many opportunities for potential for those who really dial in the first and how the use that we can see for the equation ⁓ for the applications. So ⁓ like you mentioned, you know, your your ⁓ your role when I'm with this is the role and Rogless right and the collaboration that we have between Zoetis and Joe S the ID evaluation stuff. You're you're addressing how do you build the the lifetime build for the apprenticeship type of producer, but then also dividing them the solution to say you're buying the right hyperverse, but we're also gonna provide you support and try to help you manage the fact that your cable ⁓ which not very many C SATRs used to be like doing and I think in my mind that's where ⁓ the the the quality of the C Stack thing is sort of ⁓ lies in both the the role and development of hyperverse but in the customers. Yeah. speaker-0: Yeah. And so are are there any traits that you're that you guys are are starting to to think about in development that that get you fired up right now, Elaine? speaker-1: We can build traits for a lot of different things, Jamie. and I think in my mind there's there's a lot of opportunity for some more of these ⁓ so ⁓ so I'll I'll I'll stepping on a soapbox. So historically, right, we've focused so much on revenue traits, live cap percentage, weaning weight, yielding weights, in some cases ⁓ carcass value, right? So we focus so much on those and as an industry we've historically over-emphasized selection revenue trades. ⁓ we haven't as an industry spent so much time selecting on cost trades. And so I think in my mind, right, because the the profit equation is is both revenue minus cost, right? If we're only focusing on one side of the equation, are we actually improving profit and operations? That's the question mark. ⁓ So in my mind, if if if we're gonna do a better job, we're gonna need to one address that there are opportunities now to identify cost traits. But I think the industry is largely and we're we're seeing a shift towards this, but you can actually select for profit by using indexes because it balances those out. But if the indexes don't have crucial traits represented in their rating, that's where as a as a as as a genetic supplier we need to start investing in the so ⁓ traits that I get excited about all these new health adaptability traits ⁓ some of these ⁓ these feed intake feed efficiency ⁓ all of those type traits that are we're becoming so much better at evaluating we're building data resources to fuel those predictions. If we can if we can mitigate costs associated with health outcomes and Fig Rogs, that'll save the industry hundreds of millions of dollars, right? So that's where I really did about this. speaker-0: Perfect. Well this has been really fun, Lane. I appreciate your time. ⁓ you and I have solved lots of worlds problems at the bar and I'm sure we'll continue to because it's when we get our best work done, right? speaker-1: But so speaker-0: But ⁓ really fun to get to work with you. If anybody has wants to dive deeper into the the sausage making behind genomics, ⁓ mind giving the folks your email address lane. speaker-1: Yeah, that's what. ⁓ email is lane.geist at zoetis.com. speaker-0: Yeah, and we'll we'll put it in the in the description guys. ⁓ Lane likes to talk cattle, so if you want to talk cattle, shoot him an email and see if he's got time. But ⁓ again, greatly appreciate your time, Lane, and ⁓ we will look forward to to doing this again at some point. speaker-1: Yeah, thanks Joe. Appreciate the opportunity to nerd out on informatic evaluation. speaker-0: We bet. Thanks guys for for checking into the source, and we will see y'all next week.