[Music] thank you everybody for joining us this morning I'm going to get started here so the title they gave me was practical approaches for balancing diets for amino acids and then we talked some more so there's some work towards the end where we're integrating energy to describe the amino acid requirements of cows that might not be as practical right now as we'd like but that's what we're hoping is the future in all endeavors I don't have a big thank you slide at the end but you can see a bunch of co-authors up there and there's more people behind that and then I have time to talk about right now but it takes it takes an army of good people to figure out how to get this done and my grad students of and colleagues have supported a tremendous amount of this so we'll just get going so today's discussion one of the opportunities for amino acid balancing when we capable of now I think the key to the whole thing and this is part of the whole modeling process I've been I was one of the first people to apply the same CPS to an actual feeding study all the way back in 1990 when it was still in the Lotus 1-2-3 spreadsheet and over that time it's just about my age but over that time what you find out is that the model is very powerful in many regards but you got to use it to figure out what's force limiting because that's really what it's good at and then once we know that then we can make our changes and make our adjustments you know so a little bit about basic biology around amino acids I'm actually going to go through that pretty quickly I'm going to talk a little bit about feed chemistry I'm gonna talk about feed ingredients and we'll talk about formulating for miss a plausible protein and refining formulation with amino acids they're really going to tie that together with energy all right so it will summarize quickly I think to me the the important thing here at least the way we approach this whole process is the cow is the ultimate arbitrator of whatever we do as Brian said right up front he goes when are you gonna get this thing done not soon enough for me it's kind of a love-hate relationship not with Brian but with the CNC PS but the you know I think that as we go through our processes here and we work on a model we're spending a lot of time trying to break it and find out how good it really is and we're doing a lot of cows studies to ask that question if the cows agree with us then hey we're good if the cows don't agree with us innocence we got to go back to the drawing board and figure out why we can't represent what the cow is actually doing with these nutrients so that's one of the things that's been kind of holding us up is that we want to make sure that it works right and like I said the cow the cow is our judge and we want to make sure that we're judged fairly and appropriately and when we know that we're good we're getting pretty close we can we can move forward right so to that and I I just feel all this stuff down to its simplest form we can get into the biochemistry of nitrogen metabolism and transporters and ubiquity and we can do aquaporins and how urea moves around you know but in the in the end a cow a cow can be disassembled to a couple of different parts when it comes to nitrogen the first thing is she needs nitrogen in the room and to maintain adequate microbial growth for carbohydrate digestion right not a new idea by far you know a lot of that was characterized over the last 50 60 years refined by certain people we know peptides can be stimulatory we know that pretty much all bacteria in the rumen use nitrogen protozoa eat bacteria that's how they get their nitrogen protozoa produce ammonia all right and we always forget that so in the end there's a requirement for ammonia in the room and we got to figure out how to meet that that that is a separate requirement from the amino acid requirements of the cow and I think part of our problem here is that we have continued to use this metric of crude protein to describe all of this and again Brian alluded to the next version of the model one half crude protein in the next version of the model cows are gonna consume nitrogen so you may have you may be driving cows that are consuming 600 grams of nitrogen and we're not going to convert that back to crude protein because it actually confounds a lot of the compartmental modeling that we do to describe nitrogen through the system leaving it as nitrogen is much more efficient you don't have a ratio so once we have room and requirements met the cow requires amino acids and and now we know that there's there's a lot of forms of that obviously microbes are the big one under graded theater the other one other aspects of that but you know we've we grossly underestimated the supply of endogenous protein proteins that have slough off from her room and her esophagus her tongue enzymes and things like that secreted along the way that can be 20% of her amino acid supply so I depending upon what Cal you're feeding and what her intake levels are and we can thank elaine Lapierre genuinely and a few other people who have demonstrated and actually given us the data that allows us to model this more effectively now so there's like I said there's more people on this slide that that the title slide that I didn't have on there but have made major contributions to our ability to do this and then in the end anything she doesn't need or is fed in excess is excreted in the urine and that's where the problem starts right if she's if she's being it it would a its nitrogen she doesn't need so we're feeding something that's not being used effectively so we could either be over feeding it we're feeding it in the wrong form doesn't have the right amino acid profile too much room the nitrogen all sorts of things to think about there but so we're using this concept of urinary nitrogen relatively productive nitrogen to kind of figure out how to pull back and some of that excess nitrogen that she's wasting right I think one of the big things that we all kind of know but we don't formulate using this concept all the time is it milk yield and milk protein synthesis is an energy driven event there are our energy driven events right and it's energy it's the ATP that's going to drive protein synthesis and Drive lactose synthesis it you know then you're gonna get a bunch of you're getting a bunch of downstream indicators after that right you're gonna drive it through insulin and igf-1 which we'll talk about in a minute so once you have met the energy requirement or you know where it's what the requirements should be then the outcomes we you know relies on an adequate supply of amino acids and this is all driven by appropriate production for the most part you know foreign aids converted the glucose or since stimulates insulin right and what we know is that for the most part give or take as we produce more procreate about seventy percent of that that glucose produced from procreate is going to be taken up by the mammary gland that seems to be kind of a fixed number right but you know there's been a lot of discussion about should we have more womanly available starts you're less normally available starch and then you get into the hot theory and I'm not going to go there today what we do know is that as we push more glucose to the large antenna or the small intestine that also supplies you know energy to the cow but there's a discount on that energy for lactose synthesis the glucose is taken up across the gut and based on the data of Reynolds and several others Katherine Knowlton there's other folks out there now that have looked at this there's at least an eighteen percent discount that energy do the tissue you use prior to getting into circulation of the point during viscera but what we also know is you don't generate the same kind of insulin and igf-1 tsa's from that form absorption so intestinal glucose is okay but what we really want to do is make more appropriate eight and we want to be really efficient at it and that to me is how we're gonna get all the downstream signaling that's going to help us produce more milk protein and some more milk alright so we know insulin stimulates protein synthesis and in the mammary gland this has been shown for a few years to the house and some of the Yuka lysine that gives me the clamp date that was generated unveil Bauman's lab and then others and we know that energy intake stimulates igf-1 secretion from the liver right and obviously it does that in many different ways lots of binding proteins that that's not you know that the control of igf-1 is not important here we just know that we get greater there's greater synthesis and secretion of igf-1 from the liver as energy intake increases alright we also know that protein supply per se it's not an activator of milk protein output although I think there's some data coming that would suggest maybe it is for certain amino acids but we know that there are receptors and promoter regions for igf-1 that are responsive to things like amino acids we know that mTOR is one of the gatekeepers of all of this and responds to the presence of certain amino acids in the face of khattak with energy or inadequate energy and you've got things like elongation factors that will stimulate you know the production of proteins then we know the Metheny leucine and others are involved in this process but protein can be a signaling mechanism one of the cleanest examples of that is here and it's as a calf study conducted in Illinois and Jim Drake Lee's lab so this is you've got two levels of intake I'm gonna just look at the red line on top so those calves were fed at 14 percent of body weight for different milk replacers that varied from 14 to 26 percent crude protein but they were fed aysil calorically and you can see you know that plasma igf-1 if you gave the calf more energy you know from yellow to red so we went from 10% of body weight a day in intake to 14% you can see that that energy does increase the secretion or the you know endogenous production of igf-1 probably from the liver so circulation circulating levels go up but notice what happens when there's enough energy and we add more protein but you can see in that red line we're increasing the circulation and the concentration in the circulation of igf-1 all right so again the idea that those two things are integrated is really important there's a nice clean example of that where we can see that energy is stimulating igf-1 and then as we titrate in more amino acids from a greater supply of protein that the liver is responding saying hey you've got the building blocks to make more protein you've got plenty of energy I'm going to give you greater signaling capacity to stimulate that outcome okay we know this also happens in a cow I don't know of a clean dataset in a cow where where this kind of a titration has been done which is why I come back to this really nice calf study all right so they are integrated and I think we have to keep that in mind and we have to build our system to represent that all right and we know some of this you know if we go back like I said to some of the work that was done in Dale Bauman's lab when with hyperinsulinemic you glycemic lamps you know Tim Akal just doing the clamp with no increase in an amino acid supply or any branched chain amino acids you know the mammary gland increased no protein output by 15% the Miko came along and he did the same thing and they infused casein and they have amazing they got a really nice response right so it said the signaling is there but once you give the signal you got to provide more protein then and then to come back on that you know there's a lot of there's data coming out of France and some other groups right now that would just as we put in more gluconeogenic diets we're going to increase our efficiency of amino acids and that is what you would believe from the makkal data and I don't disagree with that my concern is is that Tim Akal only did that for four days so it makes me wonder how long they're gonna have that available protein pool to allow them to do that we get some indication that the signaling is greater than that once we had some protein we get a better response which is what you see in Miko study and then in the second makkal study where they added branched chains and casein and Brian Beckett saw the same kind of response when he did clamps and IV infusion of amino acids and and he saw this insulin response in goats right and then so no no milk and protein increased significantly in those animals so we know that this is intimately involved in this whole process this very nicely says hey we got a we got to marry these two approaches and make sure that when we just grunt describe the amino acid requirements we're doing it with energy in mind because that's really what's going to drive this outcome an example of that is a study that we ran years ago in a 3200 cow dairy and just this is a farm study we're in protected lysine to to high pens went out of control to hype in spent on the RP lysine this is back in the days of CPM dairy I'm not going to get into all the numbers right now but what we did was a very very simple study just brought this product and split the Pens started feeding it and you can see here this is real data here's no production in pounds and we can see the blue line or the cows that received the RP lysine the red line and the cat are the cows that were on control and you can see some differences here over this 80 to 90 day period and then when you you know take it away you could see them come together right so a Nokia we see a little bit response when you look at milk protein production same kind of thing you initiate the treatment between the two pens and then you can see a difference all the way across till the end you take them off and they come back together right so again we've got a little bit of a milk volume response a little bit of a milk protein response and then when you look at fat in those same cows right what we see here is we see again once we start the trial we see a nice separation of the amount of fat that's being produced again it's not extreme it looks a little messy this is this is herd data these are daily daily calculations we go for about 90 days and then we stop and you can see them come back together right well why don't I show you this what's my purpose here well this to me is informative because you know one of the things that I you know we always get into this mode of saying hey if I'm gonna add a room protected amino acid I expect a protein response okay and you know well maybe maybe if that's what her inner product need is at the moment depends on like what she's deficient in and what she needs that amino acid for jail-boat lee did a study with Rajee Oh several years ago and they found amino acid carbon in in glucose they found amino acid carbon in fact they found the amino acid carbon in protein so the cow will use it in whatever way makes her more efficient I look at this amino acid supplementation is the way to just increase the energetic efficiency of the cow and I look at data like this because when we look at this outcome dry matter intake of those counts did he change that he was about you know almost 27 kilos a day milk volume increased about 1.4 kilos no protein increased about 68 kilos 60 68 grams milk fat increased about 45 grams on an energy corrected basis she she improved her productivity one point six kilos a day because intake didn't change you know feed efficiency or energetic efficiency approved by about three point three percent right so the way I think about this anymore is it how did I improve my feed efficiency because I removed the most limiting nutrient and in the end I don't believe this is a linear I'm going to get more milk protein I believe that she's going to use it to improve her energetic efficiency and in this case it was three point three percent you know at the time was twenty nine cents a kilo so was seven dollars and sixty eight cents to feed these cows middle acid cost was about twenty nine cents increased revenue was 78 cents margin was forty nine cents right there was a nice margin in the end but she had to look at all of the metrics he couldn't just look at milk protein or milk fat or milk volume by itself right so but that comes back to your milk market to and what do you paid on okay so integrating energy so I and I think I make this recommendation to to everybody now formulate an energy corrected milk because that that will capture your volume protein or fat yield changes and for the most part that's usually easier to text easier to detect than one particular aspect right and I've got lots of stories going that time today but I've got lots of stories about people coming back and saying my K I did this and I didn't get a response right and I'll ask them what did you expect us is well I expected more milk protein more milk protein output or milk protein percent you know and if you're in Italy they're gonna say especially in the parmigiano region we're gonna say well I need more percentage because that's how I get paid and I've got several cases where they got greater energy corrected milk output but it didn't do exactly what they wanted to do so they thought it was a failure right so energy corrected milk is really more informative for this type of outcome and and we need to think about that okay so formulation considerations for meal acid balancing again some of this is really simple I meant trying to insult anybody here but these are questions when somebody calls me up and says hey I'm not getting this to work right can you help me out all right so I'm gonna go right back to the very beginning and say okay where we at do you have current feed chemistry or you just used the library values right can you tell me you're using library values and I would say okay we got some work to do here did you do NDF digestibility and somebody's gonna you know everybody says what does that have to do with amino acids well to me it has everything to do with amino acids I'll come back to that a few minutes have you characterized the cows appropriately man this is a tough one because it means we got to spend a little bit of time and it's not that we don't spend time doing this but means we have to have fairly intimate knowledge of body condition scores body weights days and milk stage and lactation for all these pens and it actually sometimes means we need to have cows that have been tested for that particular pen and not just some bulk tank value for MU Ensor protein or something like that and that characterization to me is is important and can be really powerful you know how much milk is that pen really making what's the range in that and do we have dry matter intake and if you have dry matter intake and you've got good feet chemistry and you put it all together where you line up where's the actual milk line up with Emmy and MP allowable all right because then if those two things cannot agree and you're several kilos pounds off whatever happens to be until you reconcile those two things you really don't know what's first limiting and then it's really hard to make a good decision all right if you don't know what's first limiting then I don't know what we're gonna do first to improve the situation if you can't make those things align sometimes that means you need body condition score change all right and I know it's part of characterizing the cow but I put that in there because many times I'll get somebody who will say to me Mike I you know I've got diets formulated you know for 43 kilos or 90 pounds and the cows are making you know 37 or 38 and I can't figure out where that extra milk is that you know and they're down in the low 80s and come to find out you know for some reason they're not persistent right this is what we do so they says there is something wrong and then we go do some body condition scores and we find out that those cows are creating body condition score at a rate that accounts for that extra energy for body weight with your body weights very often but if you get body weights you can do the same thing so the point is is that where they're a lot of times they're still very efficient you know but we go chasing this milk not really knowing how the cows partitioning that energy and that is really important especially if you want to refine down to the gram of amino acid level and finally what's first limiting once you know all that you should be able to get to some assessment of what's for slimming and then you can make your adjustments things that drive me crazy and and I and this again I I do this simply because I weekly weekly I get calls or asked to solve problems and I I see this happening like I can't make things work nothing the energies don't line up all this kind of stuff well you find out that the cows probably weighs somewhere between 750 and 800 and 20 kilos but the input in the model or whatever rations software they're using doesn't have to be the same CPS could be the NRC could be Spartan could be any of those right I get calls about everything you know they're about they're a hundred kilos or better off on body weight but you can't be a hundred to 150 kilos off on body weight and expect to reconcile the diet at grams of amino acid efficiently and I know this is one of the hardest things for us to get but boy it's one of the biggest drivers of not figuring out how to get really efficient milk out of cow's not using actual dry matter intake you know I still see that not using current feed chemistry or just using crude protein and NDF or not really well characterized in IRS all that stuff will get you into trouble if you're trying to do refinement right and again the last point there is just what I said in the previous slide not taking the time to get the cows in the formulation system to agree with each other so you can make an informed decision that that's the power of all of this software whatever software you using that's the power of it you know it's to get those two things to reconcile figure out what's first limiting and then go ahead and make the right decision inform decision to improve things now here's an example of I just happened to have this here's a GPS prediction for 6:55 body weight at 1588 with a mature size of 1764 it actually says on an MP basis so I know this is not an amino acid yet but on an MP basis for where -38 grams on MP balance which means we would have an opportunity to bring in some amino acids right we're deficient on MP and I didn't change this sorry I didn't change this when the kilos but a lot of times I see people doing this right we got cows that are 1588 but they're modeling something at 1350 there's no deficiency there right one of the fact that probably is because that's not the cow and it looks like a really subtle thing but I can tell you this is more frequent than then we would like to admit to ourselves sometimes and we've got to we've got to get better at this right especially we're going to get two grams of amino acids all right and I said this earlier just to recap this need we need to separate crude protein from true metabolize for protein and amino acids because the cow just doesn't understand that all right back to the idea about urinary nitrogen here old data I use a slide a lot because it's pretty clean you know at Ohio State in a Penn State study and you can see they went from 430 grams of nitrogen intake to 750 for fecal nitrogen doesn't change what they don't need they excrete in the urine right and we can see that we're looking at a hundred you know hundred and forty hundred and fifty hundred and sixty grams increase in nitrogen excretion here and the the the CPS I'll come back to that has the ability to predict this with fairly high accuracy right so if you take this kind of data and and I pulled this together from some of our studies and stuff from Broderick and a few other labs I can't remember everybody right now but you know these cows are making about forty kilos and milk they're eating about twenty four and a half kilos the dry matter intake the crude protein of the diets range from fourteen to just about nineteen percent and the blue line there so on the y-axis you have milk urine and fecal nitrogen excretion on the x axis is nitrogen intake the blue line is milk nitrogen that's milk crude protein or milk true protein actually converted to a nitrogen basis and probably doesn't include the M UN's now I can't remember but it's going to be milk protein converted to a nitrogen basis to put it on an equivalent comparison and you can see the red lines the theses and the green line is the urine and when you look at most cattle there where that red circles at you know their excreting more urine and fecal nitrogen than they are milk nitrogen so what we're trying to do is is with the modeling exercises get them to get down to a one-to-one basis right and if you do the calculation on that you're looking at about forty grams of nitrogen and if you take forty grams of nitrogen and work it up to a protein basis and put it on a soybean meal equivalent you know you're talking about one point three to one point eight pounds of soybean meal difference there in terms of the nitrogen contribution at least right so so that means that you know if your excreting that much urinary nitrogen that you don't need and you still have a room it's positive in rumen and then you could pull out that protein that she's not making good use of and basically wasting because she's peeing it away and now bring back some carbohydrate and now you've got the opportunity because you should be closer to your protein requirement to bring in amino acids right it's now you've made space you've Scotland you did some cost savings got rid of the stuff that's creating an environmental problem and then you you bring that bring in some other things too one make more bugs and get more energy out of the diet and also now you can supplant that with some protected amino acids if you're doing really well you got the right forage makeup you can be in that instead of one one to one range on a urinary - milk - you're an area milk - productive and you could be one point two to one point three two one or better this takes much better management takes the right forages takes high quality forages and it takes a little bit of time to build a diet like this right and and this is a doesn't really matter which version of the model you look at they all have these predictions you know so again average a lot of the industry's 0.7 to one acceptable really good one point one or one to one outstanding would be one to five to one you can see here in this output from NDS it's one point two six to one Essene CPS output here productive and a year in area and one point four six so this is actually a really good bird and then they're very efficient with their use you can see productive and a total hand at 38% right so sometimes we can't change the overall efficiency of the diet and get more in stimulate more intake or milk protein output per se but what we can do is get rid of the stuff that she's excreting and not making use of and then bring in the higher quality ingredients again it's more carbohydrate make some more protein make some more appropriate and and then bring in our amino acids feed chemistry is important you know we talked about this a little bit digestibility is really important here you know if you have a good adjustable forage you're a hero I know it's not your you didn't cause it but you got to use it but you know when we do our modeling exercises the most important variable in predicting amino acid supply is digestible NDF and that's described I think we put that into the paper that Ryan Higgs wrote and published in 2015 you know because what happens with high a digestible NDF is you get three outcomes that we know are all better right you improved feed intake we get better room and health we get more microbial yield and and all those things are going to yield more energy more appropriate eighty more amino acids and and we're gonna get more out of those cows okay and I think this is important I'm not going to go through all of this but you know some of this is coming out now because I think you know with the quality of forages we're having a bad year right now with the spring at least in the Northeast and the rest of the country looks better but we're bringing cloudy and rainy and snowy up until the last couple days and we'll get one sunny day and then it's back to four or five days of clouds so our forages you know from last year still aren't good I don't know what this year is going to be 750 kilo cow should be consuming at least one point two percent of body weight in DFA in the fom that's nine thousand grams you know that means that 27 kilos the dry matter intake you know that should be about 33 percent and the f om in the diet in the room and they're gonna keep somewhere between seventy eight hundred and eight thousand grams in the rumen at any one time we're probably having maximum we think we have a maximum amount of und F now and those cows would cap out at about forty eight hundred and and now we have we're working towards the ability to integrate this into our calculations because of und F and the forages high that becomes first limiting otherwise total and EF om will limit fill and I don't think that the cow has a und F requirement that's a different talk but we know that rumen function will decrease when inadequate and the f om intake and content and and we need to pay attention to this right and because what happens is when we have higher quality forages they fill faster and they empty faster and you can see this these two lines you get a diet the base diet with lower quality forage and the green line with a higher quality forage and you can see that it fills and empties faster which means we get greater intake and more you know and every time that happens we get more bugs more propionate more everything ok other things that confuse us what's getting to the small intestine is it digestible ok and you know get back to figuring out what's most limiting we developed an assay we've been working on refining that si think we're done now we should be getting the paper out here soon as we went back to test offensive other enzymes to make sure that we didn't miss something it is being used although I think we've got to clean up how it's applied in some of the labs when we used this data in place of acid detergent and soluble nitrogen alright we ran a study with a height-adjustable blood meal and a low digestible blood meal house we're eating about 670 grams of nitrogen a day we were formulating you know between five and six percent of their intake in terms of the difference in digestibility which was about 38 grams diets were about 15% protein about 32% in the F 30 percent starch cows consumed the same amount of nitrogen which is good because now you can test your digestibility differences that those were not those were significantly different than we wouldn't have a valid test right so I show that what it says is what we learned is that the cows actually understand this just like a pig you know and then the monogastric live and die on an intestinal metastability and that's something that just hasn't been part of our process here a normal feed chemistry but this clearly says that if you have a high digestibility feed here in this blood meal versus a low the cows immediately know that we kind of assumed this would be true we were happy that the that the assay was will it was capable of predicting this when we looked at the milk yield you know the cows are in sorry this is in pounds the six hundred and seventy grams of intake dry matter intake the same we got about three pounds of milk volume and about three pounds of energy corrected milk difference body weights were not significantly different they all gained about the same they gained about the same amount of body condition score feed efficiencies weren't different hope nitrogen wasn't different but we took all that data and we put it back into the CNC PS alright body condition and average daily gain chemical composition of feeds with the target growth system to determine our growth requirements and we used the U n values for those blood meals is the only and that was going to be changed we put that in in place of the ADI n and what you see here you know you see the energy corrected milk at the top in pounds 92 and 88 Emmy allowable the energy level is over a hundred right which is how we formulated and some people would say well geez the neighbor that's a really bad prediction by the model and I'll say no it's actually really good because their protein limited you can see that if we use the Adi n and ndin we're still predicting close to a hundred and if we stick the unnsa' data in there you know L so this says we are protein limited the high digestibility blood made it was forty 94 pounds of MPA level milk 92 88 maybe seven so it comes in line right so we were able to do that fairly adequately but that's probably really an amino acid deficiency right lysine was going to be deficient there so so these digestibility factors and things that go unknown will affect our ability to predict limiting amino acids because in this blood meal we know we were deficient in lysine okay so our opportunity here was to recover some of that milk loss in the adjustability blood meal by adding a ruin protected lysine source that would have defeated the purpose of the study alright so when we start looking at amino acids so now I'm going to drive in for the rest of the talk it's all amino acids when we get into this you know we've got many predictions for what things should be right and we're I think we're pretty well set if we look at Metheny and lysine specifically here we can see you know rule Clinton 7 3 + 2 5 4 Metheny lysine like Smith I mean Gopal 7 2 to 5 rule Quinn 7 3 basically 2 5 then RC 7 to 4 and 2 3 8 we look at the recent version of the model 6 flat 5 we get to about 2.6 as a function of the mp9 T as a function of the MP which is a little bit higher than everybody else but not far from what rule Quinn had practical application is we have started to tie this to Graham some athenian for EM Cal of em e write our original estimations were one point one two - one point one five we have data now that says it's at least one point one five and this current version and should probably be closer to one point one nine we look at lysine we can get the same breakpoint analysis this broken stick model and we get to 7% of MP is lysine right so not just Gordon from the rest of the information that you've seen from those previous calculations where we've moved to them was using data from dope a lynn Lapierre and looking at efficiencies of use and when we pulled the datasets together and read arrived all of this we came up with fairly similar efficiencies you know our efficiencies a little bit higher for things like leucine lysine was fairly close point six seven point six nine our efficiency of use from assigning was lower than what helene had but what we did differently in all of this is we went and that calculated all of this on a gram of digestible amino acid for M calamy basis and derived these factors here right so it would say that in version 7 we want to be feeding one point one four grams Metheny / M Caliban me we want to be feeding three point three 3.0 3 grams of lysine 4 M Cal of M e if you take those values and you put them on as a percent of essential amino acid just for reference 15 one and five point seven and again if you go back to the Schwab numbers you're at 49 and 5-1 so we're very much in the neighborhood we go to role Quinn's data 14 seven and 5.3 so in a 2.7 and our values 2.7 right so we're were you know we're pretty comfortable with this approach so we took this data and we ran a study several years ago heat stress conditions it was summertime the corn silage moved nutritional nutrients on us but we had a base type that was limited in the finding limited MP but adequate and ruminant are limited rumen in we added Matheny we were still in an MP and rumen n we added urea there are adequate methionine but in now adequate room an end but limited an MP and then the positive diet was adequate and everything I call this third treatment here that Chuck Schwab treatment because he didn't believe the cows could recycle nitrogen I'm glad we added that because when the corn silage moved much lower on crude protein we needed that extra nitrogen to keep the rumen and positive nitrogen balance we did this for a hundred days measured everything did fecal collections for NDF digestibility - that calculator and these you know the bottom line here in the interest of time crude protein 13 5 13 6 14 6 15 6 right so we are doing what we want to do and that is push the boundaries of our understanding and find the limits of the cow but also figure out if we can predict that using our modeling exercises here can we get the C NC PS to predict this if you look at these optimum grams of amino acid for M Caliban and me we'll just look at masada and lysine here Matheny you know one point one four point nine three were deficient one point one three now we're up pretty close to the requirement here based on our numbers and that's with adding the Matheny we add the urea and it comes up a little bit more because the rumen was deficient in nitrogen so we get more efficient microbial growth and then at the positive we just got a good supply of the state protein plus microbial yield so we're over that and you can see that's true for the most part for all the essential amino acids if you were to look across those numbers right so what we find dry matter intake for the same and if you corrected milk 13 there are 38 539 340 all not different 41.8 right so all positive and we see a significant increase in that part of that amino acid supply part of that is just room and nitrogen balance but I think that's interesting right because if you come down here and look at note volume 30 half 13 614 615 six cows making 40 almost 41 kilos of milk on 4th and thirteen and a half to 14 and a half three protein diets right does say something about amino acid balancing I think our profiles of amino acids helped us out here you can see that on a milk volume basis right not significantly different it's not until we energy correct the milk that we pick up the difference okay so this game is some comfort to say hey you know what the cows are understanding these relationships and these ratios because the cows kind of responded now we had low milk fat you can see the facts down here as we were really in heat stress conditions in an old barn that doesn't exist anymore but had really bad ventilation protein still doesn't look good but again house where heat stressed so I think we were suffering a little bit there right we can look at their intakes of nitrogen 521 532 580 to 615 right if you look at M R P UN's you know here's where we got in trouble with those luck that let drop in the corn silage crude protein five point nine five point seven our loan anything less than six based on some older data would suggest that the rumen cannot maintain positive nitrogen balance we're not recycling enough we can't recycle enough because it's just too low in the system and in fact we have that prediction set up in the model we can predict bacterial growth depression and we predicted about a 16 to 17 percent depression and in fact you could see that in the NDF digestibility right so that all kind of came together and you see you know when we enter the urea now we're up in a range that we feel comfortable and we got a good productivity out of those rumens and cows right so so there is a limit here of how well we can go but the key to the whole thing is keeping that room and happy so it can make keep enough nitrogen there to keep the bugs doing what they're supposed to we followed this up recently Angela Pierre is currently doing PhD in our lab and you know we wanted to challenge these numbers right because that what I just showed you from Ryan Higgs was the first study that I know of where we balanced for all essential amino acids on a gram 4m calleb energy basis right it wasn't just with finding a lysine we did it for all the essential amino acids so we said can we do it again right and those were individually fed cows well now we're gonna do something completely different so here we have a study where we did 16 cows per pen 3 pens per treatment so we're doing this in pens 25% of each pen was a preemie Paris cow enrollment was 60 to 120 days and we were blocked by everything that would make them was balanced as possible right and if we come back to how I didn't show this earlier but this these are the graphs from Ryan's work and in these ratios here I need to change this it's not those aren't really ratios those are efficiencies of use of amino acid all right so the y-axis is a really efficiencies of you some you know acid on the x-axis you have the adjustable Matheny supplying grams of methionine of amino acid per M calleb and E and you can see that we we could solve these curves to find the optimum right which we did for both the room did that for all essentials what's most important here though as you can see there's dispersion around our predictions so in this first study what we did is we tested well what would happen you know if we fed our optimum but then we went up one standard deviation for all the essential amino acids and of course what would happen if we went down one standard deviation for all the essential amino acids and you can see that the dispersion is greater and lesser he went through all the essential amino acids you're gonna find you know lysine has a pretty big range here Metheny not so big few the other amino acids again don't have such a wide range so it's not even in a uniform distribution among all amino acids so that makes it tricky to do this it took Andrew a good for months to put these dives together and a little bit of head banging but you know a little bit of scotch and a little bit of thought process and even got through it but it's it is hard initially here the diets heavy corn silage I'm moisture corn and you know you can see soy + in here you can see some smart amine in there you see some blood meal in there energy booster dextrose to drive fermentation you know here's what the chemical composition looked like you know free protein 14 14 7 16 so again the negative is down one standard deviation for all essentials the positive is up one standard deviation for all the essentials at least as good as we could do it right again these kinds of studies aren't done this is only the second time that I know anybody has ever done this right and you can see you know where are you peas or at not that I like that metric you see the Emmys we tried to do this under isochoric conditions so we didn't confound it starch is right around 30 percent here's the grams of amino acid so I'm gonna highlight the lysine and the finding in the interest of time Matheny went down one standard deviation you get 71 grams and metabolizable like I'm assigning today 78 93 do the lysine 201 222 250 right so so big swings and all of those those were actually easy to match because we can use high quality ruin production sources you know we had again we had a little bit of a hiccup here I don't know about tryptophan but for whatever reason are positive diet wheat we just didn't get enough tryptophan in there everything else tends to go up accordingly isoleucine kind of plateaued here a little bit but again you know this isn't it's never done this before us to learn how to put these together and we don't have a solver to do this for us so it's all hand iteration but I am this is where we ended up you can see the lights in the tiny and you know for the most part dry matter intake this is in pounds so I can get this corrected as in pounds you can see there's a big drop in milk from the neutral to the negative right and those are significantly different when you go to the neutral to the positive though there's a trend but they're not in a P less than 0.05 they're not significantly different so it says if you were to plot this out it looks like a plateau effect you see a little bit of an increase there but it's not remarkable there's a bigger drop by going the other way so it gave us some comfort that we're not too far off when you look at your protein yield and this to me is really the most important thing yep we went down significantly when we went to the negative look at the neutral 20.78 pounds and you go to the positive 2.8 for those are not different the facts are not different the lactose is not different right true protein percentages are not different so so this was pretty comforting to us and said hey you know what these Optimum's may not be that far off you may have one or two that we've got to work on once we've got to set but for the most part we can operate close to those neutral numbers and and what we call neutral our original optimum numbers and say ok we can we can formulate around that and not worry about losing too much milk and again we're gonna work through these differences here because some of this may be the energetic side of this which is what we're gonna text next and yeah make this point I got several people that said to me once Jesus at Hamburg you know you can't you'll never be able to do this in the field because you individually feed all these cows well we did this study in a pen simply to make the point that in fact the average feeder on the average farm could probably pull this off as long as you've got had enough information and knew what you were delivering every day alright and we looked at bodyweight changes some people would say well geez maybe they're not gonna gain any weight they're gonna be too deficient in fact we didn't see any of that they did gain weight right and you can see the MP if you put it on an MP basis about 26 50 about 29 70 in just over 3,200 grams and and there were some changes in efficiency right and you go from the neg 1:49 g/kg one five seven one six Oh so again kind of a plateau in efficiency of use of the nutrients and and that says this is optimum the neutral values are probably okay right but the challenge that one more time and this is a few more slides here now and another study that we conducted last year and finished here just a few weeks ago as we did it in two blocks again sixteen cows per pen 192 cows same kind of setup 25 prepares 25 Boulton Paris and here's all the the information on that particular study on the cows so here's there's data out of France and and some data out of Canada and a couple studies here in the US that are looking at the variable efficiency of use of amino acids and you can tell from our perspective that we picked an optimum and believe that this stays relatively fixed but you know again like I told you earlier the cows the ultimate arbitrator of that so we're gonna test these things under our contract to find out if the cows agree with it or not because if the cows don't agree then we've got to go back and modify the model to capture that information so in this case we had a low fermentable starch diet and a hundred percent of our amino acid requirements not really MP it's really those amino acid requirements and then we had a high fermentable starch diet at a hundred percent of those amino acid requirements in other words at our Optimum's and then what we did we did a low fermentable starch diet and we didn't go up quite one standard deviation but we increased all the essential amino acids on a gram per unit of energy by five percent so now we're by our Optimum's were over feeding the amino acids then we did the same thing we went five percent higher on starch and we went 5% higher on the amino acids okay so we've got an increase in procreate supply and amino acids an increase in propionate supply and amino acids no and no so isochoric conditions though so Morrissey togetic diets more gluten a genetic diets again heavy corn silage upfront little less here you know we're using steam flaked corn and those high formidable starch diets beet pulp is in there some wheat meds you can see all the make up they're gonna keep going crude protein fifteen eight sixteen eight fifteen eight sixteen - okay is how they work out again we're not looking at crude protein we're looking at grams of adjustable amino acid for M Khalid M II and making our assessments that way that's how we're putting the diets together so you can you know the crude proteins just kind of float you can see this eight and B fom were in that 30 to 31 to 32 range middle starch you know and the low starch diet seventeen and a half to 18 22 so not a big difference inform edible starch when you look at the actual starts concentrations 23 and 28 to 29 I any intake or enemies were about the same its formulated you can see the math I mean seventy six point eight versus eighty seventy six versus eighty one ninety seven you know so we're going up at least 10 grams on lysine dry matter intake you know we went to those high starch diets the cows ate more and we're going back to test that this is all preliminary data this came out we were able to put this together the other day no kills you know a little bit of a difference there but not significant a little bit of a difference there but not significant where we get significance is on the high starch 105 percent of the optimum on amino acids when you do energy corrected milk now they kind of separate out and get about a kilo on the low formidable starch diets by adding more amino acids when you get a higher for minimal starch and add amino acids and you know have higher starch you see about a kilo about two kilos of milk difference again we're going back to evaluate everything this is hot off the press probably hasn't been fully edited I know that Andrews working on making sure he's got all the accurate dry matters that's probably gonna take him a few weeks to settle in but Brian wanted me to make sure that I showed you this that you know we are looking at this the implication here is that hey if you put more starch in there you get more appropriate you get more signaling a little bit more output if you give them the amino acids this is kind of like the clamp data right what we don't know here and what Andrew and I have been talking about is since they did a little bit more you've actually got greater amino acid production in the rumen so I'm not sure your efficiencies actually changed if you you know because we were testing the idea of variable efficiency as you increase propionate supply goodvery will be we just got more bugs and we got better signaling so we got more output and that's what we've got to go back and test if you can see that the cows did respond a little bit when you gave them more starch they actually did use it you can see a little bit the true protein percentage is their body weights they all did what you wanted to do with body weights okay so they're yeah low for metal starch diets at lower dry matter intake a little bit more forage and then beet pulp so maybe that filled them up a little bit a greater - protein yield and the - mineral starch diets and the MP or actually the amino acids helped with that where we're at we're going to evaluate everything so that's kind of a primer hold on you know fasten your seatbelt with what you know where we go but in a couple of weeks we'll have that sorted out we're just a little bit ahead of it right here okay so just to summarize how do we optimize sufficiency determine what's most limiting do the cows and the model agree again it doesn't matter what piece of software you're using doesn't matter do they agree because if they don't agree then it's really hard to make an informed decision in our hands you know we're gonna look at that room and in balance and urinary nitrogen excretion if the urinary nice administration is high and we've got plenty of room and n we're gonna keep pulling out soluble protein sources and so we can pull that room and end down and while we're doing that you know maybe we don't have to pull too much out because you can bring in some formidable carbohydrate use up some of that room and then make some more microbes and then think about bringing in some amino acids you know if it's just an MP effect we saw heard a few weeks ago in a case study with fellows where they were deficient and rooming in but very XSI metabolizable protein right so we actually had to go the other way on that particular diet that'll see that a lot but it does happen once you have those things in balance and you're you're good with your room and end balance now focus on amino acids for Matheny what we're using as a starting point is 1.15 grams of methane or methionine per M Cal and metabolizable energy and based on those that study the first study I showed you from Angela peer we put that back through as a version 7 study we put everything back through 6 5 5 to equal the optimum under those conditions we would actually have to feed 1.19 grams of meth I mean for EM Cal of me2 equal what we observed in that study of his to get the Optimum's right and that's partly that's just a function of in five six five five we only have bacteria in seven we have endogenous protein we have a protozoa we have full recycling it's a completely diff a calculation so we we have to feed a little bit more here to meet that output the lysine of a signing relationship is 2.7 the 1 again that's what we generated from our own data that's identical through O'Quinn's number from many years ago now the way that works it's pretty straightforward math the cows consuming 60m cows of any and it's 60 times 1.19 so that cow to meet her requirements the real requirements as we understand it would be about 71 grams and Metheny so people will ask me why do I think that's so high I don't I think maybe we got to we're getting better at getting to a true requirement I also wonder about the sistine right we would if we were monogastric species we would be talking about sulfur amino acids and not just a signing so I'm wondering if we're accounting for a cysteine requirement in that and if there's just some exchange and that's where we ended up but anyhow 2.7 times the Metheny gets us to 193 grams of lysine and again if you're using 65 you always want to calculate the defining first because that's what we derive this relationship on and then calculate your lysine don't do it in reverse you won't get the right number [Music]