[Music] so yes it's my pleasure to be here and to be part of this event on variation and via variability because it's a very important topic not only for the industry but also for research and I would like to tell you a little bit about a relatively large collaborative research project called Grain up that was recently finished which is a research project or was a research project that focused on nutritional characteristics of cereal grains funded by the German government and co-founded by the German feed industry and where several research institutions were involved I would like to give a specific focus on digestibility data here and to start with a few statistics just to remind you the amount of cereal grains that are produced and of course part of this is for human consumption but quite a considerable part is for animal feeds and I took a few values here from a German statistics that shows the total amount of cereal grains that were used in one year and the amount the estimated amount of protein and the estate estimated amount of phosphorus that is contained in this amount of cereal grains and what you can see here are the corresponding values for soybean products and for rapeseed products of course they were lower in regard to total amount but look at the values here for the crude protein and for the phosphorus there were lower or not higher at least then the amounts that are contained in the cereal grains that means of course we consider them mainly as energy sources but in terms of crude protein and phosphorus they are at least as important to know about and to be characterized and that is why this presentation will give some focus on amino acids and on phosphorus so the objectives of that research project were to provide more actual data on the reason on the current qualities of cereal grains we wanted to focus on the variation that exists between different grain species but also within one grain species and as I said before one important focus was on determining digestibility values in different species we will focus on pigs and poultry today but ruminants and horses were also considered in that project and linked to that were attempts to improve in vitro and other predicting tools because we are facing the situation of animal studies become becoming more and more difficult to to conduct at least in the you and in the meantime more than 20 journal publications emerged from this project and if any of you is interested in any details and getting copies of this PDF I have them all with me so if you want please approach me in the coffee break or later and like with your USB stick and I can give you all this information so there were a total of as you can see 134 genotypes or quota bus if you prefer that of cereal grains with this distribution among maize barley wheat triticale rye and oats and there was a specific focus and on the genotype derived variation between these grains so the selection was made by owner by our collaborating partners from the breeding industry and the important point is and that is very important when we when we see the results later to keep this in mind that all the genotypes from one grain were grown under identical conditions on the same place using the same fertilization and agronomy details and all these things on the experimental research station of our University and you see the amount that was harvested per genotype and that was actually different depending on what animal species these genotypes were used in and what you can see here is just an impression of what has been done in the field you can see different things one thing is that University is located in a very nice surrounding and yeah you can see that all these things have to be done in a relatively small-scale experimental kind of scale here with the seeding of the grains and then you can see the stripes here of different colors in different stages of vegetation and then this harvesting procedure here also with these very small machines so this has been a tremendous amount of work done by our colleagues from the research station for all these samples so when we come to the results now this is a very simple beginning and it basically picks up what we've heard before and also basically basically confirms what is known for the proximate composition of the different cereal grains here are only the mean values for starch fat fiber crude protein and ash and yeah as we all know the starch of course has the highest proportion here and we do have some differences as expected also in other nutrients I wanted to explain you but I can skip that because we averted before that especially in regard to starch the question of methods standardization whether it is a polar or metric method or whether is an enzymatic si is extremely important especially when it comes to comparing data between different studies so as I said all these information or detailed information for each individual genotype is fully published so we don't need to talk about too much don't need to talk about too much now and if you are interested in that it's all available so when we come to the crude protein fraction now here again in principle we see confirmed what what we know from from all the textbook values and from other studies please note that these concentrations are given on dry matter basis here the crude protein concentration in maize considerably lower than in the soft grains but you can see on the first glance that there also is quite considerable variation within one grain genotype derived variation within one grain or though fertilization and every single standardized when we have a look at the oil or ether extract or crude fat fraction we see the known higher values contained in maize and also in oats lower than in the other grains again some variation and I would like to point you to this extremely high variation here for the maize but that is simply because some so-called speciality mesas were contained in this data pool and they were bred for higher hoyle content for different other purposes not only for feat purposes when we now go a step forward and look to the amino acid profile of the grains this is expressed here as grams per 100 grams of protein in the lysine concentration then we can see a clear differentiation between the different grains we do have relatively low values here for maize and for wheat and the highest for oats considerably higher in the lysine content in old protein but again substantial variation within one grain and this picture is totally different when we take as a next example methionine where we have only a differentiation between maize on one hand and the soft grains on the other end that means that we have depending on which amino acid we look at we have a completely different relative values here among the grains that we consider so the question is whether any correlation could be seen or relationship between the crude protein content of the seeds and the lysine content in the protein and when all data are taken together here you can see the clear differentiation again between different grains and but but there seems to be yeah a trend of a relationship saying that roughly 0.1 grams of lysine are contained less which each 10 grams of crude protein being higher in these genotypes please keep in mind again this is not related to differences in crude protein concentrations that are caused by different fertilization for example this is only caused by the by the difference different genotypes that were used in the data pool and the picture is similar here in the other example of methionine where the relationship is even a bit stronger between the methionine content in protein and crude protein content of the grain so [Music] interestingly and this is from another study this is not from the from this grain up project but I would like to show you this study because it shows some conflicts of interest so to say between lysine content increase of lysine concentration in the grain and digestibility these are data for critical genotypes and here they were produced either under normal nitrogen fertilization regime or with additional extra nitrogen fertilization and when we have a look here at the lysine concentration in the seeds then you can see here a clear effect of nitrogen fertilization on lysine concentration in the seat and these seats were then tested in Sector mised laying hens this is the animal model that we often use for amino acid digestibility studies in poultry then you can see that the digestibility of lysine in all three triticale genotypes the digestibility of lysine was significantly lower meaning that we may get some more lysine in the seat through fertilization but when we have a look at digestible content of lysine in the seats its it was not different so we always have to keep in mind that when when we change one one trait we may have negative effects on another so as I said before we also had a closer look at phosphorus and now we get back to the grain up dataset you can see the total phosphorus concentrations in the different grains again well known that mace has a relatively low total phosphorus content but we also have the differences here between the different soft grains and again variation and here because sometimes people ask how does this look against established table values these are values from the NRC pick booklet were more or less the mean values are relatively close to what has been found in this data set with the exception of barley and but I mean is it is as it always is when you compare data from single studies with values they may fit or they may not depending on where this data come from so we should perhaps not pay too much attention to it and only or always keep in mind that table values can not tell us the whole story more important these days is becoming the question of what proportion of phosphorus is contained in phytate form and you can see here that increase in total phosphorus concentration is associated with a certain increase of phytate phosphorus concentration and this is something that is not investigated in much detail in the or not not considered in the in the tables that are around you usually have only these mean values here and you can see that this not at all reflects the variation that we have invited phosphorus concentration and have and having in mind what big pressure we we have on on phosphorus excretion in some parts of Europe and this is extremely important information for the fine tuning of the diets of pigs and poultry okay so let's now come to digestibility data and what we my co-chair asking whether I still have enough time and this will be on this will will be this will have a focus on digestibility of amino acids and on phosphorus and I would like to start with a few explanations and on how we studied amino acid digestibility here for the poultry side of the story we did that with laying hens and these laying hens were all sector mice you can see how how how big the Sica of laying hens can be and that is a conceived considerable room of fermentation producing bacterial protein that is no longer produced when these removed so for trials were conducted with these Septimus laying hens and each trial investigated twenty genotypes of the gray that are displayed there and they were all conducted in the same way using a basal diet that contained some maize starch and this mates starch was replaced by one of the grains and then linear regression approaches were applied to estimate amino acid digestibility of these genotypes and in pigs we also used a well standardized approach using ileal e cannulated growing pigs you can see the surgically fitted cannula here again for trials in this case only eight genotypes of the grains because it was not so easy to investigate so many genotypes as many of you can imagine and the so-called direct method was applied where the cereal grains were used as the sole source of crude protein in the diet and allele that just was sampling according to a standard protocol that we use since many years in Germany so when we start with the pig results now you can see I've selected a few amino acids just to show the principle what you can see here is the amino acid that the digestibility for lysine methionine threonine and tryptophan for the four grains here and please consider that the y-axis has different scales for the different amino acids and there were very clear differences in digestibility of all the amino acids between the grains and the Rye was very low particularly very low in digestibility it C lies in digestibility of lysine digestibility wasn't on an overall relatively low value here for all grains and but the variation more or less was the same and in the same direction for the different no assets among the different grains and when this is compared to tabled values and here again the pic data from the NSE were taken then it is very it becomes very clear that these values are considerably lower then these older tabled values and the question was and still is what the reasons for this could be and it probably has not so much to do with changes in the grains per se over the years it probably is related to the method and to the better standardization of the methods over time I mean these table these values that were tabled in 2012 they have been determined I don't know the 20 years before or even longer and approaches and essays were not so well established at that time and you can see the values here again for barley from the NRC and from this grain a project but also from other studies this is from Einsteins lab here for barley and it was not only in the grown-up dataset but also in the u.s. work here that digest abilities were lower than the older values and the picture was very similar for wheat were also this more recent data irrespective of where these data were obtained were lower than it then that what is contained in the in the in the older books so there is something going on on the essays now after we have seen this variation the question of course is what could be the reason for this variation particularly within one grain and as I said in the introduction we were also interested in approaches to estimate or to predict this variation and a lot of statistics has been run on this kind of data and of course the question was is there any relationship between amino acid digestibility in this case as an example lysine digestibility and the NSP fraction and this is a total nsp content here displayed on the axis here and I mean you can see for wheat and for barley and for triticale there was no relationship no no significant relationship between amino acid digestibility and NSP content the only significant relationship was detected for ride so this does not mean that such relationship between an espy content and amino acid digestibility does not exist it simply means that we could not detect it in most of the grains here and that is most likely related to the relatively low variation in NSP content that we had within one grain and that again brings us back to the standardized environmental and agronomy conditions that were applied here in this experiment so when we also have a look at the results of the laying hens now it's pretty similar but on a higher level especially for lysine values here for wheat rye triticale and maize barley was not investigated in laying hens again the Rye with dis currently is currently low digestibility values here but overall relatively high values for the digestibility of the amino acids in the laying hens and that is not so easy to compare with other data from literature because what they are hardly data available and if they are available they usually are based on different methodological approaches and we all know that details of amino acid digestibility highly affect the outcome of the experiments but I think we can say that this star these results in principle in principle confirm that the level of amino acid digestibility is relatively high in cereal grains and what I can do is making a comparison with this cereal grain data with some other data that we have generated using the same experimental approach so this the Septim eyes laying hens and when we look at the mean values here for lice in digestibility this is what has been contained in the graphs already before we see this around 80 percent for wheat and maize slightly lower in triticale and much lower and rye and when we have a look at AB seed meal it's also not very high but values are much higher for example here in Reubens and piece that were investigated in this animal model and also for methane in the data are shown here methionine is of relatively high digestibility in cereal grains and even higher than the digestibility here in the in the legumes so again for this data we try to play around and run a few statistics on relationships between digestibility and certain specific fractions analyzed in the grain and tried multiple regression analysis in an attempt to predict amino acid digestibility and for this other data that were chemically analyzed were grouped in these eight pools here and then the computer played around with this data and what I can show you here is that using methionine again as an example when you select here these different number of very of variables in in these multiple regressions then a relatively high accuracy of prediction can be obtained for the estimate of methionine digestibility but have a look at these fractions that are considered here this is something that cannot be done under routine conditions and that means that this is at least at this stage an artificial situation and cannot be applied for predicting amino acid digestibility in laying hens so combining several several variables that were analyzed analyzed may lead to an acceptable accuracy of prediction but the selected variables are neither useful to explain this variation nor are they of practical relevance because too difficult to determine routinely another interesting question was how does it look like when we compare the peak and delaying end data this again only lysine as an example the laying hand data here and the peak data here we can do this comparison because these digestibility tests were done with the same batches of feed and what you can see here is that as if you take all the data together there seems to be quite a fair relationship but when you look into each individual grain here then these relationships are not existing so you can see the the a square values here for wheat critical and dry only when all data are considered together relationship can be predicted but again we have considered here we have to consider here that the range that we have had in this data set was relatively low and last point here as I said we also tried to develop or improve in vitro assays so on all the samples the boisen and Fernandez based in vitro nitrogen solubility was also applied and measured here and this is for the pig data and this is for the laying hen data the relationship between this in vitro nitrogen solubility and the determent digestibility in that case again for lysine and there was no significant relationship with the exception of the right data here for all the other grains and you can also see here for the laying and data that this is so far not a really satisfying relationship still more work to do on that so a few words finally about digestibility of phosphorus again for pigs and for in this case broilers for the pig we we have since many years a standardized assay to determine digestibility of phosphorus on total tract basis or feed raw materials and for broilers we have the WPS a protocol to study procedural digestibility and when we immediately dive into the data for the pigs this has not yet been published done by the colleagues at who at Bonn University wheat barley rye and triticale again eight samples investigated for phosphorus digestibility and you can see the mean digestibility is 60% for wheat with this variation among the eight samples barley in the known range rye surprisingly low taking taking into consideration the high phytase concentration of this grain and triticale somewhere between wheat and rice so again we see also for phosphorus digestibility this high variation here and this could not be explained by phytase activity alone and it would not be explained by phytase activity and phytate proportion together and for the broiler data yes thank you we have the pre-sequel phosphorus digestibility here for 8 samples of decay and for eight samples of wheat if we look at the triticale samples here the the orange ones you can see that they are some around what is it 80 percent for some of the samples and others were down to 60 and the wheat digestibility was lower it was around 40 between around 40 percent for 4 samples and between around 60 and 70 percent for some other samples and I choose this way of expression depending on plant intrinsic phytase activity to show you that well there was quite considerable variation in in the intrinsic phytase activity of the seed but this intrinsic fighters activity of the seed could not at all be used to explain the differences in phosphorus digestibility found here so there are there are other factors relevant and intrinsic fighters activity embroil us is not so important so just towards the end only to mention if there is anybody in the audience who should be interested in ruminant research ruminants were also considered in this project so there are data available on ruminal degradation of protein and starch and you can get this data if you want so to conclude the results from this animal studies allow for a more precise consideration of protein and phosphorus values in diets we had seen in the previous presentation that there is a lot of variation in chemical composition of all the feeds that we use including grains and we have to consider that we also have this variation on the digestibility side so running chemical analysis is not solving the problem alone we have seen in the pics that amino acid digestibility differed largely between the grain species overall the values were lower than reported in feedstuff tables but are in agreement with more recent publications this points towards sa relevance here enhance and employ less we had again and also differences in amino acid digestibility between the grain species and among the genotypes the amino acid digestibility is relatively high in laying hens except dry and phosphorus digestibility also is relatively high and so far and this may be the little bit well perhaps I don't know frustrating or disappointing message that so far we are unable to explain this variation of digestibility data simply by chemical characteristics of these grains but that is something we can continue to work on with this I would like to thank you for your attention and look forward to your questions [Applause]