 I work at Oros University for who doesn't know me. And if I can define myself in any ways as a food scientist, I would say that I am an addicted to physical techniques. So I have a problem. I like to actually measure things at different length scales. And I do, I think Nicholas said it really nicely yesterday. We like to identify a problem. And not necessarily focus on a particular technique, but we try to find what will help us in understanding the problem and figuring out what are the mechanisms around that particular effect. So my expertise is in ingredients interactions and structuring of foods, both in terms of structuring during process, during digestion, any type of structure really intrigues me. And of course, my expertise is in proteins and colloids. So this is the focus of what I will say today. And I will try to add a few things from what my colleagues said yesterday. But as I said, more from a user perspective than an expert in the techniques that are highlighted during this conference. So first of all, why am I involved with the Northern Lights and Food? Because as I said, I'm intrigued with powerful techniques. And I do believe that the techniques that we can access to or we will be able to access to when everything is up and running are of very good use to food. How? I'm not sure. And I think that I probably share this vision with many of my colleagues, both in industry as well as in academia. And I think it's important for us to really understand. And I always laugh. I put this little table here with the name of the actual facilities. And this is already the first barrier. What the heck is that? I don't understand why people have to put nicknames to these places so that it's yet one more barrier to me understanding what they are. So it's just a completely new world for us. But it's extremely attractive to be able to use something that is so powerful and will allow us to really understand the structuring and the dynamics at a detail that we could not get with any other method. So that's why I'm involved in this group. And I try to hang on as much as I can because at the end, I know that this is going to help me understand my systems and also to contribute to the challenges that we have within the new sustainability challenges that we have. It is a really good time to be a food scientist and a food researcher. And we have heard that in the panel yesterday, too. Because we have all these new problems related to new food systems that, of course, before we always did really good food science work. But now we actually have a meaning and a meaning that is more than just improving or excelling in some processes or just improving our yields. It's really about changing the food system and becoming carbon neutral. So less is more. We have to avoid waste. We have to understand better our byproducts. We have to learn how to use new sustainable sources. We need to design new diets, which includes the use of proteins, different blends of proteins into a system. And we are not used to that. We like to use commodities. So we are either a dairy scientist or meat scientist or food chemist. But we do not really understand how proteins interact with one another. We do not want to over process. And so better safety and story does no longer apply to us. And we want to create clean labels. And that's not very easy with some of the functionalities that we actually want in the food. So what do I work with? I work with processes. And I think that I have myself in my research program a pretty big challenge. We try all to have some huge impact in to become carbon neutral. We want to have changes to current processes and try to use less energy, less water, less waste, less cleaning chemicals. And I don't know if you heard yesterday, Peter Weise, at one point he said that the consumer will not see the change because the consumer, we are not used to seeing incremental changes. And if I have to say, actually, I think that the consumer will have to see changes because we cannot reach carbon neutrality only with incremental changes. We actually will have to have disruptive changes to our food system. And so although this is super scary, especially for the food industry, because it will mean new equipment, new capex investments. And of course, you cannot make a capex investments that require two or three years and a lot of money if you don't really know what's going on in the detail that you really need. And this is why I think it's important for us to give them all the information on structuring that they can possibly have. So one more point around this is that, of course, when we are talking about disruptive change, we're also talking about new sustainable ingredients that currently we're simply processing the same way as we're used to. Because that's how the equipment is in the plant. So we are going to basically make a soy milk, a almond milk using the same kind of equipment that we use, that we are used to, and that we know very well. And perhaps that's not the right way to do it. And also we have to be able to target these processes to the properties of the proteins, which are obviously very different than the properties of the proteins that we know of. So without further ado, I will tell you that as a person, as a researcher, that works in a very well-equipped department where we can do any type of analysis from the molecular to the microstructural to the supramolecular, as well as looking at processes and looking at mechanisms during processing, it requires me a huge activation energy to try to figure out when do I need to come to do it? Because truly, is it worth it? Is it worth it to when is it important for me to really understand the details of this? And I think that probably I share this with many of my colleagues in industry that say, well, but if I can understand the system with other techniques, why should I use the large-scale facilities? So this is really where I come from when I was thinking of giving this talk. And hopefully I'm being a little bit provocative here. And as you can see, I am part of this team. So I don't really truly believe that the activation energy has to be that high. But of course, it is an issue that we have to resolve. And I believe that to resolve this issue, we need education. We need to really understand these techniques and allow people like me to feel comfortable enough that what we can achieve is actually useful. OK, so I am not going to talk more about structuring. I think we've heard it for two days now. What I want to say is that today, my talk is more related to protein systems, as we have already heard from Francisco about carbohydrates. And we've heard quite a bit about emulsions and the fats from Alejandro. So I tried to focus a little bit more on protein and especially new proteins. So without further ado, 2050 dietary shift, at least 50% of our diet will have to be plant-based. And with that, what that really means is that, of course, the animal proteins will still be part of our diet, but they will have to be value-added and therefore very high nutritional value and part of the sustainable diet. We have to, therefore, improve both our current systems as well as develop new systems. And when we look at proteins pivotal to our diet, we cannot only talk about the essential amino acids and how these little legal blocks actually fit in our body, and then we break them down and we absorb them, and yes, faster or slower. But I think what is very important here is that peptides and how we break down the peptides and how the protein structure or food are the critical points of developing new diets. And let's not make the mistake of just making food that is tasting good for the consumer today without actually planning for the best outcome from the nutritional standpoint. Because if we don't do that and we just try to just please the consumer as they are used today to their food, we probably will make a mistake and we will not provide them with the appropriate diet and we will not resolve the rest of the picture, which is obesity and malnutrition or undernutrition. So how do we use new proteins? And how do we use them within the current system that we have? And how can we create novel forms of foods? And let's just try to be a little bit innovative and not only think of fitting the new proteins to what we have today. And within this protein debate of the new proteins, I think we also have to remember that there is a lot of discussion on the protein bio refinery and that perhaps using purified ingredients or isolated systems all the time is not the right way to go. And that because as we move along the pyramid of the isolation, we also create very high resource intensive processes. We have to purify. We create a lot of byproduct. So when is it appropriate to use isolates? And when it's appropriate to use less refined ingredients is also part of the discussion when we are reformulating or redesigning the foods. But this creates a huge set of problems because if we, of course, if we are using an isolate, we're spoiled because we know exactly how it will behave and we can mix it and match it with other Lego bricks in our formulations so that we can derive the final ingredient, the final food and we can control the process to the tea. But unfortunately, if we're using less refined ingredients, that will not be so useful. And this is quite challenging, for example, and this is something that Anna and I want to work on because in the non-refined ingredients, what do you normally have? You have proteins, you have fiber and you have to really understand how the fiber can be perhaps modified, maybe with the aid of enzymes or fermentations so that we can create functional fibers as we move along into the process of production of the ingredients. So a lot of in-situ processing, in-situ modifications are very targeted. And I think we are just at the beginning of this journey because if you think about it, yeah, we are using some enzymes, we're using enzymes commercially available enzymes that are full of different things in there, very low in purity. How can we really target a change in pectin modification? Or how can we make a different blockchain distribution of methylated esters in a pectin if we are using commercially available enzymes and we don't really understand what we're doing? So it's an art versus science, but I think we have to move it towards science. And I think that knowing exactly the changes that happen during these processes at the molecular and supermolecular scale will be very important. And this is why I think that the toolbox that we are growing now as a team in Northern Lights of Food will provide us with this kind of information, both in spectroscopy as well as in imaging. And it will be allowing us to really fine-tune ingredients changes. So let me give you a couple of examples of my puzzled mind when I actually read some of the papers that I see that are related to the work that is done in large-scale facilities. We know the structure of soy proteins. We, I think we have worked on soy protein probably longer than any other plant protein, because it used to be a byproduct of the oil industry and actually quite well-funded. And when there's funding, there is research. So we know quite a bit about soy proteins. And in the 80s, already, we have had an enormous amount of scattering work done on plant globulins. So this is just an example of a paper from 1985 where there was a scattering of 7S and 11S proteins from P. And of course, all the time, these type of curves were developed to understand the structure of these proteins and with complementary techniques, such as proteomics, STS page electrophoresis, even mass spectroscopy, light scattering, spectral analysis, microDSC to see if the proteins were still native. So with that in mind, basically, they all kind of sort of do the same thing. They are globular. They are about four, five nanometers in size, and of course, different in molecular weight, depending on their subunit composition. But now, here is my puzzled mind. When I look at soy milk, the soy milk particles look like this. And they are much larger. They contain fat. And this is really what I process. Is the information, of course, it's nice to know how is the protein, the subunit, looking like. But is that telling me anything about how this particular material will behave? And they are colloidal particles. They increase in viscosity as a function of concentration. And they haven't been studied in much more detail than this. So that's where I think there is a little bit of a gap here that I think we need to fail. The other puzzling thing about looking at plant proteins in a fundamental sense is that when we, this is a micro-DSC of soy proteins prepared either in the lab or in the pilot plant. And you can see that while in the lab, you can see both the 7S and 11S denaturation peak. When you're preparing in the lab, you don't see these peaks anymore. So obviously, you have already had aggregation, or maybe you just cannot dissolve these systems anymore. So lobility curves as a function of which sample do we have. This is a native soy protein isolate. This is made in the pilot plant, and this is denatured in the plant. And as you can see, if I have to heat the powder to actually dissolve it, so how do I deal with this? Because all the work that we do in the lab actually looks at non-denatured systems. And here are some electron microscope pictures of proteins that interface, of soy protein interface. And of course, when it's native, you have a very nice thin layer at the interface, but then you have all these aggregates as well. And many times now when we do these studies in the lab, my students say, well, should I centrifuge the sample before I homogenize, or should I not? And of course, if you centrifuge the sample, you will get only part of the protein population. Again, I'm just being provocative here. It's not that I'm saying, well, I don't have a solution to this. Very recent paper on the studies of heat treatment of soy proteins using X-ray scattering. And as you can see here, they basically looked at an acidic and a basic treatment, alkali and acid and different heating treatments, and looking at unfolding of the protein. And it's very nicely seeing the unfolding of the protein. The temperatures of treatment and the pHs are extremely high and low. And the times of heating where you're actually seeing a change in the structure are five hours. Who's treating proteins in food for five hours? That would be a very expensive way of doing it. Another example here of another very recent work on rice glutinine. Rice glutinines are very difficult to solubilize. So, OK, we have obviously looked at some soluble fraction of it. And the SAX profiles very clearly show a loosening of the structure with eating treatment. But look at the type of heat you have to apply to these proteins before you're actually seeing a structural change. You have to treat them at 100 degrees for 60 minutes. So these are just very interesting studies, of course. But how do I apply them to my processes now? One last one on plant proteins that I thought was quite interesting. Again, this is quite new too. They actually went one step forward and they managed to demonstrate that oat protein is an ellipsoidal shape. So it's very cool to be able to see with these techniques not only the size or trying to model the size, but you can also model their shape. And that's very useful if perhaps we can do something that is a little bit more relevant. So how can we contribute to these field of knowledge and as food scientists? I think that's quite important. So here is a picture. It was one of my favorite pictures from one of the PhD students at Wageningen. I always love when scientists are also artists. I just cannot do that. So what is really nice from this picture is that it's showing how challenging is the molecular architecture when you're looking at extraction of plant proteins from different materials and how proteins will be interacting with the other components in the system. And man, I am a dairy chemist by training. I am spoiled because in dairy, although we say it's a complex system, at least we can separate things physically. They are different sizes and they're quite separated from one another. Here I think that the easiest system that you can look at is legumes that have starch granules that are separated from the protein bodies. And you can actually do some dry fractionations with this. But when you're going to plant cells like microalgae or green proteins from leaves, for example, you have interactions with chloroplasts and with polyphenols. And I mean, the protein quality is very difficult to maintain. And here is the solubility of three commercial pea protein ingredients as a function of pH. We just did this study because we wanted to pick which protein to use for our tests. And as you can see here, the two isolates have a very low solubility in a buffer system. You have to beat the heck out of them before they start solubilizing. And this is because they are being sprayed dried and dried. The green line is the dry separation technique. So what I was saying here that you can take a legume and do just a dry fractionation. So you never actually extract heat, filter, dry, evaporate. This is just a dry separation technique. And of course, the protein status in that system is much less denatured. And it can be solubilized. But let's not forget, though, because a lot of times people talk about low processing or minimal processing. But we also have to think about food safety. And it's very interesting how sometimes we kind of forget that most of our food needs to be somehow heated at some point of the process. So when do you heat it is now it's, I guess, a question, not how do you eat it or sorry, how do you eat it and when do you heat it but not necessarily if you heat it because you do have to somewhat have a safe product. OK, so I think Ramune later will talk a little bit more about the cellular architecture. So I will not steal her thunder. But here is an interesting paper that I found that I thought would be interesting to share with you because it was carried out using synchrotron FTIR imaging. And I think this could be quite an interesting technique for us because it allows us to see differences in this case, differences in the roasting of the seed. And I'm not really like, of course, I didn't go too much in detail on exactly what they found. But I thought it was quite interesting that you could see the denaturation of the protein already as the seed was roasted. Oh, sorry, I went all the way to my end. So let me just now. OK, so what can we contribute? I think we need to study aggregation. We need to study behaviors of the systems. And we need to do this in conditions where we have thermal mechanical treatments. And what do I mean by that? Shear, concentration effects, moisture levels, all these, of course, temperature and environmental conditions. Because if we do not do this, we cannot understand how these systems will behave in realistic conditions. And maybe I have to debate this a little bit with my Unilever colleague that said, well, I want to have these systems on the line. And I agree. It's nice. But I think that if we understand the mechanisms at a very in detail, then we can perhaps relate to this mechanism to some more simple online measurements that we can do. So that's really what we have to try to do because right now we do not have the right machinery to understand this in the plant. And it is important to spend different length scales of these processes because we do not know which critical length scale is linked to which property. And I think that's critical. One very cool study of my good friend from New Zealand, Martin Williams, Bill Williams, I thought this was a good example of how we could use neutron NX-ray and the famous contrast matching that everybody talks about to really start to understand interactions between ingredients. And he's done it very cleverly looking at complexes between beta-lactoglobulin and pectins. And by basically matching, he was able to model better what is happening when you're adding a certain amount of pectin to the system and a certain type of pectin. We go back to my initial thoughts on less refined ingredients and how we do need to understand how fibers and all sorts of soluble fibers and carbohydrates interact with less refined proteins. One other example from this complex world and how people are adventuring into something that is a little bit more complicated than a one single system is the work that is, again, done in Australia on micro-gel stabilized emulsions where they basically prepared the micro-gel particles. In this case, it was beta-lactoglobulin particles or whey proteins particles. I don't remember. But they really managed to see the thickness of the interface. When we do this with pure plant protein systems, and we use ellipsometry in the lab, we have thicknesses of about 2, 3 nanometers. And it's very strange because I think that if we actually used the entire system, then we would have pectin stabilized emulsions and much larger scale. So we need to know the chemistry. And then we can study the dynamics. And we need to get detailed structure information. I hope that this gives you an idea of it. And of course, we can look at all sorts of less refined systems. I don't want to go too much in detail on the microstructures here of what we see. But basically, I just wanted to give you an example of what happens if you have a rapeseed and you try to make rapeseed milk, whatever that, like an extract directly from the rapeseed. Depending on where you're starting, you can start the pH 7, or you can start the pH 5, you can start the pH 8 and do an extraction. And then you can also fluctuate the pH. So you can start at 8 and then go down to 5 because the solubility will change. But as you can see here, also the complexes of proteins with the oleosomes and how much oleosomes you're carrying with the protein will change by doing these pH extractions. And everybody is doing these pH extractions now. But they just look at amount of soluble material, amount of protein, yields. And we do not know much about the supramolecular structure of these complexes. We need relevant environments. And here are just a couple of examples of ingenuity in the creation of relevant environments. Eleo Gilbert likes to work with starches and he has created a little RVA system to look at how starch gets gelatinized in situ. And it's quite clever. And honestly, I think it's quite low budget too. So it is possible to come up with some of these systems and really see the systems in scale. But what would happen if you actually have complexity in this? Then you start having problems with modeling with really understanding what these curves actually will tell you. This is a case, a very recent study that showed that if you add more and more P protein to a dough matrix, the dough changes. And quite a bit. So the red spots here are P proteins into this dough matrix. You can see the microstructure is completely different. So what is the effect of these novel proteins that we're adding to the starch into the star gelation? And these are important problems. And of course, we cannot study these systems in solution. So I don't have the answer to that, but maybe this clever group can think of nice ways to do this. And again, protein mixed proteins. This is a whey gel with fat globules. And this is the same formula, just changing the protein matrix. Lupin only, P protein only, O protein only, and mixed whey. This is reality. So how do we, I want to know what's going on in the systems. And I think that we have a long way to go. Another clever in situ system that I think would be quite interesting to use. We saw a lot of talks about fouling and membranes. I don't want to only know after fact, I want to know what's happening while I filter. So how do we do this? We can do this. But of course, it will take quite a few brains to get in together because by ourselves, we cannot do this. How do we hydrate powders? We have studies on neutron that show that maybe we have some differences when we hydrate a powder or in sands because, of course, you don't have the water signal. But how do we dehydrate powders? And we heard people talking about spray drying yesterday. And I would say, how do we hydrate powders? But we actually don't go to solutions. There's a lot of processes where we hydrate powders, but we never get to a soluble phase. So what happens to the proteins when you are in conditions where you have only maybe 0.5 of water activity at the end or 0.4 or 0.3 of water activity? I'm not really sure we actually understand this. Militropy of systems and the fiber formation in proteins. I am particularly interested in mozzarella, of course, but or in cheese substitutes. But of course, there's all the area on meat replacers as well. And I think this paper here that Gwen probably will talk about, so I won't say much about it, to me, is a good stepping stone in the right direction because it was able to already give us an idea of how we can get submicron information on fibers and how to connect them to tensile tests. And my last thought is that when we look at systems like that, of course, we cannot only look at submicron. We also want to look at what happens during modifications or textural strains to the system. And I think that we could learn a lot from what's happening in Stephen Hall's lab on tomography because they have done this on packaging for a long time. So of course, it's not really contrasting the same way, but maybe we can become clever about it. So with that, I'm finishing my talk and I just want to say, hey, let's learn from each other. We cannot do this alone. None of us can. And we need to have a very good understanding of the system and we need to build suitable environments because for food scientists, if we do not have suitable environments, we can only do so much. And with that, I thank you for giving me the time to speak. And I pass back the airwaves to Anna. Thank you, Milena. Thank you for excellent talk.