 Okay, excellent. Then without any further due, let me welcome you to this animal genomic special interest group that we are launching today. Many of you are familiar with this channel because it all started under the Bovreg banner and the Eurofang as a Eurofang special interest group on pipelines. At the end of Eurofang, of Bovreg, we decided to broaden a little bit the channel and to make it a bit more inclusive beyond Eurofang. And lucky enough, NFCOR, thanks to Phil Hewels and Sven Nanzec, we were able to have the support of NFCOR to launch a special interest group. And that is what we are doing today. So this new channel is really meant to support anyone doing animal genomics and using NFCOR or next flow pipelines. It's entirely self-managed, so that means all of us are entirely free to propose new topics, to organize this thing as we see fit to best serve our community. And to put it very simply, but Phil will give a bit more details later, but to put it very simply, this channel is typically the kind of place where you would come and say, I am doing this kind of analysis. Is there anyone doing the same type of analysis? Which kind of tools are you using? Which kind of pipelines? And there we will have a small community using the Slack channel, hopefully exchanging about tricks and trying to maximize harmonization in the kind of analysis we do. Now, we hope to have recurring features, especially on tool discussion and this kind of things. The most recurring feature will be a monthly seminar. It will range from at-hand topics, anybody among our community wanting to discuss some aspects of genomics. We also hope to have forward-looking talks, like the one of Krista today, and starting from the fall, a series of talks about state of the art in human genomics, because human genomics is often a driver of what is going to happen in animal genomics. And in general, any of the topics you guys will find of interest for this crowd. So, I'm really, really happy to welcome Krista Kuhn, who is our first speaker. Now, Krista was just appointed as president of the Friedrich Löffler Institute, so that's something that happened just at the end of Bovreg, and I think it's a very, very nice development. And Krista is well-known to many of you for her work in animal genetics. She's been highly involved in QTL analysis in Cal, and has been a member of the International Society on Animal Genetics since 1990. She's also an editor for the Scientific Journal Animals since 2006, associate editor for genetic selection and evolution, or genetics and breeding. So, she's someone very, very well-known in the field, and I am sure that her talk on the future of genomics for found animal will be of interest to all of you. And without further ado, Krista, the floor is yours. Thank you, Cedric. Can I share my screen? I think so. Jose. Jose is the host. Yes. Can you see my screen? Yes. Your presentation mode. Let's see if I can. Perfect, Krista. It's not in presentation yet. Almost. Is the presentation mode no for you? No, no, no, no. Which one is it? It's the one where we see the slides, you know, the one where you work on your slides. Better now? Exactly. Now it's great. Now it's perfect. Okay, because for me it's the opposite way around, but okay, I'm working with you. Yes, that should be. If you're on the map, that's exactly the confusing thing. Okay, okay. Thanks, Cedric, for the invitation, and actually I feel a little bit, or embarrassed, a little bit shy, because all the competence in the field of NFCore and Bioinformatics is on your side, and I'm really, really, well, only a side liner in this field. Thank you for introducing me, but I really don't feel like an expert, but you asked me to give a presentation, maybe some future aspects of animal genomics, and I suggested to go even a little bit beyond. And maybe I can give somewhat one or two ideas about what might be needed in the upcoming future. So let me first give you a brief overview of what you can expect. So very briefly, you mentioned Pritik Lofler Institute, I've very briefly introduced where I work, then I go to what's Bofreck and Fang, although I see from the audience that most of you already know about it. I wasn't fully sure about the audience, so I thought put in some slides on this, then generally how I think the idea of workflow managers, next flow and NFCore has taken over in our field, and then the core will be a future expectations for and from NFCore, and then maybe also challenges for the animal genomics group and beyond. So let's briefly start, if you go into the center of this little island, this is where I'm working, at the moment at the world's oldest virus research institute, and the very far east, northeastern corner of Germany. And to give you a background why I might be interested also in the future about NFCore, we are an independent federal authority, so directly under the Ministry of Agriculture and Nutrition, but mainly we are working on animal diseases, animal welfare, animal husbandry, animal nutrition and farm animal genetics, and these are all the fields where I think animal genomics will have an impact in the future. And this is also because some of the ideas I got from my colleagues who are working on viruses on animal science, including farm animal genetics, animal welfare, husbandry, and animal nutrition, where I think a lot of the future demands will probably come from in interdisciplinary projects. One special feature we have, which is rather unique, we are one of only four institutes worldwide. We can work with BSL4 pathogens that's up to things like Ebola or a Crimean Congo virus. So we are really a very high-security laboratory, not only in terms of IT, but also in terms of other biological viruses. As Cedric mentioned, things started with Bovrak. Bovrak is a Horizon 2020 funded consortium as a contribution to the global farm, and one of the objectives was to establish new laboratory and bioinformatics tools because we wanted to annotate the functional genomic elements in the genome, particularly those that are highly relevant to, there's some background noise. Is this somebody speaking up? Because at least in my speakers comes up. Everybody's muted, except you and I right now. So I'm not sure where the noise comes from. Okay, I go ahead then. So we wanted to annotate the function active regions in the genome, which are relevant to modulate the plasticity and variability of the farm's animal genomes. So it's about the annotations. There's a lot of basic research, but also then to channel this knowledge, to develop prototype models, to integrate biological knowledge about the regulatory variation in genomic prediction schemes. That's already the application side of this basic knowledge. And also we wanted to provide training and dissemination of these tools. So Cedric knows that right from the beginning we thought, how can we best tailor our data analysis that it fulfills our needs? Because we have the challenge that we had many different layers in the functional annotation, including the transcriptome, chromatin excess, we want to map regulatory regions, go for the methilome, epigenetic response, the EQTLs and many, many different layers for many samples, even on top of each other. So this definitely required a lot of standardized, reproducible pipelines, which in the end can then also be further exploited and used, for example, for biology, informed genomic prediction, for the real applicants, the ones who are our stakeholders. Animal genetics is not about, genomics is mostly about animal breeding, maybe animal husbandry, which people who are not really not a deal about the main shape of the genome and maybe not even interested in the genome function itself. So the driver was, how can we best develop pipelines and it was quite obvious in the beginning, we had three different projects, cluster projects, BOPRAC was one of them dedicated to found, to the cattle, whereas we had gene switch on the monogastric like pigs and a poultry, and we also had the fish community there with aquafine. And they, I think we all faced with similar challenges, because in the end, once the DNA is there and the essays are there, the data analysis shares many, many similarities across species. And this I think was also one of the things which persuaded the projects in the more recently funded period of Horizon 2020 in Europe, for the Ruminant, Roni, Mo, and RumiDen to join this, because the bioinformatics group, I think was maybe besides the ones on genomic prediction, the one, the group which shared most common interests and also most comparatively and interactively collaborated. Why was this concept of workflow managers so interesting? I think the main things also for future application is that it's reproducible, containers and workflow managers in general, they enable that data analysis can be scaled. So we have one sample, one dataset, but you can have hundreds. And on the other hand, as reproducible so that people can, in the end, reproduce the deposited datasets, and then on top of those continue with their analysis. And the other element is, I think it's also quite general that it works with containers, which contain all the necessary elements. And when you have many, many different, I think I don't have to tell this to most of you, but people outside often struggle that the main work in pipeline analysis is converting one structure, one data format into the other one. And streamlining this has a huge advantage because it saves time for the subsequent steps which we wanted to conduct. I felt that Nextflow and NFCore made quite a development during, from the start, when I think we didn't even have this DSL2 module option. I think in the start, the focus was mostly on the developer side, but now I've just got this from your website. And of course, you put into a view also, facilities and end users. And I think for my talk, for the next minutes today, I would like to focus on these end users. I don't see myself as a developer, and I can't speak for facilities, but I think this move forward for users is, I think, is a major leap and will probably, at least from my perspective, as a non-biometrician, one of the, will be one of the major impacts that NFCore and Nextflow will have. So in the meantime, you have many, many pipelines developed already for users to take over. So what are then the expectations for and from NFCore? You probably might have had a look into this FANG 2.0 paper, which we published several years ago, which a little bit depicts the future of animal genomics in the, I think, in the short term. So that's the major things which will happen, which will need data analysis. This is definitely the FANG GTX project, which integrates the different expression EQTL analysis with information on functional annotation and also information about features within cell types and tissues. We will also see large phenotype collections, so many data from many animals that are tested in well-described environments. And this is, I think, also a change a little bit to what the core animal genomics work is about, where we mostly deal with very few animals. Now we see an extension to a large animal phenotype. In vitro biorepositories in the field of more and more concern and more and more public concern about how we conduct animal research. So in vitro research will definitely be a major challenge for the scientists working on wet lab experiments in the future. We will have the idea that not all cells are the same, so the single cell atlases and all the the technological developments we have, this is a very, very dynamic and fast, the emerging fields. This also will then contribute to challenges. We will have high throughput genotype and throughput in vitro systems trying to evaluate potential genomic variants with potential, which potential impact might be associated. And last but not least, the pan genomes and the comparative genomics. So I think this paper really depicts, depicted the, from my perspective, more immediate challenges or perspectives of animal genomics. And I would like to spend a few minutes on what I think might be the ones beyond this initiative, which is to some extent or a major extent, brought forward and also pushed into the future by the Eurofound initiative, the research infrastructure project, but also the tank to the 2.0 initiative on the global scale. So from my point of view, we will face several movements into the future. The first is that for in the past, we also based most of the analysis starting from one reference genome, the species. So the first one, then UMD3, ARS 2.0, 1.3, and now ARS 2.0. So it's the frequency of the change in the reference assembly, I think, is more and more speeding up. And at the moment, Bob Rack, you know, Daniel Fisher, we discussed, shouldn't we redo everything now with the most recent ARS 2.0 compared to the ARS 1.3, which we used for all Bob Rack work. So redoing all the analysis we did, which comprises hundreds and hundreds of data sets, will be definitely something which will require a huge workload. But I think the real challenge comes from the pan genomes. So we will face that in the past, we had this one reference genome, but we will see that we have a dynamic pool of pan genomes. Today, I looked into the NCBI, and at the moment, we have 27 cattle genomes, for example, deposited. And we saw that, for example, this nice paper from Hubert Haustru, from the ADHDH, demonstrating how nicely pan genomes can contribute to the phenotype mapping in our species. And so we will probably go into the era of a dynamic pool of pan genomes, which have to, which has then to be included in the subsequent steps of data analysis. Then we have the question, currently, we took hours for granted that we have one genome per individual. I think this is really, well, something we took for granted. However, for example, immunologists, they're already well, right now, well, that, for example, immune cells, they don't have a single genome, they change. And there's some fluidics also with response to endogenous retroviruses, which then creates changes in the genome between the different cell populations. And finally, we definitely will see that there is no such thing like junk in genomes, which altogether creates an increasing complexity as a starting point for analysis. So this is all something which I think we will create a lot of work for the future of animal genomics per se. But I think we will see more and more the demand to go beyond. Many, many people will use our genomic technologies and our knowledge in other disciplines. And we have to work interdisciplinary because the increasing demand comes from people who have not that much knowledge about animal genomics, but need the information. This is, for example, in the field of physiology. Increasingly, 10 years ago, there was almost no paper with any aspect of, let's say 15 years ago, no paper, at least in veterinary or animal science physiology with a lot of genome work. Now transcriptomics, epigenomics is a commonplace feature in our physiology to work. We see that veterinary medicine moves into the field of animal genomics. I'll come to this in a minute. And also all the people who work on biodiversity, who in the past many times worked on more of the structural or phenotypic features, they all need easy to use approved pipelines. They can't spend months and years to develop pipelines to get themselves knowledgeable to how to integrate all the different things, spend lots of wasted time, these experts for pipeline establishment, they want to do the to go through data analysis per se, but they need easy to use approved pipelines. For this, they need reasonable parameters and options. So which different parts of pipelines would be reasonable to glue together? They need a basic characterization of the deposited data sets to pave the way for future application. And as I said, this will need modularization. And that's why I was really very, very much amazed by this development to have this DSL to modularity option. I think this is a major leap towards application beyond the core genomics data analysis, because we're probably able to achieve progress to integrate multidisciplinary approaches, which I think bring the real information from the data sets we put together. Then there's some aspects where I think also the next four and NF4 community can have a substantial impact and create benefits beyond the scientific community. Let's say the basic science scientific community. On the one hand, we have the industry interests, for example, in the field of biology informed breeding, but we also have governmental public interests. For example, when it comes to the endogenous resources and obeying Nagoya rules, so where for example, governments, they don't want to share sequence data or industry, they don't want to share genotypes. And this next flow NF core philosophy of also modularizing and sharing paths between partners would also enable to have embargoed private data sets to be integrated into common data analysis. And I think this is something which should not be underestimated. It will make available data sets properly, which at the moment at least will be very difficult to include into large network projects. And for the perspective of breeding, once genomic and able breeding is really put into practice, the breeding organizations are the beings to be evaluating evaluating centers, they need standardized versioning to create reproducible data. For example, when one of the components of a pipeline is change, they can't frequently follow up a change again and again and again of their evaluation pipelines, they need standardized reproducible core pipelines to integrate biological knowledge into their evaluation routines. Because for example, they also do some rechecks to go with confirmation populations. And if the initial steps frequently would change, this would create a major problem for them. A new future in mind, I apologize for this in my new field of veterinary diagnostics. I learned that there is a very strong need for standardized diagnostic pipelines. There's a lot of accreditation and reference handbooks, which sometimes even at EU level have legal binding procedures how to do some of the diagnostics. And for this, for example, sequencing of to identify virus strains, phylogeny analysis or antimicrobial resistance genes, I just brought here a very recent picture from the highly pathogenic avian influenza, which currently is some quite popular or unpopular in the US, because it also concerns a dairy cattle. Their speed is essential. You can't put together pipelines and it takes three weeks, four weeks, 10 weeks. And on the other hand, it has to be reproducible and easily and partly reproducible. And workflow managers established pipelines would fully fulfill these needs. They could be standardized. They would be in place easily and would be reproducible worldwide. So speed and reproducibility would be things which are very essential to these application. And I think this is something which the animal genomics community together with the veterinarians can achieve. And I said, as I already said, this need this modernization process of integrating these multidisciplinary approaches. To some extent, however, you might also think, how does this fit with commercial alternatives? Because, for example, in the field of veterinary diagnostics, but also in many other fields, you know that more and more once things become popular than also commercial enterprises enter the scheme. And you might discuss between yourselves, how do you find yourself placed in this intersection? And finally, one of the challenges I would like to bring up for the animal genomics NFCORE community, which is on the one hand, you want to make progress. And it's a very dynamically evolving field. On the other hand, you also want to keep reproducibility and standardization. How long do you want to keep, for example, next for versioning? How easy would you like to do this? Would you like to make it? How long would you want to keep these versions alive? That's probably something which needs to be discussed. Then the definition, who do you see as your main customer or your key interest groups? You already brought this on the website, these three different layers, the developer, the user and the facility. So what is the main customer, the applicant or the developer? So what would be your main future driving force? It's also maybe for this new NFCORE Animal Genomics Group. Then the question, how do you include new technologies that comes to artificial intelligence? Or when there are key commercial products, how do we place ourselves in this area? And then something which maybe some of you already encountered, there's always the concern about IT safety, at least in my institution, maybe because it's very governmentally close. The core IT frequently complains about containers and use of containers. Then one thing which is more on the applicant side is, on the one hand, we want to make the application as easy as possible. On the other hand, we have also to be a little bit concerned about unreasonable analysis or unreasonable application of pipelines. The more easy you make it, the more public it can become, even to people who probably are not well aware of what they are really doing. And finally, I think it needs a discussion about the level of data analysis up to which the NFCORE community will go. I think we already discussed this. So Summit 10, for example, there was this, in Bovrak, we had initial discussion about the RNA-seq pipeline. And you can see now that more and more and more and more and more evolve into subsequent applications. But to some extent, we also had the discussion that Cedric and his group said, okay, we would like to stop here at some point because it becomes too diverse. It cannot be put into this standardized, reproducible workflow manager system. We have to make a cut from where then people have to move into all the different directions which are possible. It was probably something which needs discussion within the group to what extent of data analysis the NFCORE community or at least the animal genomics community would like to go. With this, I'd like to thank you for your attention. Thank you again for the great next flow and NFCORE impact that was made to the Bovrak project. I think it was really a major impact that you made to our project. And I was very glad to have Cedric and his team in the project right from the beginning because I think we really distributed the knowledge not just within the EU, but also beyond and can create really a legacy to the community from our Horizon 2020 projects. For that, I'd also like to thank my partners in the Bovrak project for their frequent discussions, input into all the ideas we developed in Bovrak. And I think with this, I spent a little bit more than 20 minutes, but I hope with Cedric that maybe I could give some ideas, even though I really mentioned in the beginning, I feel a little bit embarrassed about all the competences in bioinformatics. Thank you so much, Christa. I could not hope for a better introduction and a better launching for this channel. And as you have seen on the program, we now are going to have a round table. And we did put some topics, but your last slide contains a lot of food for thoughts and then we are going to bounce on this. This is really nice. Before moving to the next step, let's see if we have some questions from Christa. I'm trying to have a look at everybody. It's not always easy for me to see. Do not hesitate to unmute yourself if you want to ask something to Christa. I have a better view. While people are gathering their thoughts, I had maybe, Parvitt, you were raising your hand? No? Okay. So I had one important question to me. So as you know, one of the big things that is happening in genomics now is the Earth bio genome. 1.5 million species are going to be sequenced. And one of the most striking and dramatic results I have seen these last years was the primate pan genome, you know, how much medical knowledge can be unlocked by sequencing our 150 relatives. As you know, the primate genome is just a relative, not naive, but relatively simple use of this data is as precise in the recapitulation as pathogenic mutations as the state-of-the-art methods we had centered on human. Now, have you started thinking of how the sister species are going to become useful to project like ours in animal genomics? I think we already see this. We already see this. I mentioned this paper from Hubert Paul's group already. And also, there's a recent one, another one from this group, indicating when they looked at the viscent genome coming up with some missing pathways. So one of the things which definitely need a lot of attention from my perspective would need for the attention is the entire chapter of ruminants. So what makes ruminants, maybe it's my personal interest, but what really makes the ruminant different in many aspects. And I think ruminants will be also key for sustainable systems agriculture because they can make use of things that we humans cannot eat and will have a major impact I think on future food supply. So I think and we already made quite a lot of progress in sequencing of those genomes. So I'm quite confident that this will help. Another aspect which I would be interested in, I think for hundreds of years we had mouse as mouse models as models for humans and we probably all agree that the mouse is probably not really a good appropriate model for human. And my expectation is that once we know more about the genomes of other species, we might come up with much better models for specific aspects. So it's probably not one fits all, but as we know from for example, ophthalmologists, they know that cattle eyes are a very suitable model, pathological, vertical model for human eye. And in terms of genomics, my hope is that we will come to see that for specific aspects of human medicine, of human physiology, we will see much better suited animal models in our large funders of species. That's the other side beyond then agriculture. Yeah, I actually called Dr. Greenmore and it's increasingly clear that the data being gathered in all of these species and itself to be used for human medical purposes. Do we have any other question for Christa? Actually, if not yet, I have one more question actually. So historically the animal genomics has tended to follow human genomics like in some way, both reg and all of our projects can be described as a kind of encode for found animal, right? And then following on that lead, you now have GTX for found animal GTX and so on. So do you think this trend will keep going? That is to say that the state of the art, the frontiers in human genomics will make their way into animal genomics? Or do you think that animal genomics is gathering its own momentum and that it may go in direction that are specific to this field, that may inspire human genomics, in fact, the other way around? To some extent, we will always follow because the big money and the big money and also that it's for technology development and so on, that this will definitely be always be an inhuman medicine. So that's not no question about it. Many technology developments will come further for humans. One field actually where I think that could go the other way around or at least some inspiration is epigenomics because compared to humans, we have a chance to have really specified environments, to keep animals in specified environments and really then come up with better and more precise data about the consequences of the environment because we can keep it so constant, which you can't in a human. For example, we can really, we can start from who is mated to who, how pregnancies are followed, how the neonates are treated, this is all something which we can monitor very closely or even determine or give them specific environments. So I think there, the animal genomics community in the field of epigenomics, at least we can provide more reliable data than can come from the human field. I see. We have a question from Silva, but just finishing on my question. So then my impression has always been that human genetics was a little bit boring because mostly descriptive. And if it was down to human, we would have strong suspicions genetics work, but we will not be entirely sure. While when it comes to farmed animals, we know it works because that's how we've been feeding the population over the last millennium, improving the genetic, using genetics to improve the breeds. And so we have a proof of principle. Now, do you think I'm always surprised to see how segregated human genetics and animal genetics tend to be? Do you think that this genomics development will bring these two communities a little bit closer together? Actually, they are. We see a lot of animal geneticists in the human field. They try to hide it a little bit. For example, they just have to go to Rothen, where they even share one position in human and animal genomics. So actually, I think we did a lot of certain validation of the human genetics field from our side. Maybe those people don't put this in the forefront now and they don't talk about breeding values, but they talk about genetic predisposition to get a disease. But I think they are already, at least in the field of, maybe in the field of quantitative genetics, it's more than in molecular genetics, but in the field of quantitative genetics, they are very close. Sorry, Silva, you had a question. Not sure. It was to come back to your question. It's a very interesting discussion. And I agree if that is what you were thinking, Cedric, that experimental genetics is easier to perform on animals than on humans, obviously. And in that topic, I mean, for instance, maybe developmental genetics is also something that is for which the animal models have something to play and could be useful to help as, you know, historically, yeah, development has been studied a lot on animal models. And maybe also for everything that is related to evolution and this kind of studies where we have like a broad spectrum of animals and species, there's something that could be useful for the scientific community in particular, that is beyond the human only scope. Yeah, I mean, we could afford by domesticating animals to create a huge diversity. So for example, if you go to dogs, you have dogs with one kilogram, you have dogs with 80 kilogram, imagine a slim female of, let's say 40 kilogram, a very slim female human with 40 kilograms, multiply this by 80, and then you would end up, you can't imagine this. So this is an example, how diverse we created, how much diversity we created by domestication within one species. It's incredible. Do we have any other questions for Krista? So if not, we thank her again for this very, very nice and inspiring talk. Thanks a lot, Krista. And just I look for the emoji for there is an emoji for this, which I will find a reaction. Yes.