 Hi everyone. Thank you very much for joining and thanks for the great introduction. It's really exciting to have so many people join the Hackathon this year. It gets bigger and bigger every time we hold it. And I think this is certainly the future of events like this. To have them online and get people from all over the world to join. It's really exciting to see. Let's see if I can share my screen now. So I've got a fairly short stack of slides today. I'm going to be talking a few times this week. But I'll start off the week just with a quick introduction to NFCore. As I said, there's quite a lot of people amongst you who are fairly new to NetFlower and to NFCore. So if you've been around for a while and you can go and make yourself a cup of tea or something. If not, just treat this as a refresher for what NFCore is and how we all work together and how this thing is organised. So my name is Phil Jules. I work in Sweden at the Science and Wildlife Laboratory for SILoFLAB. And I work at a platform at SILoFLAB, which is the Swedish National Genomics Infrastructure. So we work with very large amounts of genomic sequencing data from research groups all over Sweden. And this kind of drew us to start building these pipelines with NetFlower several years ago. And so that's a NetFlow hackathon, a NetFlow meeting back in 2017. I think I started talking to a few others in the NetFlow community about how we could standardise our various NetFlow pipelines and kind of start to work together, namely Alex Peltzer, who was at the time at Cubic. And also with some group in Singapore and a few others. And so because of that, NFCore was born. And the idea was really to try and stop everyone from having to rewrite the same versions of the same pipeline since come together on a set of kind of best practices. So with that, and of course, this is a tagline on the homepage, which gets kind of trotted out every time I give a talk. It's a community effort to collect the curated sets of analysis pipelines built using NetFlow. This of course only makes sense if you know what NetFlow is. Hopefully some of you have been in the NetFlow training this morning, or at least following along with it, or be kind of familiar with it. But it's a workflow manager, which allows you to chain together different steps and analysis pipeline. Importantly, it works with virtually any underlying computing structure. So that's built in support for cloud as we saw, as well as traditional HPC. So we use it with a slur. Job schedule, for example, and all kinds. So it works virtually everywhere. And it also has built in support for software packaging. So Condor, Docker, and Singularity are the three which we work with the most in the NFCore community. And this means that you don't have to install any software which is specific to a given NetFlow pipeline or NFCore pipeline. Each NFCore pipeline comes with all of the software in these. So all you need to have is NetFlow and then one of these three. That means that a reproducibility and a portability of every NFCore pipeline is it's superbly high. And if you've been around in bioinformatics for a while, it starts to solve an age-old problem of when you want to run something new. It takes you a day to just get all the different dependencies and everything. That's the thing of the past now. So that's the next flow. And of course, what is NetFlow? The heart of NetFlow really is just a set of guidelines. It's a community who comes together around these guidelines and we kind of agree. This is the way things should be done. These are the best practices we're going to adhere to. That's ready. That's it. Guidelines and kind of adhering to them is kind of tough to manually regulate. So we started building a set of tools to kind of automatically check these things and also to help you work with these things and kind of help the tools. So there's nothing to do with the running a pipeline itself per se. They're all fairly generic. And these are written in Python because that's what we're most familiar with. But they help you as a developer and also as a user as well as simulator. Finally, of course, what most of us come to NFCore for is there's a set of pipelines that we kind of provide and build as a community. And if you're a user, then that's kind of the main thing you come for. You don't really need to worry too much about the guidelines and the tools and everything. If you go to the NFCore website, you'll find this big list of all the different NFCore pipelines along with descriptions of what they do and what they work on. I don't want to say to this slide a few days ago, so I hope it's still valid that we have a currently have 25 pipelines with a release, which means that they're effectively stable. 14 pipelines under development. So no, no active release yet. But that doesn't mean they're not usable. And then three old pipelines, which have now been archived. And this is kind of a typical, any NFCore pipeline can be any one of these groups. Unlike some other bioinformatics kind of listing services. And of course, not somewhere where you can build your pipeline come along and kind of add it to a database so that people can find it. That's not the idea. We don't want to just collect as many pipelines as possible, kind of the opposite actually. We want to have just a curated single minimal set of pipelines. So we want to have one pipeline data type. We want to have one RNA seed pipeline, not 20 that you have to choose from. That one pipeline can have lots of different options and we run in different ways and be configurable. That's great. But we want it to be a single set of pipelines. So it is very clear as to which one move them out for each type of data or which question we want to ask of the data. All of the pipelines that were listed there should be very, very reproducible. Hopefully, at that point, as well as the software, which is wrapped so very reproducible as well as the way that next flow can be moved around and everything. Every pipeline also comes with a digital object identifier, DOI, which is specific for the pipeline. And then there's a specific DOI for every single release of that pipeline. So that means if you use a pipeline, you can cite that DOI in any publications and anyone can go and find the code from that pipeline from that specific release. If they run that, the software container is tied to the release. So if you rerun your analysis, you'll be running exactly the same versions of the software with exactly the same versions of the code. And if you specify the same version of next level, you should be basically completely reproducible. Everything we do with NFCore is held under open source licenses with MIT. So your rack is kind of strongly encouraged for you to take these chunks of code and use them for whatever other projects you want. You go off and build off of NFCore. That's completely fine. And with the NFCore tools, the helper tools, we try to make them flexible enough that they'll work with any next flow pipeline. This doesn't have to be specifically NFCore. So of course that's our primary target. And finally, all of those pipelines, because they are all past those automated tests because they're all being checked by the NFCore community, we can say that they all adhere to best practices for vine robotics and open computing. I'm very happy to say that the NFCore project was published quite recently in Nature Biotechnology and came out eventually in February. This was a big body of work worked on by much of the core team and it took a long time to get through for various different stages of preprint and then review. The whole paper was completely rewritten several times. So if you have read the preprint, it's worth going back and rereading the final published version because it's a very different beast. It's quite easy to read the paper itself is just a short letter, so it's not very long at all. And we kind of go through the basics of how NFCore is set up in these three periods of kind of deploying pipelines of this community participation and then kind of different features we have for development. If you look in the supplementary materials, that's much longer than the paper and that's sort of a paper in itself really that goes into many more details and specifics of how we run NFCore community, the automation that we set up, how we keep everything synchronized and working actually how this thing works at scale. There's some nice graphs in the back which of course I like because I made them. They're showing the size of the community and how we've grown over time and how quickly we respond. Now, when we made these, we of course knew that they'd be out of date the moment we set them into a file. And so we did it in such a way that all of these are actually live on the NFCore website. So if you're into that kind of thing, you can go and check out this URL and you'll see the live data of exactly how many Twitter users follow as we have right now. So this is really fun and you can see on these graphs that the NFCore community is really still going fast. We haven't plateaued and every single day we're getting more and more people using pipelines and more and more people involved in contributing to those pipelines. To give you a feel for what that looks like back in 2017, we were just three. Three people, three institutes, three people on GitHub and that's when I created the NFCore GitHub organization. Within a year, things had already started to take off a lot faster than I expected. Parallel, the main developer of Nexplo had already kind of retreated a couple of tweets and started to get a bit of a following bit of an interest. And by 2019, it was clear that this project was really starting to take off and had a place in the bioinformatics community. Now in 2020, we're starting to have in the high hundreds to thousands of people involved. The Slack community is massive with lots of people coming to Slack to ask for help and also to kind of discuss how to contribute. Lots of people making issues and committing code on GitHub and a lot of people following what we're doing on NFCore. Top one, the institutes, that's the contributors page on the NFCore website. It shows a list of all the different institutes across the world where people are using NFCore. So if you haven't already, it's very much appreciated if you can go and add yourself to that page. It's just a quick pull request to a YAML file with your institute name and logo. And it really helps to kind of show that we're global and that people are using these pipelines. And if people will go to that site and they find the associated name in their institute, they recognize they know who they can go to for a helpful advice. So these are those contributing institutions. It would be great if we can get a few more in some of those blank places in the map, for example. And it's brilliant to see what started off as quite a European centric projects. And it still does have a European focus where you can see we're starting to make inroads all over the world. If we look at who's visiting the website, there's a bit more of a spread picture, but it's still sort of roughly conforms to a similar kind of spread. And when you look at these stats, I mean, it's just, I think it's fantastic of how many different people we have and how many repositories and just what a huge, huge body of work this represents. Right, I'm going to stop there and pass on to Federico. Federico's going to talk about how to actually use these pipelines and how to run them if you're new to NextFlow and new to NFCore. There's four main places that I'd recommend everyone to get involved with if you're interested in NFCore. And all the work, all the code is done through Git and GitHub. So we have the NFCore organization. If you just drop us a line, either on the instructions on the website, there are NFCore slash join, then we'll send you an invite and then you have access to work with all of those repositories. The main place we communicate is Slack. So it's an NFCore Slack. Again, if you go to that URL, you can get yourself an invite just to click the button. And we organize all of our discussion in the different channels of the NFCore Slack organization. We broadcast what we're doing over Twitter. So if you're a Twitter user, hop on there. And then increasingly we're starting to use YouTube to put things up there. So for example, this talk is being a live stream to YouTube and it will stay there so that people who aren't able to join us live today will be able to watch this and relate to it. We're also hoping to put out more kind of tutorial material there and other help material. So just go to the website, check out that URL and make sure that you're involved in all those four different channels. And with that, I'll leave you for now and I'll pass you on to the next one. Thank you for your attention.