 The theme of the session was sort of building open or building communities, and so I'm going to give an overview of how we've been working with one of our communities, a spinal cord injury community, and how we've been working towards building, sort of, as Carol had talked about, a research commons for this community. And so again, it's a collaborative effort, really across four different labs when you break it down in terms of Mary Ann and myself, sort of, more on the informatics side, UCSF, University of Alberta, and then also Jessica Nielsen, who's moved from UCSF to University of Minnesota, are really in the domain and working with this data. And as Carol had a slide on the comments, I don't have to go over a lot of this really in detail, but really what we're trying to create is a shared space for these researchers to really work with their data in different contexts. And that means not just publishing a final data set, but maybe as they're collecting data with the graduate students in the lab running studies in animals and so collecting the data as this data comes in and really trying to create that common space. And how do we, as Carol said, sort of bring these fair principles into this community? Again, it's not forcing these things on the community, but really trying to figure out how to do it in the context of the work that they're doing. And so the commons that we've been putting together is the open data commons for spinal cord injury. We have now more than 40 labs signed up within this shared environment. There are a number of data sets that are in the space, and I'll explain a little bit about what that space is. And it's really focused around this concept of the lab, and so that was one of the first things in working with the community. How did they want to organize themselves in this commons? And how did they want to manage sort of the movement of data within the commons? And again, all of this was driven through discussions, and I'll talk about this in the second half, about how we went through the various sort of community engagement pieces to build out those commons. And the other thing was, well, how do you explain data sharing to the researcher? And this is, I think, Mary Ann came up with this, and this is one of the best analogies that we've seen in terms of data, comparing it to a restaurant. So when you get in your ingredients, that's sort of the private space. You have the back end where things show up on the loading dock, then things are in the prep kitchen, and you can see a little bit more of what's happening, and then when data becomes public, it goes out sort of into the dining room. So you have these different stages in terms of how the data flows. And what we've really been trying to do is focus, so on the left side there, you see sort of like an instructional screen with a couple options given to the researchers. So how do we guide the researchers through these flows of how you want them to contribute data so that their data ends up within the system? And then making sure that the actual data that you have access to grows. So what I'm showing here is that you'll see that on the bottom there, there's a data set that's private access. That means that I have access to it in my lab that I'm a member of. And you can actually be a member of multiple labs. So you can actually have data coming from multiple labs that you might be collaborating with. You'll see then that there's a green one, the public data set. So there are some public data sets that have come out, and I think the resolution is a little low, so it might be a little harder to see. And then the larger pull out are the yellow items. These are the ones that are shared within the commons. So these are the data sets that people have shared outside of their lab with sort of the researcher community. And so these are the ones that are not completely public, but shared within the commons researchers that are part of the system. And one of the pieces that we've been trying to pull together is how do you actually put into place methods to actually do some of, for example, the alignment of data elements. And how do you make that something that's intuitive for a researcher to do in terms of one of the big goals of projects like these, when we're talking about interoperability and reusability, is that when one describes age or when one describes an outcome measure, that if someone's describing the same outcome measure, that these data sets can be pooled across research labs. And that's just a process that is not very fun to do. But how can we embed that sort of in the process of this workflow? And so what we've been doing, we've been tying this to some of the systems we have so we can sort of have community input into the data elements that are described. People can assign those and then actually, and this is actually a focus of the next phase of the project that's just begun in terms of trying to make these workflows more intuitive, hopefully a little engaging for researchers so that these actually happen. The last part is sort of the data publication piece. So once the data set is published through the system, it gets its DOI. We have the schema.org metadata, which you see on the bottom right. So trying to make sure that the data set is meeting some of the basic principles that we've been talking about. But all this is sort of built upon a number of efforts with community building. And so we've been very fortunate that we've worked with these other labs, UCSF University of Alberta, University of Minnesota that have been very engaging and also engage well with the broader community. So we've had very good interactions with sort of the broader spinal cord injury community. And again, spinal cord injury community is large, but it's not too large. It's a decent size. And back in 2016, when this project first was starting up, there was actually a workshop, an initial community workshop that was put together with funding from NIH and also some private foundations to bring together the researchers and the funding agencies to discuss what this open data commons might look like and really talk to the researchers about the possibilities but also really get their input as what they want and what they would use to really sort of set the stage for sort of the initial prototype development that was sort of funded in this period. And there was actually a paper written sort of about the workshop. So Calhanna et al. in 2017. And they did a couple assessments of the participants. And again, most of them were from the spinal cord injury preclinical world. There were other clinical researchers there, bioinformaticians, non-governmental funders and governmental funders. So it was a good mix of people who were attending. And then looking sort of at the community profile, how do people share data? And really, it was sharing sort of within the laboratory, within the local environment. So those are those first four bars. That's sort of where the majority of the sharing was, but not with the broader community. And what were they actually using in terms of managing their data? And it was paper and spreadsheets. So this is sort of the community that in terms of what we had to, as Carol said, giving tools that the researchers won't use is not going to work. So how do we work with the paper, the spreadsheets that they were using to develop that system? This initial meeting led to sort of some initial developments of alpha releases and things that we would test internally, again with the internal groups that we were working with. And then we went sort of to the American Spinal Injury Association in 2017. And again, tried to sort of get some feedback from the community, but really to start providing some information about what fair was to the broader community and provide information on what the system was going to be like. And then also that ended up also being a way to sort of recruit individuals, for example, to the steering committee, people that were interested in contributing to sort of guidance in terms of what we were developing. Now one of the sort of critical pieces for any sort of project to I think move forward in many instances is that you have at least one or two success stories that you can point out what the benefit is of data sharing. And many communities have these, and it's important to highlight these. So one of the ones from the spinal cord injury work was Jessica had done some data analysis across a number of data sets. And getting these data sets together was actually a huge amount of work because she had to deal with all the spreadsheets, all the paper. She termed it dumpster diving, right? You'd go to a lab to collect their data and they'd throw you a box of papers and spreadsheets and disks and she had to take that and put and collate it into something that was usable to do this analysis. And so out of this analysis, there's actually something that was really interesting that came out and looked at sort of what is the effect of hypertension in terms of predicting long-term recovery from spinal cord injury based on the findings across a number of different studies. And then this has moved on now to further studies looking at this. But again, this is one sort of example of how pooling together data, sort of the community coming together can actually provide insights that haven't been found before. And, you know, working, you know, providing, you know, such examples, I think, you know, helps sort of galvanize, you know, how, you know, people can see benefits from, you know, being part of the commons. So, you know, following on from sort of some of the initial meetings, you know, one of the things that was instituted was the steering committee. So out of that first workshop, you know, in the subsequent workshops, we had a steering committee that we'd get together with periodically. And we'd sort of use them also as a sounding board for policies and procedures. So how does one work with the community? What are the guidelines? You know, how does someone, you know, if someone comes in and wants to register a lab, what's required for that lab to be recognized by the community? All right, so some of these, you know, community decisions, you know, for how, you know, one operates, you know, virtually in this space. And so the Ferris Sky ahead meeting that was held at SFN in 2017 was really a stakeholder meeting for the project. It was a pre-meeting at SFN and, you know, it was really something where we sort of provided progress, again, to funders and to steering committee members that were there. And it was really, you know, a place to talk about, you know, some of the things in terms of also how do we move forward? Again, because we were building a prototype, you know, something, you know, that, you know, people could shoot arrows at to say, yes, this works, this doesn't work, and how do we want to, you know, move this forward? And to sort of help us or help guide us into, you know, defining sort of what the future roadmap for the project would look like. And, you know, building on that roadmap, you know, the next year, we actually had the first of what we were calling a datathon at SFN. And so this was an activity that was held at SFN, but we've also held a number of these activities at smaller meetings, at some of the other spinal cord injury meetings, and we've also had some of the postdocs who were working in some of the labs go out to other labs, you know, in terms of working with individual laboratories and how to use, you know, the system in terms of working with the data. And so really, you know, this was, you know, a first, you know, initial meeting where, you know, we could try and demonstrate some of the benefits, you know, we had some hands-on participation in the meeting, again, you know, at SFN you don't have that much time, but again, it's a good place to sort of, you know, get people together, you know, get them signed up, you know, answer some of the initial questions. And so really, what we've been working through is sort of this, you know, interaction between, you know, what we're developing sort of in the context of, you know, the collaboration we have, you know, in terms of, you know, the participants who are developing the commons and engaging with the broader community. And the next phase of the project really is trying to build on that. And one of them is this community curation process. So again, one of the, you know, Carol, I think she had like the little magic wand when it went from the lab to the repository, right, that some magic happens and, you know, your data goes from the lab and ends up really nicely in a repository. And that is a lot of work and a lot of effort, but again, how can we start moving some of those tools, some of those processes into the work of, you know, basically working with your data in the system, you know, when you're initially doing this. So the first mile, you know, trying to get these procedures working earlier in the data collection process. And so developing a community curation process and also what does it mean to have data that's, you know, in the community and data that's public. So what is sort of the, you know, the review, you know, to publish data sets, you know, through the commons. And you know, one of the ideas, you know, that's, you know, been being looked at, you know, has been discussed with the steering committee and others is sort of the formation of a curation board, you know, sort of similar to an editorial board, but in this case for data sets, you know, that there's some minimal standards that these data sets have to meet. And again, you know, FAIR is very new to all communities that's out there. And so in this community, this is one of the things that we're trying to define, you know, what does that mean? You know, other things, integration of analytic tools, now that we have a base, you know, for people to get their data in, you know, tying them into some of the, you know, tools for, you know, large scale data analysis, machine learning across data sets is something that, you know, is seen as a way to engage, you know, different types of researchers within the system. And again, broadening out to, you know, a larger community, more outreach, more outreach, and more outreach. So I'm really trying to, again, build this in the context, you know, of the community. And you know, I just want to thank, you know, all the collaborators, you know, the steering committee who's been, you know, giving us very good feedback and who's engaging, you know, with us. And, you know, they, you know, both the foundations that have provided, you know, a lot of support, you know, for the development. And then also, you know, everyone that's supported, because it's sometimes hard to get meetings together, you know, in terms of funding meetings, you know, to bring the community together. And this is something that's critical to do. And it's not something that's very easy to put in, you know, your grant application. So we've gotten a lot of support from foundations and also some from NIH, you know, to hold these meetings, to host these meetings, so that we can bring people together and engage them in the development, you know, of this system. So again, you know, thank everyone. And thank you for listening. And, you know, I do have a demonstration table upstairs. So if anyone wants to stop by and chat, feel free to do so.