 So I'd like to welcome everyone to this workshop for Medical Research Institutes. My name is Kristin Kang. I'm a senior research data specialist at the Australian Research Data Commons. The workshop today is presented by ARDC and AMERI, as part of an event organised with the assistance also of ARC and NHMRC and the Unis listed on screen. So the first part of this event was held last week. It was a webinar with a presentation by Justin Withers from ARC, and he was discussing the 2018 update to the Research Code of Conduct and quite specifically the guide for the management of data information in research. So to supplement that webinar, there was a workshop held specifically for universities last week to discuss how they were implementing the code and to share their experiences with each other. Today is an opportunity for MRIs to share how they're implementing these research data policy requirements and how they're approaching their data management responsibilities. But first, I'd like to acknowledge and celebrate the first Australians on whose traditional lands we meet and pay our respect to their orders past and present. So as I was saying, last week was a presentation by Justin on the crux of the data management guide and its context within the Research Code of Conduct. The key takeaway from that is that the responsibility for managing research data is shared by researchers and research institutions. Today, we're going to discuss, or it's going to be an opportunity for MRIs to discuss how they're responding to this responsibility. Specifically, we want to find out what have you been able to achieve at your MRI in establishing data management policy and culture. And importantly, how have you been able to achieve this? And of course, they're not easy things to achieve, and I don't think there's any institution out there that feels like it has everything 100% where they'd like it to be. So today is an opportunity to share our experiences and to learn what has been working at each other's organisations. Hopefully these insights will be useful, if not informative, for shaping your own MRIs practices. So the agenda for today, we're going to start in a little while with a brief presentation and that's just to get our minds focused and oriented to the topic and that will be followed by a short Q&A. There are going to be two breakout sessions because today's workshop, we really want to focus on discussion and sharing our experiences. We're going to briefly regroup after each session and share what we've learned. And then before we close, we'll discuss some ideas for next steps you can consider in developing your data management practices further. And just as a note, because it comes up quite often in these online forums, the session will be recorded and the slides and the recording will be shared as appropriate. So just to really hammer this time, today we're going to be looking at some key foundational aspects of data management practices at your MRI. Those are establishing and coordinating the policies, responsibilities and culture around managing research data. As I said, we're going to spend most of our time discussing this via two breakout sessions. In the first session, we want to be discussing data management policies, establishing buy-in at your MRI for data management and coordinating your data management practices within your MRI. In the second session, we'll be discussing your approaches for establishing clear roles and responsibilities for managing data and also asking how you've been able to raise awareness and train your staff and what success you've had. In these discussions, just we'll be asking you to share what you've achieved, where you've been able to achieve and get success in these areas, but importantly, we want to know how you've achieved these outcomes and what you can recommend to your colleagues at other MRIs. To round up those discussions, we'll also ask you to consider what you think your next steps need to be in developing your data management capability further. Now, I'll just make a quick note that all breakout rooms are going to be the same and each group will discuss the same topics. You'll be automatically allocated into a room and there's no need to nominate a topic or room to join. Okay, before we jump into those group discussions, Rad and Neba has kindly agreed to share some insights into how our Telethon Kids Institute is addressing its data management requirements and to see how they're dealing with their data management responsibilities. The presentation will be followed by a short Q&A, so can you please enter any questions you might have into the chat channel. That channel will be monitored by ARDC staff and will collate your questions to the Q&A afterwards. Now, obviously every MRI is different and needs to find what works best for them, but sharing our experiences can be a valuable way of achieving this. And I'd like to thank Rad and Telethon Kids for kicking off those discussions today by sharing their experiences. So without any further ado, I'd like to hand over to Rad and invite him to give his presentation. Thank you. I'm just going to share my screen here. Just want to make sure that everyone is seeing my screen. Excellent. So thank you for having us. My name is Rad and Neba. I'm head of research data strategy here at Telethon Kids Institute and I'm here with Tara McLaren, our head of research development. And today we are going to talk about what we are doing in data management in Telethon Kids, the success that we have and also all the sort of problems that we are encountering and how we are tackling them one by one. So we are trying to answer the questions about where we are in terms of maturity in our data management. So within that well-known data maturity model that's being used for a lot of organizations in the world to just use it as a metric to, you know, see the way their research institution about the data management, where they are in the process. So the reason why we are relying on the data maturity model is that it gives us a lot of insights on the things that we are doing in the right way and the things that we need to focus more in terms of moving that needle from the initial stage that you are seeing here where most of the organizations are across the globe and towards going slowly to a more optimized and managed sort of state in managing the data. So the focus that we are, the reason why we are focusing on that data maturity model is that it's going to sort of tell us what kind of governance, what kind of data governance we should have and all the things that we need to develop internally to tackle all the problems that we might encounter. And as you see these numbers, they are sort of a global in all the organizations, not just healthcare. So it tells you that fewer of the organizations are really managing well their data as an asset within their organizations to be able to be more proactive in terms of data governance. So when I'm mentioning governance here, but I'm just meaning data governance, not any kind of other governance that have a lot of governance cores within research institutes. But from that perspective, what we all aim for is to be more proactive. And at least if we attempt to do a lot of the work in data governance toward moving that needle, even if we are in the middle section where we are pre-emptive and sort of how we manage our data, that's also a big win for us. So what I need to mention here, and especially that it's a trap that we noticed, and it's very well known in the data management area, is that when you implement all the policies and changes and data governance towards moving towards a more advanced stage in terms of data governance, you can come back. So it's actually, if you don't keep working on improving all your standards and you don't consider the whole thing as a constantly moving efforts towards being the best in managing the data, you can actually retract and go back because the technology is moving fast, the constraints are moving, we are facing a lot of regulatory compliance, so we need to be all the time working towards improving our data management capabilities to be really effective in managing the data at the institutes. At Telethon Kids, we have different research areas, but when we are tackling these as from project-based perspective, we notice that our researchers are facing a lot of problems, especially in the data volume that is being generated in all these areas. So what I'm presenting here is a sort of template that we are trying to use, not necessarily in biology, but other research that we are covering, but we are using this as a proxy to show that a lot of the data is not just about science, but also about other areas that are tightly related to science, there's the governance and the procurement and the communication, and all these are related to, that they have impact on their research outcome, but also our scientists, they are data generators, which means that once a data is being captured in one of our systems and being used in another domain, then it's going to be transformed into other shape and formats, and we are constantly generating these amounts of data. So volume is one of the components that is causing a lot of problems in terms of data integration, insight generation, knowledge management, and the end goal of all these pipelines is to the scientists to be able to generate insights and publish high-quality research, and we, I mean, from all the interviews that we have with our scientists here and also from our network, integrating this data is the most challenging task for the scientists to be able to respond quickly and accurately to any questions that they may have, so this is a big problem that we are trying to tackle here at the Institute. So in order to do that, you need to be prepared from management and data governance perspective, and this is what we are trying to achieve here at the Institute, so we are developing, we are constantly working on that, a simple framework, but at the same time, something that might be moving and changing over time, so the first step is to, that we found very effective is to try to adopt some sort of data principles we work by, and we are working closely with our researchers to try to implement the fair standards at the Institute, so these are the first building blocks for our strategy, because from these data principles, we're going to generate all the data policies, procedures, procedures and guidelines to frame our work from governance perspective, so once we have these sort of guidelines, we develop the data strategy, and the data strategy can be complex because it's touching several areas, it's not just about research, but it's about people, it's about management, it's about infrastructure technology, so all these are building blocks for our data strategy, and the outcome of this is obviously the data governance that is really specific to the Institute and should be specific to any medical research institute, because that's the sort of DNA of how the organization is managing its own data, so we can have a template, but it should be always sort of personalized to the Institute's needs and the research institution needs, so we are trying to work on these areas all together, and we try to keep it simple even though it's a marathon, it's not, it's something that's always changing, and we try to keep it simple because when it gets complicated usually it doesn't work, so we are trying to work all of us together, and one thing to mention here is that what you see here is really complex in the sense that it's involved, it involves a lot of the organization's core business units, it's not just about researchers or people managing the data or IT, it involves people in research development, procurement, communication, contract, legal, all these areas, so it's really a group exercise. So we are using all the things that we can use in order to shape these data principles, but at the same time we're trying to map this to the principle of responsible research conduct, so if we take all the HPs of the code of conduct and we see what kind of things we can map from the fair principles to these principles, we find the overlap between all the sides of the fair principles, which are the findable, accessible, interpretable, reusable, and we add the S that we are trying to personalize for the institute because we are really focusing on the security and data privacy, so this is where we kind of adopting a little bit the fair principles to telephone kits, and we are mapping these to, we're finding the overlap with the code of conduct and we are trying to drive our efforts from policies implementation perspective in order to be compliant in both areas. In terms of institutions, all what we are doing for data governance and we are trying to implement these things, but at the same time we are learning along the way, so there's a lot of ongoing work to implement these recommendations that come into the code of conduct, so covering these areas from an institution perspective, we are doing a lot of work to try to implement that from raising awareness among our community, the science community, and also doing a lot of workshops and trainings in terms of data management, best practices, the good things to do about data, basically to make our scientists more aware of the risks and security and the good values about using the data in a good manner. So we are taking a project-based strategy right now, so we are tackling these problems in silos, and I think this is the bottom-up approach that we are having, but being aware of the strategy and values of the institution at the same time, so we are trying to meet along the way, and we are trying to train our base of users and scientists to build this data search program in a continuous manner, and along the way we develop all the data policies. We have a couple of them that we are revising and a couple of them that we are in need to implement from scratch, and this is to respond to the changing landscape of regulation around our research. We have a lot of constraints in terms of governance and compliance, especially our scientists are working not just in Australia, but with partnership with Europe and US and other countries, so we need to be aware of these legislations outside Australia, inside Australia, and try to find that overlap. That's going to impact the policies that we are having and impact in terms of writing new ones or updating the ones that we have in order to be compliant, and that also has an implication on the technology that we are using and how we implement these policies to be sure that we have all the safeguards in terms of data security and privacy. So with that, I'm going to end over now. So as Kristen mentioned in the introduction, this is actually a responsibility for both the institution and the researchers, and we're really lucky at Telefun Kids that we've actually bought right on as our head of research data strategy, and he is bringing together all of those silos for the top down and bottom up approach. But the other part that we're doing is really clarifying for researchers what their responsibilities are around the implementation. So the four areas that we're looking at are retention and publication. So we have really good policies around that, but we are constantly updating them to make sure that they're aligned with all of the changes to data locally and internationally, managing confidential and sensitive information with our big focus on stewardship as well as security, acknowledging the use of others data. Publishers are taking an increasing role in driving researchers to make sure that their data is open access where allowable, and encouraging us to open access publishing. And also we've just updated recently our authorship policy to include the credit taxonomy, which is the contributor roles taxonomy where researchers may be contributing data to a project, but not necessarily authorship, which we think is really important. And then we also use the famous carrot and stick approach for encouraging our researchers to engage with the relevant training. So a lot of Rad's time at the moment is spending out talking to research team, but we also do implement some stick approaches when it comes to engaging with that relevant training because it's really important. We also have very well defined regulations and responsibilities when it comes to breaches of the code and these are in line with the code of conduct. So we are building a strong voice safety culture to allow for concerns to be raised and at our organization we're tracking that and it's becoming increasingly we're recognizing that there's more voice safety. We are reviewing the roles and responsibilities of those involved in the management of the breach. So as with data it falls across a number of portfolios, it falls across grants and research development, governance, legal and the HR team. We are making sure that our processes for receiving and managing concerns and complaints are well defined and available on the external website and with our focus at Telethon Kids on Aboriginal health, we're also making sure that they are culturally accessible for our Aboriginal communities to raise any concerns. And we're currently reviewing our processes for managing research conflicts of interest. I think that's something that as an organization we can become better at. I'm back to Red. Thank you. So what we really try to do here at Telethon Kids and I think this is a natural evolution of things once you figure out where you are in the process and what kind of things you'd have to do. So having that project-based approach we realize that we need to focus on a couple of buckets of work that we need to implement. It might look linear, as you can see it here, but it's not really linear. It depends on the case. But the first building block is setting up the foundation, the data governance units or core. That's going to be the link between the business, the IT and the science community. So that's really the glue that's going to put the framework about how things work from data management perspective. The second problem is tackling the data integration problem. That's purely technological, but also it's really taking a lot of directions from the data governance itself because what we're trying to achieve here and we are building this as we speak is try to embed the data governance into the technology itself so that we can get to the point where we're going to have an automated data governance that requires less human intervention in terms of updates and maintenance. And this should embedded into the technology so that we tackle issues like automated data privacy, tackling all the personal identifiable information and all these constraints that come with the data governance. And it's not stopping there because once you have the data integration in place, you need to extract the information. And this is where using the eight P's from the Code of Conduct and the Fair Principles all of it together, we need to make that data available for scientists in a click of a button. It's not an easy process, but it's achievable and doable. So we need to make that data tracking and analysis as seamless as possible for our researchers because they want to answer the questions very quickly in their response to their problems. So when we get to the point where we have the data catalog in place, this is where we start to integrate other types of data with research data and we get into the data streaming sort of capability and this is where things are really well managed and the infrastructure is tightly related to the governance and everything is well organized. So the last point which is AI. So AI is not just as an outcome for our pipeline that we are trying to set up. It exists in all the steps that I'm talking about here, but it's more of how all this research data is going to be used by the scientists. So this is where we enable scientists to be not just data stewards but also they transform this data into other types of data which is more concrete in terms of analysis and insights generations and it's a cycle because once you have this data generated through AI, it's also another data that needs to be managed. So we close the loop and we start doing all these kind of things from the beginning. So this is what we are trying to achieve. It's a marathon as I mentioned. There's a lot of things to tackle and to adjust, but it's a learning experience and we are trying to do that as fast as possible and as accurately as possible to be able to enable the scientists in the institute. Thank you. Thank you so much and my apologies Tara, I didn't realize that you were co-presenting, so didn't have your name in the slide to begin with, but thanks guys. And yeah, really great to hear how you're going about things in a very proactive as opposed to reactive manner as your first slide alluded to. We are a few minutes over time at the moment and we wanted to allow a lot of time for breakout room discussion. There are a couple of questions that have come in so maybe I'll ask those quickly, but if I could just ask you to, if there are quick responses, that'd be great. If there are longer responses, just let me know. We'll leave it later. So one question came in about have you had a positive or what's been the most positive reaction by researchers about data management training or what aspect of the training has had the most impact? So one-on-one is worked so far. So Rad's done a lot of work with identified some teams that have complex data needs and so through lots of conversations with those teams, we've got people now knocking on Rad's door to actually have conversations about data management. So I think strategically that training has been one-on-one to start with and what we will do as we move forward is, I guess, move down into the EMCR stage and so the researchers who are coming through the pipeline, we're ensuring that they're getting good training from their supervisors, but they're getting good training from the organization as well. Yep, that's wonderful and definitely I guess the experience I had when I was a research data manager at NSW. Next question, will you be primarily working with new data sets or are you retrospectively working with all data sets as well? So both. So that's some of the projects where we need to retrospectively sort of work on and fix. There's a lot of things that we need to sort of go back and readjust, but so with all these, all the projects that already started before we implement our strategy, it's easier to start from the beginning with new projects when things are well organized, but it doesn't mean that we don't retrospectively work on other projects that already started. And from experience, we found that there's a lot of positive feedback from these scientists and they are all willing to do the change actually. Yeah, so a really exciting project we're working on is supporting the three longitudinal cohort studies in Western Australia to actually look at better data management and integration through a sort of shared portal system. And that's a project that if it takes off, we hope it can be expanded across Australia to integrate many longitudinal cohort studies. Fantastic.