 Okay, so good morning everyone and welcome to this course which will introduce you to OpenBIS and we're very happy to have Patrina Barriari and Rossislav Kuzdiakiv giving the course today. So we'll talk about how to manage research data with OpenBIS. So first of all I just want to say a few words about who we are. We are the scientific IT services of the ETH Zurich. We are a large section of the central IT services of ETH Zurich. We have over 40 people in our group that have different scientific competencies and we all have or most of us have a scientific background. We do different things at ETH. We actually have four groups in our section, so we go from maintaining the clusters at ETH, the high performance computing to developing custom softwares for research groups to helping groups with data analysis for example and we also provide services for working with confidential data and services for data management and today we will particularly focus on the research data management part. So we develop also our own platform that is called OpenBIS for this and OpenBIS I would define it as a complete solution towards fair data management. There are three components in OpenBIS. One is the inventory management which you can use to manage your samples, your materials, the reagents, equipment, standard operating procedures if you have them. Then there is the electronic laboratory notebook component which you would use to describe your experiments, describe all the processes and the data analysis and finally the data management component. So in OpenBIS you can store data connected to your experimental descriptions or to the samples and data can be of any size really, so there is no limit for this. The limit comes actually from the storage you have available in the back end and you can also use these components independently so you don't have to use the full, all the features of OpenBIS. You can also use some of them depending on what your needs are. So some information about OpenBIS. OpenBIS has been around since quite some time now so the development started at ETH in 2007 and nowadays it is used in different quantitative scientific disciplines. So when we started we started working primarily with biologists but nowadays it is actually used in different fields. So for example material science is another big area where now we have several users but also in other departments at ETH this is used and here you see a few of the institutes where OpenBIS is currently used. So primarily it's used in academic institutions not really in companies and we have it at ETH Zurich of course but also other institutes of the ETH domain like MPA and PSI and other universities or institutions in Switzerland and also we have several users in Germany specifically but also in other European countries. So this is a picture of the life cycle of research data which you may have already seen in some if you took some data management courses for example this shows what happens to the data in the life cycle of a project. So usually you start by generating data in one way or another by taking some measurements of something and then you need to process this data, you need to analyze this data, you need also to preserve this data maybe not all of it but some of it and then the most relevant data will be published and hopefully also reused by someone else to progress in with our knowledge and scientific discoveries and in the center you have the data manager and planning because you should always at the beginning of a project you should always plan how you're going to deal with your data what you're going to do with your data where you're going to store it and how you're going to describe it and so on. And where does OpenBeast fit in this data life cycle? So how can we help you with OpenBeast with managing the data throughout the life cycle? So we can take a look at the different components. So for the data creation and data processing basically the three components of OpenBeast that I mentioned at the beginning can be useful. So we see here a few screenshots of the inventory component of OpenBeast. You can use the inventory as I said at the beginning to manage for example your equipment which can be any type of equipment that you use in the lab so you can create some folders which can be created by the admin of the system and customized tailored to the needs of the lab. And you see here for example some mixers and then you see a table with all your equipment. Then you can have the samples for example or different materials. So again in this case this can be customized. So we provide a version of OpenBeast for life sciences which already has some predefined types. But we also have a generic version of OpenBeast which is empty essentially so all the folders and the structure need to be created by an admin of the system. And also the life science version despite having some default types it can also be further customized. And also in this case so you see you have different folders of different collections. In this case we have a collection of chemical AD mixtures so you see all the list of your samples. And then you can also have a collection of protocols. So these are the procedures. If you have standard procedures that you follow for some measurements then you can store them in the inventory part of OpenBeast and they will be accessible to everyone in the group. And these folders access to these folders can be controlled so you can decide who has access to what and what type of access you want to give. If you use OpenBeast for managing samples you can also keep track of the samples, the positions of the samples in the different fridges and freezers that you have in the lab. And you can also use the barcodes to keep track of the samples. So this is a functionality that is now more and more used in OpenBeast. So we've recently also done some further developments and improvements on this functionality. Then we come to the electronic laboratory notebook part. So in OpenBeast by default every person gets their own folder where they can create their projects and their experiments, experimental steps and so on. And you can control the access to this folder so again you can share it with the other with your colleagues or collaborators and you can decide what type of access you want to give it. If it is read-only access or if you want other people to also be able to write in your own folder. And this is as I said by default but in some cases there are some groups that prefer to have an organization by project rather than by person and this is also possible. This is supported by OpenBeast. And then you have here some forms where you can describe your experiment. This is one of the default forms provided by OpenBeast that is called experimental step. And in OpenBeast you can also create, you see this section at the bottom here that is called parents. So basically these are connections to other entities that you have stored in OpenBeast. So for example this is a flow cytometry experiment and I can connect it to the samples that I use to measure in this experiment to the protocols that I followed and to any other materials that I have used. And these things would be stored in the inventory and I can make these connections and then OpenBeast builds this graph so you can see the history of what has been done. And in OpenBeast we also have an audit trail so whenever you modify something this is tracked in the database and you see it in the history of the entity. So you can have access to this table that is called entity history and you see all the modifications that have been done by home and when they were done. And also when you delete something so you can delete if you have admin rights so it is possible to delete things. However, everything again also here is tracked in the database so you will always see whenever something is deleted you will always see what has been deleted by whom and when. We come now to the data management part and we see that basically in OpenBeast you have data always connected to some entity so it is either connected to an experimental description or in some cases maybe it makes more sense to attach the data to the samples that you have measured so it depends on the data model essentially. But the data has to be connected to something you cannot just have data lying around without any description. How do we upload data to OpenBeast? We have different ways to do that. The simplest way would be to upload it via the web interface so you have an upload button and you just click on that and then you will upload your data. And this is fine if the data are not too large so we are talking about a few gigabytes. Then you have other options like the Python application programing interface which is called Pybis or Obbis which is our command line tool so if you prefer to work programmatically then this would be a good option for you either using the command line or Python scripts or the third option, the fourth option when you have large data sets this would be more recommendable is what we call the OpenBeast Dropbox and this is not to be confused with the commercial program Dropbox. This is essentially a folder where you would put your data and this can be either a manual process or an automatic process so the data can be directly transferred from a measuring instrument to the Dropbox folder. And then OpenBeast has access to this folder. It constantly monitors it and whenever it finds that there are data in this folder when this data will be transferred to the final storage of OpenBeast and in all cases then you will have the data will end up in OpenBeast attached to as a set either an experiment or a sample. So now we come to the data analysis part so OpenBeast itself doesn't provide anything for data analysis but what we do is we provide a connection to some tools that are commonly used in science for analyzing data and these tools are the Jupyter Notebooks and MATLAB. So the Jupyter Notebooks this is a picture of one if you're not familiar with them they are basically documents that combine text code and the results of your code so you have everything in one single document. They are very powerful they support more several languages more than 40 languages among which are in Python which are there in our experience the most commonly used in the scientific environment we work with and in OpenBeast basically what we do is we can provide a Jupyter Hub server that can be connected to an OpenBeast instance and what you can do is that you can launch basically notebooks Jupyter Notebooks from within OpenBeast so you can open notebooks at different levels and this will also be seen later during the day. Rosti has a training session on this and the idea is that you open the notebook to analyze data that you have stored in OpenBeast and then once you have finished you will store the notebook back to OpenBeast connected to the data that you originally analyzed. It is also possible to use local Jupyter installation so not necessarily use the Jupyter Hub server and then using this tool you can basically connect you can have an extension and then you can connect to OpenBeast and download your data from OpenBeast do the analysis and upload the notebooks back to OpenBeast and for MATLAB we provide a tool that essentially allows you to do the same so you can connect to an OpenBeast instance download the data that you want to analyze run your analysis write your MATLAB script your MATLAB code and then upload everything back to OpenBeast. So for the data preservation at ETH we provide a connection to the tapes for the long-term storage but this is not at ETH we provide this as a service but anyone in any places where a similar backend is used for tapes this would also work so what we do is we provide a link to strongbox and for the data publication we provide a connection to two data repositories at the moment one is the ETH research collection and the second one is the nodo so ETH research collection is of course only relevant for the users at ETH this is the data repository of ETH and essentially what you can do is you have an export to the research collection you have a tree where you can select what you want to export so the idea is that you will have all your data in OpenBeast because this is the tool that you use daily where you store everything and then when it comes to the moment when you need to publish this data share this data with the community you have an easy way of doing this so you can select what you want to what you want to share what you want to transfer and then you will need to provide details on what type of submission disease and also the retention time so for how long the data should be stored in the research collection and then when you click on this export selected what happens is that in the background a zip file is created with the things that you have selected and it is transferred to the research collection and you are also transferred to the interface of the research collection where you have to finish your submission because you need to provide additional information such as for example the licensing or other information other metadata and in the connection to Zenodo works in a very similar way so you have these export to Zenodo option and basically what you would do is that you select what you want to export in this case you need to provide the title of the submission and then again when you click on export selected what happens is that a zip file is transferred in the background to Zenodo and you are redirected to the Zenodo interface where you would need to finish your submission so I hope that I have shown you that OpenBIS can help you in the different phases of the help you and support you in the different phases of the data life cycle so now I just have a few words on the services that we provide so we provide OpenBIS as a service inside ETH we started doing this in 2018 and essentially we have three types of services one is called the research data hub which is basically one OpenBIS instance where any of the ETH groups can request access these are some limitations because the groups cannot have administrative rights here so this is suitable for groups that do not require extensive customizations and also for groups that are not working with sensitive data similar and this is a free service that we provide to ETH then there is the departmental data hub which is fairly similar but so different groups can get an account on this instance and this is this would be dedicated to departments so at the moment we have only one of these instances at ETH so the groups of a particular department can have an account so it can be customized this instance for the needs of the group of the of the department again it's not really suitable for sensitive data and finally we have the last option which is the research data node and this is a private instance let's say so it is an instance that is dedicated to one group or one project also and in this case the group can have administrative rights to this instance so it's more suitable for those cases where one needs to have extensive customization of the system and also it is suitable for working with sensitive data because in this case what we would do is that we would actually provide this instance inside our secure infrastructure that is also managed by us which is called Leomed and then we provide services also for the Swiss academic community so not only at ETH but also in Switzerland in general and actually these services were established thanks to a program a project by Swiss universities and the project ran between 2019 and 2020 so we officially started the service in 2021. What we do with this service is that we provide OpenBison as a solution hosted on the cloud so this is as a cloud provider we use switch engines and what we can do is that we can provide one OpenBis either for one group or for several groups for a department or an institute and optionally these OpenBis can be connected to a Jupyter Hub server. There are we also have cases where some institutions prefer to have a local installation of OpenBis but maybe they require our support so in that case what we can do is we can have a small support contract and basically we can provide support both technical support and user support for the for using OpenBis and getting started with OpenBis. Then we always provide training and best effort user support and here you see a few examples of the current customers. So what do we provide with these services? Well we take care as a set of the initial installations either on ETH infrastructure or on switch engines depending on who are the customers whether it is ETH group or non-ETH group. We take care of the maintenance of the infrastructure of the upgrades of the of the software and so on. We provide consulting and we can provide tailored data modeling so at the very beginning when you start setting up your your system we can help you with the customization and finally we provide regular user trainings and user support. So this is a brief overview of what OpenBis can do for you and how we from SIS can help you in using OpenBis the services that we provide here on the last slide I have some contacts and useful information like the website of OpenBis website of SIS and then where you can contact us and this brings me to the end of the presentation and if you have any questions I will be happy to take them.