OpenStudyBuilder #2 (Sandbox)

OpenStudyBuilder #2 (Sandbox)

The OpenStudyBuilder project is progressing with a sandbox environment where people can register to play around with the open-source solution without needing to install anything. This provides an excellent opportunity to get a more detailed impression of the metadata repository (MDR). The sandbox allows for investigations on how standards can be maintained, differences identified, how studies can be created with all protocol information and much more. The API is also accessible to see the capabilities for connecting tools.? And lastly, a database browser allows for check out details of the biomedical concept.??

Sandbox

Getting Access

To get access to the Sandbox environment, you can simply send a mail to [email protected] with the subject "Request Sandbox Access". Your e-mail will be used together with Microsoft authentication to access the various tools via browser. Please be aware that your mail might be exposed when you perform changes as all changes are tracked for a version history.

The Application

The core application can easily be started with a browser. The sandbox system has already an example study which can be loaded, browsed and changed. Under the "Define Study" all relevant study protocol and trial domain information can be managed. The "View Specifications" section is available to use the entered information in different formats, e.g. for the protocol.

Screenshot of the study part from the OpenStudyBuilder

The general and sponsor standards can be maintained in the "Library" part of the application. Next to the functionality to browse standards like the controlled terminology from CDISC, also sponsored preferred names can be stored alongside. Extensible codelists can be added by elements and additional new sponsor codelists like for different visits can be created. Dictionaries can be maintained and browsed as well.

A core feature of the MDR and the underlying graph data model are the concepts. Standard activities can be grouped and linked. Even unit mapping for UCUM, CT and sponsor units can be maintained. The "Syntax Templates" are a fantastic feature which allows to use generic texts for protocols with semantic meaning.

Screenshot of the library part from the OpenStudyBuilder

The Documentation

The tool documentation is available via the sandbox as well. This even does not need any authentication as it's a simple webpage which can be accessed here. Next to a user guide, you also find useful information about the architecture.

The API

A main advantage of the OpenStudyBuilder is the openness for interfaces. The API is very powerful. Everything what the application does uses APIs which updates the graph database. This means that everything can be automated via scripts. Data can easily be loaded via the API. All the data and standards data is loaded into the solution via import scripts (mdr-standard-imports and data-imports). These can be adopted to import other data.

All information available can also be exported via API and then used by other tools. When you browse the API Swagger documentation, you can see and execute various API endpoints available.

Screenshot for API usage - get a study

The Graph Database / Biomedical Concept

The heart of the application is the graph database which is using Neo4j. It contains a biomedical concept linking all kind of data which allows a high level of connectivity for automations. The NeoDash can be used to browse the concepts.

Screenshot for NeoDash showing Diastolic Blood Pressure concept

Next to NeoDash there is also the DB Browser. This can be used to browse the Neo4j database directly using the query language Cypher. After logging in via SSO (in case of issues clear cache or do a hard reload of the page), you can select the sandbox database. You can either use a direct query or get along by selecting a node and follow the connections.

Screenshot for Neo4J Bloom Browser showing graph data

Finally, Bloom is also available which allows to create and store Scenes. This is very valuable if you want to come back to a visualization or share this with others. In one of our next newsletters, we are going to have a closer look into the concepts.

Quicklinks for the Sandbox (needs registration)

Documentation and Contact

For more information, please visit the?project homepage. For instructions on how to install all components together or just single components, see the single readme files in the?GitLab repository. Finally, there is also the documentation portal with the documentation coming along with the tool itself – this is accessible through the Sandbox system?here.?Use the opportunity to get access to the Sandbox environment via mail: [email protected] - Request Sandbox Access.

Please follow our newsletter here in LinkedIn, join us in?Slack, write a message to the OpenStudyBuilder team via mail [email protected] or contact us at conferences and similar! In case of issues, you can also use GitLab Issues to report these.

Stuart Malcolm

Work smarter not harder! Standards + Automation = Efficiency

2 年

Looking forward to meeting up with you and Mikkel at Phuse next week.

Volkan Kocak

AI Quality | Implementing GxP compliant Data Science & AI Process in Novo Nordisk | Black Belt in GxP IT Quality Process | Ex-consultant | GxP |

2 年

Thanks for sharing. Which part of an GCP process does this solution support?

回复

要查看或添加评论,请登录

Katja Glass的更多文章

社区洞察

其他会员也浏览了