Sharing ARCs via the DataHUB

In this guide we explain the different ways offered by the DataHUB to share an ARC with collaborators.

☑️ You have a DataPLANT account
💡 See the DataHUB manual for tutorials

Before jumping into the details of data sharing, we need to introduce a few DataHUB concepts, settings and terms that tend to be confusing for new users.

Project = ARC

• In the DataHUB, ARCs are called "projects"; they are the same.
• An ARC can be shared with individual users (invited as "members") or a group.

DataHUB Groups

• A "Group" is a group of users (then "members" of that group) with specific permissions.
• A group (e.g. your lab or consortium) can share ARCs.
• Groups help you communicate with all group members and manage permissions for the ARCs shared in that group.
• Depending on their assigned roles, members of the group can view, edit, delete, and / or maintain the ARCs of that group.
• If someone has access to a group, they get access to all the ARC in the group.
• The group Owner(s) can adjust group settings and manage group memberships.
• Groups can have subgroups (with a subset of members). Subgroups have the same features as groups.

• In the DataHUB, namespaces help organize related projects.
• Every user has a personal namespace, where they can upload or create new ARCs.
• Every group and subgroup has an own namespace, respectively.

You can check the namespace by looking at the URL. For example:

💡 "Personal" is not the same as "private", see visibility.

The visibility of ARCs and groups can be managed individually for each ARC (see ARC settings) or group see (Creating a Group).

• Private – Access must be granted explicitly to each user or a group.
• Internal – Can be accessed by any logged in user.
• Public – Can be accessed without authentication.

💡 By default every ARC and every group is set to private.

If you create or upload an ARC to the DataHUB, you are the Owner by default. When inviting new members to an ARC or group, you can choose between different levels.

• Guest – Has the least rights. They will not be able to see the content of an ARC (only the wiki page).
• Reporter – Has read access to your ARC. This is recommended for people you ask for consultancy.
• Developer – The choice for most people you want to invite to your ARC. Developers have read and write access, but cannot maintain the project on the DataHUB, e.g. invite new members.
• Maintainer – Has read and write access and can add or remove members, except for yourself. This is the minimum role recommended for inviting group leaders as it allows them to add their group members for data upload or analysis to the project.
• Owner – An owner has basically all rights to the ARC. They can delete or transfer the ARC and manage user memberships.

💡 A detailed list of all permissions for the individual roles can be found here

Data in the ARC can be annotated with metadata. Metadata annotation in the ARC is based on the ISA metadata model and includes several points to add the the contributors to the overall investigation (→ "Investigation Contacts"), one linked study (→ "Study Contacts" of that study), or one linked assay (→ "Assay Performers" of that assay). For all three types of contributors, contact information (e.g. first and last name, affiliation,ORCID, email) can be added.

Although the contacts added to the ARC's metadata may be the same persons as those added as members to the ARC in the DataHUB, there is currently no automation involved:

• Contributors added to an ARC's metadata are not automatically invited to or notified about the ARC in the DataHUB.
• And vice versa, members invited to an ARC in the DataHUB are not automatically added to any ISA metadata in the ARC.

Creating an ARC with contributors and uploading it to the DataHUB is conceptually, the same as creating a manuscript with authors for publication and sharing it via any cloud service. You can invite members to your DataHUB-shared ARC just as you can invite collaborators to your cloud-shared manuscript and these may not be the same legal people as added to your investigation contacts (ARC) or author list (manuscript).

💡 Members, groups, and roles set via the DataHUB can easily be confused with contributors added as ISA metadata to an ARC, e.g. collaborators added to investigation, study or assay. These are not the same and they are not automatically synced.

To suit a wide range of collaborative scenarios, the DataHUB offers multiple, flexible ways of sharing an ARC. This flexibility can be confusing at first. Generally, ARCs can be uploaded to the DataHUB and (a) associated to the namespace of a user account (the default) or (b) the namespace of a group.

The figure below is supposed to give a quick overview. Once an ARC exists in the DataHUB, you can choose between these options to share the ARC with collaborators.

1. You can invite individual users (1). This is helpful to share an ARC with selected colleagues.
2. To share an ARC with a group of users (e.g. a lab or consortium), you can invite a group (2).

In both cases (1 and 2), the ARC "stays associated" with the original owner only (visible by the namespace and URL address). Furthermore the roles and permissions can be set for individual users and groups.

3. Alternatively, you can create a fork of your ARC (3). This generates a copy linked to the original ARC, but now associated with the group. This can be used to share an ARC at a certain stage, without sharing the full progress after that stage. However, since the two ARCs (the original and the fork) can now be developed independently. This can easily lead to divergence and requires a bit more technical expertise to keep both ARCs in sync (if desired).

4. Finally, you can transfer your ARC to a group (4). This moves the ARC to new namespace (that of the group).

In both cases 3 and 4, you must have at least maintainer access to the group. If you have maintainer access, you can also directly create or upload an ARC to a group.

DataPLANT Support