A list of Gen3 features

Gen3 is an open-source platform that enables the standing-up of data commons for managing, analyzing, and sharing research data. Below you can find an exhaustive list of Gen3 features.

Dictionary and structured data

• Explore the graph in the dictionary page of Windmill:
  • Search keywords
  • Visualize possible paths between nodes
  • Download JSON or TSV format for node templates
• Explore project graphs and nodes in Windmill:
  • Visualize graph model
  • Browse/download data for each node
• Structured data submission (create, update, delete records) in JSON, TSV or CSV:
  • Submission page of Windmill
  • Gen3 SDK (Python, R)
• Export structured data as a JSON, TSV or CSV file
• Structured data is stored in Postgres databases
• Flattened data:
  • Structured data can be flattened and stored in Elasticsearch by using our ETL tool
  • Explore the Elasticsearch structured data using Guppy
• Generate fake data specific to a dictionary for testing
• Import and export using the Portable Format for Biomedical Data (PFB): (Work in progress)
  • Import and export a dictionary
  • Import and export structured data

Data files

• Data upload through the Gen3 Client
• Indexing data files:
  • Create data file records and mint GUIDs
  • Index a new version of the same file
  • Get the most recent version of a file
  • Index data files that are stored outside of the Commons
• GUID resolver
• Data download:
  • Gen3 Client
  • Files page of Windmill
• Core metadata endpoint to get information about a data file
• Metadata Service

Data exploration and analysis

• GraphQL endpoint and UI to explore structured data via Peregrine (graph data)
• GraphQL endpoint and UI to explore flattened data via Guppy (in ElasticSearch)
• Explorer page of Windmill
  • Filters to select a cohort
  • Charts and visualizations: count, pie charts, bar charts, stacked bar charts, heatmaps
  • Download clinical data for the selected cohort
  • Download a manifest for the selected cohort
  • Export the selected cohort to Workspace/PFB/Terra
• Workspaces using Hatchery, Workspace Token Service, and Manifest Service
  • Jupyter notebooks (Python, R)
  • RStudio
  • Publish Notebooks as .html version on the Resource Browser. See here for an example.
• Discover Datasets at a high-level on the Discovery Page. See here for an example.
• Run complex CWL analysis workflows (Work in progress)

Access management and security

• Gen3 follows the OpenID Connect (OIDC) specifications
  • Use OAuth clients
• Gen3 supports logging in through Google, NIH, ORCID, Microsoft and Synapse
• Create precise roles, policies and groups of users to limit access to resources
• Assign policies to users:
  • Individual users
  • Groups of users
  • All logged in users
  • All anonymous users

Automation

• An open source collection of tools to simplify and automate the creation and management of a Gen3 Data Commons
• Use Compose Services to create a Gen3 Data Commons sandbox
• Kubernetes job dispatcher
  • Automatically index new files in a bucket