README.md

Project configurations for the Data Curator App

This repo contains DCC configurations for the Sage-hosted multi-tenant Data Curator App. The prod and staging branches configure their respective version of the app. To contribute, fork the repo and create PRs into the staging or prod branch, depending on which app you want to update. The staging and prod branches are maintained separately; they have diverged and are not intended to be merged, so make sure to start changes from the desired branch.

The file tenants.json controls which DCCs appear in DCA. It contains a json object tenants which holds one json object per DCC. Each DCC object requires:

• name: The name of the DCC to appear in the DCA selection menu
• synapse_asset_view: The synapse ID of the DCCs fileview for DCA. Must include "syn"
• config_location: Filepath to the DCC's DCA config file. More details below.

Each DCC will have its own directory to store configuration files. At minimum it will have dca_config.json. This file can be validated against schemas/dca_config.schema.json. At the top level it contains three json objects, dcc, dca, and schematic.

dcc contains various configurations about your DCC and data model:

• name: name of DCC in dropdown menu
• synapse_asset_view: synapse ID of asset view.
• data_model_url: data model URL (RAW github file).
• data_model_info: URL to a description of the data model, such as release notes.
• template_menu_config_file: URL to DCA template dropdown file.
• logo_location: URL to logo file.
• logo_link: link to DCC website.
• dcc_help_link: link for users to find help about DCA.
• portal_help_link: link for users to find help about their portal.

dca contains DCA-specific customizations:

• use_compliance_dashboard: TRUE or FALSE. Only FALSE currently supported. Data compliance dashboard.
• primary_col: center header color hex code.
• secondary_col: right header color hex code.
• sidebar_col: left header color hex code.

schematic contains several objects for passing parameters to schematic commands:

• global

  • data_model_labels: class_label or display_label Schema labelling option

• manifest_generate

  • output_format: excelorgoogle_sheets. Schemtic manifest get option.
  • use_annotations: TRUE or FALSE. Schemtic manifest get option.

• model_validate

  • restrict_rules: TRUE or FALSE. Schematic model validate and submit option.
  • enable_cross_manifest_validation: TRUE or FALSE. Use cross manifest validation.

• model_submit

  • table_column_names: class_label, display_label, or display_name Schema labelling option
  • annotation_keys: class_label or display_label Schema labelling option
  • table_manipulation: replace or upsert. Schematic model submit option.
  • manifest_record_type: One of table_and_file, file_only , file_and_entities or table_file_and_entities. Schematic model submit option.
  • hide_blanks: TRUE or FALSE. Schematic model submit option for blank annotations.
  • file_annotations_upload: TRUE or FALSE. Annotate associated files/entities with values in the manifest. Schematic model submit option.

Each DCC also needs a template configuration file. This controls which data types are accessible to users and if they contain record-based or file-based data. It can be generated by hand or automatically from a data model using the dca-template-config-action github action. See an example workflow in a data model repo. This json file can be validated against schemas/dca_template_config.schema.json. The top level of the file needs a json array of objects named manifest_schemas. Each object in the array requires the strings:

• display_name: Display name of the data type
• schema_name: Name of the schema
• type: record or file. If the manifest has Filename as a column, type should be file. When annotating records inside a spreadsheet-like file, such as specimen or patient information, type should be record.

See our example dca_template_config.json