Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

This The DCC has derived standards for how content is annotated in Synapse and this page describes how the DCC manages the content used to annotate files (management of the keys, definitions, and ontological references ). Sage Bionetworks has derived standards for how content is annotated Synapse. The latest approved Synapse Annotations are available in a searchable Table of Annotations. used to annotate files.

Schemas

The synapseAnnotations Github repo includes schemas for individual metadata terms. The Each term is stored in a JSON mini-schema and the terms are organized by modules in the each terms/<module>/ folder. There are subfolders for each module, and the term names include the module. The mechanisms that manage and implement these annotations are described in the README.

Annotation schemas in JSON

Templates for adding new terms are included in the term-templates/ directory. Each individual term is stored in its own JSON mini schema. This is a valid JSON Schema, such as the following:

Code Block
languagejson
{   $schema": "<http://json-schema.org/draft-07/schema#",>
    "$id": "<https://repo-prod.prod.sagebase.org/repo/v1/schema/type/registered/sage.annotations-experimentalData.specimenID-0.0.1",>
    "description": "Identifying string linked to a particular sample or specimen",
    "type": "string"
}

Schema editing procedure

  1. Create Github issue if needed in https://github.com/Sage-Bionetworks/synapseAnnotations/issues

  2. Term information is stored in JSON schemas in the synapseAnnotations/terms directory.
    NOTE: Terms specific to MODEL-AD are stored in: synapseAnnotations/terms/neuro

  3. Create a JSON schema file or edit an existing one.

  4. Copy and Paste schema template.

  5. Edit $id line with Modulename and Key name.

  6. Version numbers start with 0.0.1 and should be incremented with each update.

  7. Add new term, a brief definition, and an ontology citation for the included definition.

  8. Create a branch name formatted like ‘username-edit-type/‘

  9. Add a brief description of changes to commit

  10. Data checks will start to run in the background. All data checks must successfully pass.

  11. Once desired changes are committed, initiate a Pull Request and select DCC curators as reviewers.

  12. Once the changes have been merged into master, need to run script to update the metadata dictionary with the latest version of terms. To register these schemas in Synapse run the register-schemas.R script.

Ontology Lookup Service: https://www.ebi.ac.uk/ols/index

AMP-AD Validation (dccvalidator + dccmonitor data).

Edit JSON Schema

  • Copy a template from: synapseAnnotations/term-templates

  • Edit new file

    • Edit $id line with <MODULENAME> and <KEY>

    • Increment version with each update

    • Add Term, Definition, and a source for Definition (https://www.ebi.ac.uk/ols/index )

    • Commit changes to a new branch

  • Schema validation checks will automatically start to run in the background.

  • Address any errors until validation passes

  • Open a Pull Request and select DCC reviewers

Update Schemas

  • Once the PRs are merged, it is necessary to update the Metadata Dictionary

  • Navigate to: /Users/ryaxley/Documents/GitHub/synapseAnnotations

  • To register the schemas in Synapse run the register-schemas.R script

    • Help: ./register-schemas.R --help

    • Register all schemas: ./register-schemas.R terms/*/*.json

    • Verify if this script runs automatically as part of the Github workflow

  • To update the Synapse annotations table:

    • ./update-annotations-table.R

Generate Template

  • Github sysbioDCCjsonschemas repo

  • SysBio Metadata Templates: https://github.com/Sage-Bionetworks/sysbioDCCjsonschemas/tree/master/schema_metadata_templates

  • Construct a JSON schema file to generate XLS template

  • Make sure to register latest versions of JSON schemas

  • To create a Synapse table of all terms in metadata templates:

    • create_Syn_table_from_Syn_schemas.py

  • For example, to generate a Metadata Template:

    • create_template_from_Syn_schema_sorted.py

  • Code Block
    cd ~/Documents/GitHub/sysbioDCCjsonschemas/code/python/.
    python3 create_template_from_Syn_schema_sorted.py \
            sysbio.metadataTemplates-ad.manifest \ 
            manifest_test.xlsx excel \
            ~/Documents/GitHub/sysbioDCCjsonschemas/schema_metadata_templates/AD/manifest_metadata_template.json