Skip to content

Releases: opentargets/json_schema

2.0.6 - Updating patterns for fields

17 Mar 11:53
859f624
Compare
Choose a tag to compare

To accommodate evidence data better, required string patterns got updated.

  • variantId: now the pattern allows mitochondrial chromosomes, and IUPAC ambiguity codes to handle multiallelic variants.
  • studyStartDate: now the YYYY-MM-DD format better handled.
  • studyStopReason: the field no longer allows having leading and trailing white spaces and new-line, CR characters in the text.

2.0.5 - Providing additional links are now available for COSMIC.

26 Feb 15:44
Compare
Choose a tag to compare
Adding urls to COSMIC, it will allow having links to analysis pages o…

2.0.4 - Beta coefficients for Genetics Portal and other updates

23 Feb 17:14
c21a174
Compare
Choose a tag to compare

Updates:

  • beta is added to Genetics Portal.
  • The upper and lower boundaries of the confidence interval are also added.
  • The same definitions are applied for the confidence intervals of beta and odds ratio.
  • alleleOrigins is no longer an enum to better accommodate EVA evidence.
  • diseaseFromSource field is added to Sysbio.
  • In reactome and uniprot_variants, functionalConsequenceId has been changed to variantFunctionalConsequenceId

2.0.3 - Updating accepted fields for ChEMBL and IntOGen

12 Feb 20:53
d11d7d0
Compare
Choose a tag to compare

No major update in this release. Fixes to accommodate ChEMBL evidence:

  • diseaseFromSourceId has been deleted, as the presence of a MeSH ID is always a intermediary mapping step coming from ChEMBL; it is not present in the actual resource.
  • targetFromSource has been added to store those cases where the target is a protein complex. If the subunit interacting with the molecule is unknown, all subunits should be listed. If known, the particular subunit will be included.

Minor changes to intOGen after rewriting the parser:

  • diseaseFromSourceId has been deleted
  • studyId has been deleted

2.0.2 - Supporting multiple pathways

11 Feb 13:51
a893290
Compare
Choose a tag to compare

Support for storing multiple pathways is added.

Example:

"pathways": [
    {
        "id": "R-HSA-2894862",
        "name": "Constitutive Signaling by NOTCH1 HD+PEST Domain Mutants"
    },
    {
        "id": "R-HSA-2644606",
        "name": "Constitutive Signaling by NOTCH1 PEST Domain Mutants"
    }
]

2.0.1 - First release of the slimmed-down schema

01 Feb 15:56
9c6046a
Compare
Choose a tag to compare
  • From the 2.0.0 pre-release, the opentargets.json has been finalised for all data sources.
  • The OT_network_schema.json is added to the master branch

New, streamlined json schema for the Platform rewrite

12 Jan 12:38
Compare
Choose a tag to compare

There are numerous changes in the schema from the previous version.

Changes:

  • The schema is structured by data sources instead of data types.
  • targetId is removed: targets are mapped by OT pipelines. Instead, providers use targetFromSourceId
  • uniqueAssociationFields is removed: evidence uniqueness is defined by OT pipelines.
  • diseaseId is removed: diseaseId is now assigned by OT pipelines. Instead providers can use diseaseFromSourceId, diseaseFromSource or diseaseMappedFromSourceId for diseases where the EFO identifier is available.

Please contact us for further information.

1.7.5

09 Nov 12:58
b292143
Compare
Choose a tag to compare

Variant schema does not require id or rs_id.

Note that the data pipeline may still need the ids for some data sources. This change does not mean that they can be omitted, only that they are not compulsory.

1.7.4

06 Nov 14:15
fcad3d7
Compare
Choose a tag to compare

Changes to variant object schema to support updated PheWAS evidence:

  • Added rs_id to report the rs id of the variant when id uses a different id or notation
  • id is not compulsory anymore, rs_id can reported instead or both if wanted

1.7.3

15 Sep 08:20
07b05a2
Compare
Choose a tag to compare

Multiple changes made to accommodate additional ClinVar information (kindly contributed by @tskir):

  • Enum of clinical significance values updated to match values in most recent ClinVar data release (version 2020-08-24) (PR #95):
    • Typo fixed
    • - removed as it does not appear in the list of ClinVar significance values
    • Values sorted alphabetically
  • Enum used for ClinVar review status and supported both in genetic_association and somatic_mutation (PR #96)
  • mode_of_inheritance supported both in genetic_association and somatic_mutation (PR #97):
    • Verbatim values so that they are not ClinVar-specific
    • Single value and array of strings accepted
  • last_evaluated_date supported for somatic_mutation evidence strings as well (PR #99)
  • gene2variant.date_asserted made optional (PR #99)

All changes have been ported to the draft-04 schema and fields used in multiple places have been moved to definitions to avoid duplication.