Implementations

The libraries and applications listed below have implemented the GA4GH Variation Representation Specification to store and exchange variation data. They are listed here to demonstrate utility and as a resource for those considering implementing VRS. These packages are not supported by GA4GH.

Libraries

Libraries facilitate the use of the VRS, but do not implement a particular use or application. Although there is only one library currently, it is expected that others will eventually appear as VRS is adopted.

vr-python: GA4GH VRS Python Implementation

The GA4GH VR Python Implementation is an implementation for the GA4GH VRS. It supports all types covered by the VRS, implements Allele normalization and computed identifier generation, and provides “extra” features such as translation from HGVS, SPDI, and VCF formats. See vr-python notebooks for usage examples.

VRS MAY be used without using the Python implementation.

Applications and Web Services

Applications implement VRS to support specific use cases. Projects known to implement VRS are listed below. Descriptions are provided by the application authors.

ClinGen Allele Registry

ClinGen Allele Registry [1] provides identifiers for more than 900 million variants. Each identifier (canonical allele identifiers: CAIds) is an abstract concept which represents a group of identical variants based on alignment. Identifiers are retrievable irrespective of the reference sequence and normalization status.

As a Driver Project for GA4GH, ClinGen Allele Registry implements two standards: RefGet and VRS in the first implementation.

The API endpoints that support data retrieval in this two key standards are summarized in the following table.

HOST: https//reg.clinicalgenome.org/

API Path Parameters Response Format Example  
RefGet        
[GET] /sequence/service-info - Refget v1.0.0 /sequence/service-info  
[GET] /sequence/{id} id => TRUNC512 digest for reference sequence Refget v1.0.0 /sequence/vYfm5TA_F-_BtIGjfzjGOj8b6IK5hCTx  
[GET] /sequence/{id}/metadata id => TRUNC512 digest for reference sequence Refget v1.0.0 /sequence/vYfm5TA_F-_BtIGjfzjGOj8b6IK5hCTx/metadata  
VR        
[GET] /vrAllele?hgvs={hgvs} hgvs => HGVS expression VRS v1.0 /vrAllele?hgvs=NC_000007.14:g.55181320A>T /vrAllele?hgvs=NC_000007.14:g.55181220del  

Support for GA4GH refget and VRS specs provided in ClinGen Allele Registry is independent from VR-Python. Support for this community standards is implemented in ClinGen Allele Registry through extension of code written in C++.

BRCA Exchange

BRCA Exchange [2] proposes an API endpoint which will share the variant list in VRS JSON model. Behind the scenes, all variants will be represented according to VRS, in a separate table of the BRCA Exchange database, and the contents of this table will be served by the BRCA Exchange API. A stand-alone executable will leverage these data to integrate the BRCA Exchange variant set with the ClinGen allele registry.

VICC Meta-knowledgebase

The Variant Interpretation for Cancer Consortium (VICC; https://cancervariants.org) has a collection of ~20K clinical interpretations associated with ~3,500 somatic variations and variation classes in a harmonized meta-knowledgebase [3] (see documentation at http://docs.cancervariants.org). Each interpretation is be linked to one or more variations or a variation class.

As a Driver Project for GA4GH, VICC is contributing to and/or adopting three GA4GH standards: VR, Variant Annotation (VA), and the Data Use Ontology (DUO). VICC supports queries on all VRS computed identifiers at the searchAssociations endpoint (vicc-docs). Features associated with each interpretation are represented as VRS objects.

Example queries:

References:

[1]Pawliczek P, Patel RY, et al. ClinGen Allele Registry links information about genetic variants. Hum Mutat 11 (2018). doi:10.1002/humu.23637
[2]Cline, M.S., et al. BRCA Challenge: BRCA Exchange as a global resource for variants in BRCA1 and BRCA2. PLoS Genet. 2018 Dec 26;14(12):e1007752. doi:10.1371/journal.pgen.1007752
[3]Wagner, A.H., et al. A harmonized meta-knowledgebase of clinical interpretations of cancer genomic variants. bioRxiv 366856 (2018). doi:10.1101/366856