Variant Data Access

  • We will allow participating laboratories to initially upload their variant classifications into the system without turning on the sharing mechanism. This means that variant data will not be shared by them or with them until the laboratory has indicated their willingness to share data.
  • We intend to support programmatic access for all variant classifications (without text-based interpretation summaries) so that labs can integrate this information into their bioinformatic pipelines. Laboratories will only have access to their own data and to the consensus data of the centralized COGR instance. A comprehensive list of shared data points is located below under “Aggregate Variant-Level Data Fields”.
  • Each laboratory retains rights to their own data and this data can only be accessed and downloaded by authorized users within that laboratory. Individuals from the project management team are authorized to access each instance including all internal data with permission, in order to assist individual labs and to generate discrepancy reports.  Data uploaded to the COGR’s centralized repository is automatically attributed to the contributing laboratory.  If you choose to share your data with the COGR network, your variant level data can be viewed by other  laboratories on a variant by variant basis. Individual laboratories can only download their own variant datasets and not variant data from other sharing laboratories. This function exists to discourage the download of other laboratories datasets without authorization, i.e., no bulk downloading (aka. ‘leaching’) of data from other members is allowed.. If members are found to try and exploit data and the principles of the COGR initiative, the members will be warned, censured and if necessary asked to leave.
  • Variant consensus data (decided on by consensus workgroup meetings), will be made available for use by consortium members through the ‘consensus’ COGR instance. This consensus variant data will be made available to patient advocacy groups in an aggregated form on the COGR website ( Specific laboratory related data or other information will not be made available to the broader community.
  • Variant interpretation summaries must be viewed manually through the user interface. Each laboratory must make a variant by variant decision whether to “promote” another laboratories variant interpretation and make it their own. The purpose of this rule is to discourage wholesale copying of another laboratory’s knowledge base. Each laboratory must review and accept other laboratory interpretations before incorporating them into their own knowledgebase.
    • Rules for use: It is acceptable to bring in and use (with or without modification) variant interpretation summaries from clinical reports without acknowledging interpretation history. However, other uses require permission (e.g., publications, abstracts, third party sharing). The system will track importation of variant interpretation summaries. Note: If several labs contribute to a variant interpretation, all can claim full ownership. A future enhancement will show better visibility of the complete audit trail of a variant’s interpretive history.
    • It is recommended that if a lab uses any networked data to inform their own interpretation they should annotate what date and from what lab they imported the data.
  • You may withdraw from the collaboration and have your variant data removed from the system at any time. To withdraw from the consortium you can do so by notifying the COGR management team. Any variant information that is shared and used to form a consensus variant interpretation will remain part of the COGR consortium.
    • If a lab leaves the network, it is acceptable for them to maintain information on the variants they created or reported using the system, but other information (e.g., interpretation/classifications on non-reported variants) from the network must be removed/destroyed. It is the labs decision whether they would like to have their data removed from the network or if they would like to leave their data in the network with a note that this variant will no longer be updated.
  • In the event the COGR Consortium is terminated, a ‘data migration’ plan will be formulated by discussion. This will guide the preservation of the consensus data and data already in the public domain for future use by the broader community. Members will be able to download their data without restriction and assistance offered with bulk extraction from the individual member’s instances if needed.
    • Future builds of GeneInsight will enable individual laboratory variant data or COGR consensus variant data to be pushed to ClinVar or other public databases by agreement of the data owners.

Leave a Reply

Your email address will not be published.