What is Cafe Variome?

There is a considerable need and desire for networks of diagnostic laboratories/disease consortia to be able to check each others databases for the presence of mutations and/or phenotypes they observe in their own patients. But this is countered by the understandable reluctance and impracticality of sharing the content of each group's database with other labs, or indeed with the wider world.

Given the above, a way forward has been devised based upon enabling the 'open discovery' of data (rather than data 'sharing') between networks of diagnostic laboratories or disease consortia that know/trust each other and share an interest in certain causative genes or diseases.

To that end, the Cafe Variome platform has been developed. We are reaching out to groups and projects that could benefit from such a system, allowing them and their fellow labs to securely share data between each other.

Sharing Policies

Once data has been discovered it can subsequently be accessed under three different models, to give data owners fine-grained access control over their data. Cafe Variome should not be thought as a database but instead as a "shop window" for what exists in various data sources. It is designed to enable users to ask the question "where can certain data be found?", seeking to access those data under one of the three models listed below, as stipulated by the data owner/submitter:

Sharing Policies Summary

Variants can be released using one of the following three sharing policies (fine-grained access control in the Cafe Variome interface allows data owners to modify permissions for whole datasets down to the individual variant level):

openAccess: like a journal, where variant records are publicly available for access

restrictedAccess: the user requests access to variant from the data owner/submitter, the user must either belong to a pre-approved group or must request access from the data owner in order to access the variant record(s)

linkedAccess: only reports the existence of a variant and the user is linked to the source database in order to access the full record