“GrabBlur” is a tool that we have developed to collect, aggregate and share SNV (single nucleotide variant) data of hundreds of samples with a special trait/phenotype in a public database while keeping each individual sample unidentifiable.
First, GrabBlur helps a submitter to combine SNV data from samples with phenotype and sequencing information (VCF files) using a local web interface or command line operations. Examples are HPO (Human Phenotype Ontology) terms, sample traits, information about applied sequencing technology and the identity of the submitter. Most of the information is optional: the submitter can decide for himself what he wants to and can share.
In the second step GrabBlur merges and aggregates the data. For example, if it turns out that an SNV is rare - every submitter sets the proper threshold for his own data - all sample information is replaced with contact information from the submitter. In further steps, the data is highly aggregated so that it can be shared while adhering to data privacy. It is not possible to recover the exome/ genome of a sample or to identify a single individual in the dataset.
After checking that the individual genetic information is sufficiently “blurred”, the data can be uploaded into a public database where the sample sets will be aggregated again, thus adding another “blurring” step.
If another researcher finds an interesting SNV in the dataset he should now be able to get in contact with the original submitter to exchange further information on the carrier and clarify, for example, whether the latter’s phenotype matches with phenotype of their own patient. A webinterface exists for querying the database as well as deriving gene-wise SNV information.
For more information visit our GrabBlur Homepage.