Connecting diagnostic labs: Cafe Variome and DNA sequencing software

During the FP7 project GEN2PHEN (http://www.gen2phen.org/) we developed the concept of sharing pointers to diagnostics data to facilitate the public exchange of the existence of information, avoiding the problem of directly exchanging diagnostics information. A minimal set of data was defined to point to variant/mutation information generated by diagnostics laboratories: required data are exact genomic positions of the detected variants and the source (which lab or user generated the data, or publication or public data base), optional information may include the amino acid or protein change, phenotype information, or an anonymised patient identifier. The project also developed a standard to exchange the information: VarioML (http://varioml.org/). We present Café Variome, which has been implemented as a federated market place for these pointers: a public instance has been created (http://www.cafevariome.org/) and populated with public data (such as dbSNP, 1000 Genomes…) or pointers to data (HGMD, LSDBs…). Users can either chose to submit directly to this public instance, or setup a local instance to share detailed data with selected collaborators (such as a national network or a disease-specific collaboration) and connect this instance to the public instance, sharing only pointers. The system has comprehensive management tools for fine-grained user control as well as data sharing control. Because manual data submission is error prone as well as a major barrier to usage, we implemented public APIs to allow both the submission and automated query automatically of data via secure internet connection. We will present two examples of this automatic transmission of pointer information: Gensearch for capillary DNA sequencing and GensearchNGS for NGS DNA sequencing.