Fig. 1

RVS architecture and workflow. All new variant data in VCF format gets populated into a staging area, where novel variants are registered with RVS. Novel variants are exported to the compute cluster for annotation with snpEff etc. Data are imported back into the production tables of RVS. Large studies will also trigger the upload of (sub)population frequencies. Variants in RVS are assigned to each new or updated source, allowing multiple sources per variant