Genotype-phenotype databases provide information about genetic variation, its consequences and its mechanisms of action for research and health care purposes. Existing databases vary greatly in type, areas of focus and modes of operation. Despite ever larger and more intricate datasets--made possible by advances in DNA sequencing, omics methods and phenotyping technologies--steady progress is being made towards integrating these databases rather than using them as separate entities. The consequential shift in focus from single-gene variants towards large gene panels, exomes, whole genomes and myriad observable characteristics creates new challenges and opportunities in database design, interpretation of variant pathogenicity and modes of data representation and use.