Figure 4

The Age-Phenome Database's overall structure. The Age-Phenome database comprises of 4 main data-tables: (1) Evidence, (2) Evidence-age, (3) Evidence-phenotype, and (4) Age-Phenotype relationships. In the Evidence table each evidence instance has a unique identifier and is characterized by: 'Evidence type', 'Evidence text', 'Analysis description' (for evidence computationally derived from clinical data rather than literature), 'Title', 'Publication', 'Publication type', 'Publication year', 'MeSH terms' (for evidence taken from literature), 'ID in source', 'Source name' and 'Source URL'. The 'Evidence text' field holds an evidential text fragment. In the evidence-age table, each entry records the age or age range that the evidence instance is linked to, using 'range min' and 'range max' fields to describe ranges as well as single ages (using identical range min and max values). In the evidence-phenotype table, values for the 'disease' and 'Additional phenotypes' fields are taken from the Disease Ontology, and a locally compiled additional phenotypes list (phenotypes such as different blood, available as Additional file 4), respectively. In the age-phenotype-relationship table, each evidence instance is linked to one of five different relationships ('Age of Onset', 'Age of Diagnosis', 'Age of Occurrence', 'Age of Observation' and 'Age of Evaluation').