Fig. 1

Illustration of phenotype files available in large cohort datasets. There are multiple files that contain the phenotypic information but for the sake of simplicity here we only show two. The green squares represent the header of the file which contains the column names, and the turquoise-coloured squares represent each individual in the study (i.e. the participant ID). In a study, we may wish to use data stored in the columns outlined in blue in File 1 and the columns outlined in purple in File 2. The researchers would combine the files into a new file using the participant ID. Once the phenotype file is created, the data often needs to be pre-processed before use. For example, in the WHI dataset, occurrence of cancer is recorded as number of days into the trial which cancer was discovered, however most studies will need this information in the format of age diagnosed