Schematic diagram of the dataflow between FMAj and the relational database. (a) FMAj generates two types of image descriptors that are stored in the database. First, metadata of the time-series images contain biological parameters of the live samples (e.g. genotype and fluorescent marker) and details about the imaging experiment (e.g. objective, laser excitation wavelength, pixel size). Second, descriptors of segmented muscle ROIs contain polygons, shape features, annotations provided by the user (e.g. cell type) and tracks, i.e. assignments between ROIs in subsequent frames. In the data flow diagram, forward arrows indicate storage of data while backward arrows indicate retrieval of data. Raw and processed images are stored on the local hard disk. Their file locations and metadata are stored inside the MySQL database. (b) Simplified version of the entity relationship diagram of our database. Each block represents a table and each line indicates a relationship between tables. The notations at the end of lines define the type of relationship, such as one (1) to one (1) or one to many (M). For example, the muscle cell details table has a One to One relationship with muscle cell type. Table names in orange store experimental and biological metadata, while those in green store results from image segmentation like the annotation or calculated features of muscle ROIs.