Additional file 3: of Integrative analysis and machine learning on cancer genomics data using the Cancer Systems Biology Database (CancerSysDB)

Figure S2 Interactive workflow of mitochondrial pathways. Shown is the Tricarboxylic acid cycle (TCA) pathway for KIRP cancer patients. The central view of this workflow is a bee-swarm scatterplot, which contains the averaged log2-fold changes of patient groups according to either tumor stage, gender or vital status. Each dot is represents the averaged log2-fold change of one gene that has been assigned to the chosen function. Functions can be selected on the right-hand side of the scatter plot. The dashboard below the scatter plot can be used to change the averaging according to a different feature ((a), which shows averaging according to stage), to display information on the composition of the selected feature ((b), which informs the user that all individuals of stage II, which was hovered over in this case, are male and that one individual is dead, while three of the patients are alive); or to further select individual patients and thus modify the averaging shown in the scatter plot ((c), where only female patients were chosen for stage-dependent averaging; as female patient data are only available for two stages (I and III), the scatter plot is changed accordingly). (PNG 679 kb)