Figure 1

Schematic representation of MUDPAC. In Step 1 a ranking score is calculated as a function of Mutation Factor (MF) and Interaction Factor (IF). MF is essentially the difference between non-synonymous and synonymous mutational impacts averaged across all samples; IF takes account of the mutational landscape in the functional vicinity of the gene being scored. A ranking gene list is then computed based on the score assigned to each gene (darker red indicates higher score), which is then used to find candidate pathways. In Step 2, the coverage rates of the Top 60 candidate driver pathways are calculated based on the number of samples where there is at least one non-synonymous mutation in the corresponding pathway. Greedy algorithm is applied to find a set of pathways whose composition and pattern, uniquely identifies a breast cancer subtype in a maximum number of samples.