Fig. 2: Workflow depicting the statistical pipeline to detect genetic interactions.
From: A systematic analysis of genetic interactions and their underlying biology in childhood cancer
Starting with the original mutation files from the TARGET and DKFZ data set, samples and alterations are first filtered after which a mutation matrix is created to record which samples contain mutations in a certain gene. Here, the candidate gene pairs JAK1–JAK3 and PTCH1–SMO are shown as examples. The matrix serves as input for the Permutation and WeSME test to detect significantly co-occurring or mutually exclusive pairs of altered genes. The resulting candidates from both data sets are merged into a genetic interaction map of childhood cancer.
