ABSTRACT
The transition from a normal cell to a cancer cell is driven by genetic alterations, such as mutations, that induce uncontrolled cell proliferation. This chapter describes NetNorM, a method proposed by Le Morvan et al., that transforms mutation data using gene networks so as to make mutation profiles more amenable to statistical learning. NetNorM was shown to significantly improve the prognostic power of somatic mutation data, and to allow defining meaningful groups of patients based on their mutation profiles. The chapter shows that NetNorM normalization improves survival prediction and patient stratification in some cases, and proposes the interested reader to the original publication for more details. Exploiting the wealth of cancer genomic data collected by large-scale sequencing efforts is a pressing need for clinical applications. Systematically assessing and monitoring somatic mutations in cancer therefore offers the opportunity to help rationalize patient treatment in a clinical setting.
