DR-369 A method to infer substitution mutational signatures while correcting for local mutation rates biases

Vulnerability of the DNA to damage is influenced by (epi)genomic features and tissue specific properties of the DNA repair machinery. As a result, local mutation distributions differ across cancer types. Currently, implemented methods inferring mutational signatures disregard such diversity, thus limiting the comparison of mutational processes across cancers. We have implemented a solution overcoming this limitation, and plan to use whole genome-sequenced cancer data from HMF consortium to validate our method. We expect to release mutational signatures better suited at describing the activity of mutational processes across cancer types, thereby facilitating the investigation of their etiology and the translation to the clinic.

Maria Cartolano, University Hospital Cologne, Germany