Lude Franke

UMC Groningen

In short, the research of Lude Franke entails the following:

Identification of (ultra-)rare functional non-coding somatic mutations in cancer using sequence-based deep learning models

This project aims to uncover how ultra-rare non-coding mutations contribute to cancer development. While most research has focused on the 2% of the genome that codes for proteins, the remaining 98% - the non-coding genome - remains largely unexplored. Using advanced AI-driven, sequence-based deep learning models, this project will improve the detection of non-coding somatic mutations that alter gene regulation. By analysing more than 24,000 whole-genome sequenced tumors and integrating insights from large eQTL studies, the research will identify which non-coding variants have real downstream molecular consequences. The improved models will then be applied to several rare cancers, with the goal of pinpointing novel non-coding driver mutations, understanding their biological role, and assessing their clinical relevance for prognosis and treatment. Ultimately, this work supports earlier diagnosis, more accurate risk assessment, and improved therapeutic strategies for patients with rare cancers.