Cancer remains one of the most complex diseases to treat, especially when tumors become aggressive or therapy-resistant. This project develops deep learning and network science methods to map gene regulatory networks in cancer at unprecedented resolution. By integrating multiple types of molecular data, we aim to reveal how regulatory programs rewire across patients, tumor types, and tumor subpopulations. Applied to large public datasets and the Finnish iCAN cohort, our models are expected to uncover hidden cancer subtypes and drivers of aggressiveness and therapy resistance, linking molecular mechanisms to clinical outcomes and laying the foundation for network-based precision oncology.

2025