Multi-omics and artificial intelligence to fight cancer

Abstract

We now have many modalities to explore the tumor and its environment: many types of omics, imaging, clinical records... One burning question is to convert this heap of data into knowledge useful in clinics. This requires ad hoc algorithms which are capturing the underlying biological nature of the data while being computationally efficient which means keeping a reasonable level of complexity. I will present some examples of our research in that direction: -How to reconstruct evolutionary path from clinical data? -How to integrate multi-omics to gain biological insights and clinical perspectives in a devastating pediatric disease: medulloblastoma. -How to model the spatial dimension of the tumor, a critical though neglected aspect of cancer.