Learning interpretable networks from large scale biological and clinical data
Single cell transcriptomics and bioimaging techniques produce massive amounts of data at multiscale levels. However, this wealth of heterogeneous data remains largely under-explored due to the lack of unsupervised methods and tools to analyze them without preconceived hypothesis. This highlights the need to develop new Machine Learning strategies to better exploit the richness and complexity of the information contained in biological and biomedical data. In this talk, I will outline some network reconstruction methods and applications to a wide variety of biological and clinical datasets. In particular, our group has developed reliable machine learning methods to infer interpretable causal networks from large scale cell biology data as well as biomedical data. I will present applications to single cell transcriptomics, live-cell images of ex vivo tumor ecosystems and the analyses of medical records of elderly patients with neurological disorders and of breast cancer patients.
Physico-chimie Curie (UMR168)
Laboratoire d'imagerie translationnelle en oncologie (U1288)