Cédric Vincent-Cuaz

ProtScape learns context-specific protein interactomes across 200 cell types, improving interaction reconstruction and identifying disease-related mechanisms.

The paper introduces SI-SimCLR, a self-supervised learning framework for reconstructing cellular phenotypic landscapes from microscopy data without segmentation or annotation, revealing substate organization and connections in astrocytes.

This paper extends pre-training of histopathology foundation models to specialize them for tumor tissue, revealing two coexisting breast cancer archetypes defined by RNA splicing and TGF-β dysregulation.