Self-supervised deep learning uncovers the semantic landscape of drug-induced latent mitochondrial phenotypes
P. Natekar, Z. Wang, M. Arora, H. Hakozaki, J. Schoeneberg
This study introduces MitoSpace, a self-supervised deep learning approach to map drug-induced mitochondrial phenotypes, revealing a more complex morphology landscape than previously thought.
bioRxiv · Sep 13, 2023 · PreprintPharmacology & Toxicology