Modern drug discovery approaches often use high-content imaging to systematically study the effect on cells of large libraries of chemical compounds. By automatically screening thousands or millions of images to identify specific drug-induced cellular phenotypes, for example, altered cellular morphology, these approaches can reveal 'hit' compounds offering therapeutic promise. In the past few years, artificial intelligence (AI) methods based on deep learning (DL) [a family of machine learning (ML) techniques] have disrupted virtually all image analysis tasks, from image classification to segmentation. These powerful methods also promise to impact drug discovery by accelerating the identification of effective drugs and their modes of action. In this review, we highlight applications and adaptations of ML, especially DL methods for cell-based phenotypic drug discovery (PDD).

The endoplasmic reticulum (ER) has evolved multiple mechanisms to maintain homeostasis under stress conditions. A recent study by Efstathiou et al. identified a novel mechanism of silencing ER-associated RNAs by the exogenous RNA interference pathway. This adaptive response reduces protein flux in the ER under stressful conditions.

Lysosomes are essential catabolic organelles with an acidic lumen and dozens of hydrolytic enzymes. The detrimental consequences of lysosomal leakage have been well known since lysosomes were discovered during the 1950s. However, detailed knowledge of lysosomal quality control mechanisms has only emerged relatively recently. It is now clear that lysosomal leakage triggers multiple lysosomal quality control pathways that replace, remove, or directly repair damaged lysosomes. Here, we review how lysosomal damage is sensed and resolved in mammalian cells, with a focus on the molecular mechanisms underlying different lysosomal quality control pathways. We also discuss the clinical implications and therapeutic potential of these pathways.