A supervised approach to generate robust cardiac organoids

Organoids efficiently form complex structures, improving our ability to study organ development. The embryonic heart is inaccessible and current models fail to accurately represent the effects of mutations, medications, and environmental factors on the specialized functions of different cardiac compartments. The use of in-vivo studies for a better understanding of early mammalian embryonic development has multiple limitations. In-vitro embryo-like models based on embryonic stem cells have improved greatly in recent years and can be used to study and manipulate early developmental processes. Here, we aim to establish a cardioid platform that recapitulates cardiogenesis, the formation of different heart cells and morphogenesis of the heart parts. We employ embryonic stem cells carrying fluorescent reporters to live monitor the formation of the heart tissues in gastruloids. We are going to use a supervised machine learning approach to characterize the factors affecting cardiac morphogenesis in these gastruloids. By analyzing the learned models, we will devise interventions that will reduce the inter-organoid variability, to provide robust, reproducible cardiac organoids.