Summary Deep learning is a significant step forward for developing autonomous tasks. One of its branches, computer vision, allows image recognition with high accuracy thanks to the use of convolutional neural networks (CNNs). Our goal was to train a CNN with transmitted light microscopy images to distinguish pluripotent stem cells from early differentiating cells. We induced differentiation of mouse embryonic stem cells to epiblast-like cells and took images at several time points from the initial stimulus. We found that the networks can be trained to recognize undifferentiated cells from differentiating cells with an accuracy higher than 99%. Successful prediction started just 20 min after the onset of differentiation. Furthermore, CNNs displayed great performance in several similar pluripotent stem cell (PSC) settings, including mesoderm differentiation in human induced PSCs. Accurate cellular morphology recognition in a simple microscopic set up may have a significant impact on how cell assays are performed in the near future.

Human embryonic and induced pluripotent stem cells are self-renewing pluripotent stem cells (PSC) that can differentiate into a wide range of specialized cells. Basic fibroblast growth factor is essential for PSC survival, stemness and self-renewal. PI3K/AKT pathway regulates cell viability and apoptosis in many cell types. Although it has been demonstrated that PI3K/AKT activation by bFGF is relevant for PSC stemness maintenance its role on PSC survival remains elusive. In this study we explored the molecular mechanisms involved in the regulation of PSC survival by AKT. We found that inhibition of AKT with three non-structurally related inhibitors (GSK690693, AKT inhibitor VIII and AKT inhibitor IV) decreased cell viability and induced apoptosis. We observed a rapid increase in phosphatidylserine translocation and in the extent of DNA fragmentation after inhibitors addition. Moreover, abrogation of AKT activity led to Caspase-9, Caspase-3, and PARP cleavage. Importantly, we demonstrated by pharmacological inhibition and siRNA knockdown that GSK3β signaling is responsible, at least in part, of the apoptosis triggered by AKT inhibition. Moreover, GSK3β inhibition decreases basal apoptosis rate and promotes PSC proliferation. In conclusion, we demonstrated that AKT activation prevents apoptosis, partly through inhibition of GSK3β, and thus results relevant for PSC survival.