Researchers at CeMM, the Research Center for Molecular Medicine of the Austrian Academy of Sciences, have developed knowledge-primed neural networks (KPNNs), a new method that combines the power of deep learning with the interpretability of biological network models. KPNNs learn multiple layers of protein signaling and gene regulation from single-cell RNA-seq data, thereby providing a much-needed boost in our ability to convert massive single-cell atlas data into biological insights. These findings have now been published in the renowned scientific journal Genome Biology.