The molecular logic of retinal cell types

The mammalian brain is assembled from thousands of neuronal cell types, which are organized in distinct circuits to perform behaviorally relevant computations. Transgenic mouse lines with different cell types selectively marked would facilitate our ability to dissect functional components of complex circuits. We performed a screen for cell type-specific GFP expression in the retina using BAC transgenic mice from the GENSAT project. Among others we identified mouse lines in which the inhibitory cell types of the night vision and directional selective circuit are selectively labeled. We quantified the stratification patterns to predict potential synaptic connectivity between marked cells of different lines and show that some of the lines enable targeted recordings and imaging of cell types from developing or mature retinal circuits. Then, we selected some of those mouse lines and genetically profiled the labeled cell types to reveal if adult cell types have a unique genetic fingerprint. Our data suggests that each cell type has a unique genetic signature that is likely to be a consequence of a unique transcriptional code in each cell type. A hierarchical clustering of the cell types showed that each cell type cluster to their corresponding biological cell class. We found genes exclusively expressed for cell classes and for biologically relevant combinations of cell types.