ABSTRACT Neural circuit formation requires a series of highly diverse and specific cell-cell recognition steps, many mediated by cell adhesion molecules (CAMs). Indeed, mutations that disrupt CAMs or their regulation are associated with circuit level neurodevelopmental disorders from dyslexia to schizophrenia. Our model is the mouse retina, an extension of the central nervous system where ~100 types of neurons organize into dedicated circuits that encode the features of the visual world. We focus here on the gamma-protocadherins (?-Pcdhs), 22 CAMs expressed from a single gene cluster that generate many thousands of distinct homophilic recognition complexes. The ?-Pcdhs are critical regulators of neuronal self-avoidance in starburst amacrine cells (SACs), and cell survival and in many other types of neurons in the retina. The mechanisms through which the ?-Pcdhs serve these functions are unknown, as is the importance of ?-Pcdh isoform diversity. We used a CRISPR/Cas9 approach to generate an unbiased allelic series of mouse mutants with between 1 and 21 intact ?-Pcdh isoforms. From these, we learned that one isoform, ?C4, is essential for neuronal survival, suggesting that this isoform functions differently from the other 21. We propose to define the mechanisms of self-avoidance and neuronal survival, and to use our allelic series to determine the level of isoform diversity required for normal neural circuit formation. Our central hypotheses are that: 1) a high level of ?-Pcdh isoform diversity enables neurons to distinguish between ?self? and ?non-self? to mediate self-avoidance while permitting interaction with neighboring neurons through mechanisms common to all isoforms; and 2) neuronal survival, in contrast, requires interactions specific to the ?C4 isoform. In Specific Aim 1, we will use a strategic subset of our reduced-diversity mutants to determine the extent of isoform diversity required for self/non-self discrimination in SACs, neurons essential for the motion detection circuit in the retina. We will analyze this circuit at two levels: A) morphology of contacts between SACs, and B) the electrophysiological function of direction-selective retinal ganglion cells, the downstream neurons in the circuit. In Specific Aim 2, we will define the molecular mechanisms of self-avoidance using in vivo gene delivery to manipulate candidate pathways and map essential domains. In Specific Aim 3 we will uncover the mechanisms through which ?C4 promotes neuronal survival. We will use retinal electroporation to map critical protein domains, complemented by a discovery-based proteomics approach to find isoform-specific protein interactions for ?C4. These studies will allow us to better understand how the ?-Pcdhs contribute to cell-cell recognition and neural circuit formation in the retina and provide insight into processes disrupted by neurodevelopmental disorders.