Project Summary The goal of this project is to create a unified framework for understanding the relationship between neuronal gene expression and connectivity in mouse visual cortex, by using morphology as a key linking modality. There now exist publicly available large-scale data sets that measure both these modalities in mouse visual cortex. One dataset is a large set of Patch-seq experiments from single cells, which provide measurements of gene expression, electrophysiological properties and morphology for individual cells. A second dataset is from large scale Connectomics using electron microscopy, which provides neuronal morphology; fine-scale and detailed cellular and ultrastructural properties; and measurements of the connectivity between individual neurons. Our first aim is to analyze these two datasets in an explicitly integrative fashion, in order to build better classifiers of neuronal types, along with tools to translate type predictions between each data modality. Our second aim builds on the first, by using those tools to characterize cell-type specific connectivity of mouse visual cortex. This will allow us to describe how that connectivity relates to the likely molecular composition of individual cells and provide insight into which molecular distinctions drive differences in cell type specific connectivity patterns. Our third aim is to redistribute the results of our analysis back into publicly available data repositories and create tools that allow other researchers to query the gene expression, connectivity, electrophysiological and morphological descriptors of neurons in the datasets, as well as apply those same tools to their own data. Like a Rosetta stone for cell types, this will enable researchers using disparate methods to integrate their data with other modalities and foster a rich environment for understanding the role of cell types in brain function and disease.