In the dendrogram construction described above we used several steps of exclusion genes either from all clusters or from the neuronal clusters in particular. This was done due to our observation of background levels of gene detection which seemed to be depend on very abundant cell types the dissected region (e.g., oligodendrocytes in hindbrain or enteric glia in the enteric nervous system). This is likely because of floating RNA coming from dead cells or doublets either with abundant cells or parts of broken cells. Still, due to the risk of misinterpreting the data we also constructed the dendrogram based on similar procedure but without any gene exclusion. The resulting dendrogram was not fundamentally changed from Figure 1C, but included a few key differences which we believe are technical artifacts. First, enteric neurons clustered together with the enteric glia probably due to fact that enteric glia were extremely abundant in the tissue. This created a big enough change that the other PNS neurons created a separate branch disconnected from the other neurons. Second, the olfactory bulb inhibitory neurons were placed next
