Zirkuläre RNAs sind eine Klasse endogener, tierischer RNAs. Obwohl sie hoch abundant sind, ist weder ihre Funktion noch ihre Expression im Nervensystem bekannt. Es wurde ein Katalog zirkulärer RNAs in neuralen Proben erstellt. Es konnten tausende zirkuläre RNAs von Mensch und Maus entdeckt und analysiert werden. Zirkuläre RNAs sind außerordentlich angereichert im Säugetiergehirn, ihre Sequenz ist gut konserviert und sie sind häufig gemeinsam in Mensch und Maus exprimiert. Zirkuläre RNAs waren generell höher exprimiert im Verlauf der neuronalen Differenzierung, sind stark angereichert an Synapsen und oft differentiell exprimiert. circSLC45A4 ist die Hauptisoform, die in humanem präfrontalem, embryonalen Cortex von diesem genomischen Lokus exprimiert wird und eine der am höchsten exprimierten zirkulären RNAs in diesem System. Induzierte Verminderung der Expression von circSLC45A4 ist ausreichend, um die spontane neuronale Differenzierung einer humanen Neuroblastomzelllinie zu induzieren. Dies kann durch die verstärkte Expression neuronaler Markergene belegt werden. Verminderung der Expression von circSLC45A4 im embryonalen Mauscortex verursacht eine signifikante Reduktion von basalen Progenitoren. Außerdem wurde eine signifikante Reduktion von Zellen in der kortikalen Platte nach Depletion von circSLC45A4 gemessen. Weiterhin konnten die Ergebnisse im Mauscortex dekonvoliert werden. Dies zeigte die Zunahme von Cajal-Retzius Zellen. Es wird eine Methode vorgestellt, die RNA-Sequenzierung von Einzelzellen mit räumlicher Auflösung zulässt, ohne vorherige Kenntnisse des Systems zu benötigen. 3D-seq vereint die Applikation eines physischen Gitters mit kombinatorischem Indizieren, so dass Einzelzellen individuell und räumlich markiert werden können. 3D-seq wurde an koronalen Schnitten von adultem Mausgehirn etabliert. Die Daten wurden zur Reproduktion des Gewebes in silico genutzt. 3D-seq ein leicht zu adaptierendes Protokoll, das an jedem Gewebe angewendet werden kann. / Circular RNAs (circRNAs) are an endogenous class of animal RNAs. Despite their abundance, their function and expression in the nervous system are unknown. Therefore, a circRNA catalogue comprising RNA-seq samples from different brain regions, primary neurons, synaptoneurosomes, as well as during neuronal differentiation was created. Using these and other available data, thousands of neuronal human and mouse circRNAs were discovered and analyzed. CircRNAs were extraordinarily enriched in the mammalian brain, well conserved in sequence, often expressed as circRNAs in both human and mouse, and sometimes even detected in Drosophila brains. CircRNAs were overall upregulated during neuronal differentiation, highly enriched in synapses, and often differentially expressed compared to their corresponding mRNA isoforms. CircRNA expression correlated negatively with expression of the RNA-editing enzyme ADAR1. Knockdown of ADAR1 induced elevated circRNA expression. Together, a circRNA brain expression atlas and evidence for important circRNA functions is provided. Starting from this catalogue a circRNA, circSLC45A4 was identified. It is the main RNA isoform produced from its genetic locus in the developing human frontal cortex and one of the highest expressed circRNAs in that system. Knockdown of this conserved circular RNA in a human neuroblastoma cell line was sufficient to induce spontaneous neuronal differentiation, measurable by increased expression of neuronal marker genes and neurite outgrowth. Depletion of circSlc45a4 in the developing mouse cortex caused a significant reduction of the basal progenitor pool and increased the expression of neurogenic regulators like Notch2, Foxp2, and Unc5b. Furthermore, a significant depletion of cells in the cortical plate after knockdown of circSlc45a4 was observed. In addition, deconvolution of the bulk RNA-seq data with the help of single cell RNA-seq data validates the depletion of basal progenitors after knockdown of circSlc45a4 in the mouse cortex and reveals an increase in Cajal-Retzius cells. Taken together, a detailed study of a conserved circular RNA that is necessary to maintain the pool of neural progenitors in vitro and in vivo is presented. The developing mouse cortex is a good illustration for a highly spatially organized tissue and why knowledge of spatial information for each cell can be of great importance. However, obtaining transcriptome-wide and spatially resolved information from single-cells has been proven to be a challenging task. Current state-of-the-art experimental methods are either limited by the number of genes that can be detected simultaneously within a single-cell or require preexisting spatial information. Here, 3D-seq, a new experimental technique that allows unbiased, high-throughput single-cell spatial transcriptomics is introduced. 3D-seq combines a physical grid with combinatorial indexing to label single cells of any tissue in a unique way and thereby preserving the approximate spatial localization of any given cell. 3D-seq was applied to coronal slices of adult mouse brain, more than 70 cell types were identified and the 3D-seq data was used to reproduce the tissue in silico with single-cell resolution. Furthermore, 3D-seq is easy to adapt, can be applied to any tissue and can be combined with other technologies.