Extended Data Fig. 14: Schematic summaries of MOp-ul outputs by area (PHAL) and for different cell types revealed with cre-dependent AAV tracing in different cre mouse lines on a whole brain flatmap of the rodent brain5 (also see Swanson, Brainmap 4.0 in http://larrywswanson.com/?page_id=1415).

Related to Fig. 2. These data shows that each of the MOp cell types (L2/3 IT, L4 IT, L5 IT, L5 ET/PT, L6 CT) display a discrete subset of MOp projections. Please note that these results also showed MOp-ul axons targeting several previously unreported areas, e.g., the capsular central amygdalar nucleus (CEAc), bed nucleus of the anterior commissure (BAC), globus pallidus external segment (GPe), contralateral thalamic nuclei (PCN), and cerebellar interposed nucleus (IP; Suppl. Information). Further analyses of the connectivity matrix (Source Data Fig. 2, formatted matrix tab) and images (Extended Data Fig. 12) reveal the predominant PN types constituting new and established MOp-ul output channels (also see Fig. 2d). For example, projections to SSp-ul originate from both L2/3 and L5 IT neurons labeled in the Cux2–L2/3, Rbp4–L5 and Tlx3-L5 populations. Projections to CEAc arise from L5 IT cells (Rbp4-L5 and Tlx3-L5, also see Extended Data Fig. 12), and projections to GPe are primarily from ET cells (Rbp4-L5, Fezf2-L5, Foxp2-L6, Sim1-L5). Anterograde experiments also confirmed (in Cux2-L2/3) the population of L2 neurons projecting contralaterally to TEa, ECT, and PERI identified by retrograde tracing (see Extended Data Fig. 7), and a Tle4-L6 CT projection to contralateral thalamic nuclei. Moreover, the sparse cerebellar projection to the IP nucleus we observed with PHAL, AAV-GFP, and AAV1-Cre anterograde monosynaptic tracing is also labeled in Rbp4-L5, but not other ET lines (Source Data Fig. 2, Extended Data Fig. 12).