This CSV contains the similarity scores used to construct the heatmaps. Let me know if this format works or if you'd like it structured differently. Some notes about the metrics:
dist = the Euclidean distance between cluster centroids in an integrated human-marmoset-mouse space. Pairwise distances are calculated within classes (i.e. GABAergic mouse type vs. GABAergic human type)
similarity = a reverse scaling of the dist metric (since greater distance between cluster centroids implies "less similar" types). Used on the heatmaps as this was more intuitive
This CSV contains the similarity scores used to construct the heatmaps. Let me know if this format works or if you'd like it structured differently. Some notes about the metrics: