Plots

Figure 1	Figure 2	Figure 3	Figure 4
Aggregated metric results over all datasets.	Twelve spatial quality and temporal stability metrics evaluated for 9 DR methods run on ten datasets. Note that here we've added the results of convolutional methods to the analysis.	Detailed analysis of distances and movements produced by all DR techniques on the cartolastd dataset.	Examples of instability in TF-t-SNE (a,b) and TF-PCA (c,d,e).

Figure 5 (UMAP)	Figure 5 (t-SNE)	Table 3 supplement
UMAP projection of projections map showing the similarity of all evaluated techniques on all datasets (Sec. 5.2). Note that here we've added the results of convolutional methods to the analysis.	Alternative version of figure 5 where the projection method is t-SNE instead of UMAP.	Scatter plot showing correlation between traits and quality metrics (complement of Table 3).