FIGURE 1.

Overview of the U‐CIE algorithm and application to single‐cell RNAseq data (scRNA‐seq). U‐CIE uses a two‐step process to encode high‐dimensional data as colors. The first step is to use a state‐of‐the‐art dimensionality reduction technique (UMAP) to reduce the data to three dimensions, since human color perception is inherently three‐dimensional. The second step is to fit the resulting point cloud into the three‐dimensional polygon that represents the displayable part of CIELAB color space. This is done as an optimization process, which shifts, rotates, and uniformly scales the point cloud to make it as large as possible while penalizing for points protruding outside the polygon