Extended Data Fig. 4: Integration of HCS datasets (related to Fig. 3).
(a) Overview of the number of reference drugs and samples in each category and dataset used for the integration analysis. (b) Summary of sample sizes (top) and feature dimensions (bottom) for each reference dataset. (c) UMAP visualization of integrated embeddings from different integration methods, with color annotations indicating reference compound categories (top), datasets (bottom left), and cell lines (bottom right). (d) Heatmap visualization (left) and value distribution comparison (right) of original profiles for active and inactive samples in datasets #6 and #7. P-value: two-sided Mann-Whitney U test. (e) UMAP visualization of the CLIPn embeddings, highlighting compounds from the mTOR and PI3K inhibitor categories (colored dots).
