Fig. 5: IPD stratification.

a Two-dimensional PCA plot of the monogenic PD and IPD datasets. b Unsupervised clustering of monogenic PD and IPD datasets. c Two-dimensional PCA plot of the monogenic PD datasets and IPD datasets, considering every individual IPD sample separately. d Unsupervised clustering of monogenic PD and IPD datasets, considering every individual IPD sample separately. a–d Log2 FC of individual experiments considered (PDvsCTRL) of the genes included in the merged list of the top 100 significantly differentially expressed from each individual experiment. e Similarity network fusion analysis on integrated imaging and transcriptomic features (top 100 significantly differentially expressed genes) from 6 IPD and 3 CTRL cell lines. The computed similarity score among different cell lines is within the [0, 0.1] range in this set. The self-similarity has a value of 0.5. The colour palette shows the similarity among the cell lines: “white” for values close to 0, “yellow” for low values within the range [0, 0.1], “red” for values closer to 0.1, and “blue” for self-similarity (0.5). f Two-dimensional PCA plot of IPD and CTRL samples, considering the average cell expression of the top 100 significantly differentially expressed genes between every IPD line against all six CTRL cell lines used for single-cell RNA sequencing.