Additional file 1 of Coordinated single-nucleus responses for quantitative disease resistance involve a calcium-associated switch in transcriptional noise

Additional file 1: Fig. S1. An overview of nuclei and reads per nucleus before and after filtering. Fig. S2. Clustering and cell-type assignment of the nuclei at three clustering resolutions. Fig. S3. The hierarchical relationship between cluster levels and the cluster sizes for each sample. Fig. S4. Identified cluster specific markers for each of the three clustering levels. Fig. S5. t-SNE plots showing scores of CNMF factors computed on all nuclei, with each sample plotted separately. Fig. S6. Complete list of abundance and enrichment p-values of gene ontologies in single nucleus clusters. Fig. S7. Genes involved in camalexin and indole-glucosinolate biosynthesis and their expression values. Fig. S8. Similar trajectories were obtained by estimating the pseudotime trajectories for all cell types together and for the epidermal and mesophyll subsets separately. Fig. S9. Global correlation for genes significantly regulated along pseudotime trajectories supported a good agreement in the transcriptome reprogramming along pseudotime ~ 0.21 to 1 and Tagseq expression between 0 and 24 h post inoculation. Fig. S10. Relationship between different measurements of within nucleus gene expression variability and sequencing statistics. Fig. S11. Distributions of entropy and number of reads per cluster. Fig. S12. t-SNE plots showing entropy and variance metrics per nucleus for each sample. Fig. S13. Relation between genes differentially expressed and entropy increase upon inoculation. Fig. S14. Size distribution of Ca clusters correlates with single nucleus RNA-seq cluster sizes along pseudotime trajectory 2, supporting a correspondence between pseudotime and distance to the fungus. Fig. S15. Isolation of Arabidopsis thaliana nuclei by flow cytometry.