Table of contents generated with markdown-toc
Large data files used by the code in this repository are stored on GEO.
| Description | Filename(s) |
|---|---|
| Raw count data before filtering | <library_name>.barcodes.tsv, <library_name>.counts.mtx, <libraray_name>.genes.tsv (GSE161771 |
| Anndata object combining all libraries after mitochondrial fraction and total count filtering | mito_total_counts_filt_raw_27563x40930_200517_10h29.h5ad |
| Cell annotation file (i.e. adata.obs) | obs_info_27563x32_201025_14h44.tsv (or .npz here*) |
Angela E. Zou (AZ)
Marius Messemaker (MM)
Nicolas A. Gort-Freitas (NAGF)
Rapolas Zilionis (RZ)
The indrop.py pipeline was used for obtain cells x genes matrices. Yaml files detailing the parameters used can be found here.
Notebooks starting with "part*" focus on the analysis steps up to filtered annotated data. Data filtering involved repeating cell visualization several times. For example, drawing and clustering a kNN graph of T cells only revealed two distinct double populations within T cells, the removal of which required repeating the visualization of all CD45+ cells.
| Methods | Figure panel(s) | Comment | Relevant notebooks | Contributions |
|---|---|---|---|---|
| Filtering on total counts and mitochondrial fraction | NA | Filter data and combine individual mtx files into one AnnData object | part1_read_and_filter_data.ipynb | RZ |
| Cell annotation using a Bayesian classifier | NA | Classify single cell transcriptomes using a Bayesian classifier as described previously [2,3,4] | part3_classify_cell_by_published_profiles.ipynb | RZ |
| Initial visualization of the data using SPRING [5] | Interactive explorer online | 2D visualization using a force directed layout of a kNN graph | part3c_spring_plot_1000_umi.ipynb | RZ |
| Doublet detection | NA | Obtaining doublet scores using Scrublet [6]. Modified version using precomputed PCA: for consistency, the same PCA-transformed data as for drawing the kNN graph in SPRING visualization is used | part4_detect_doublets_in_each_emulsion_precomputed_PCA.ipynb | RZ |
| Spectral clustering of the SPRING plot | NA | Divide the kNN graph into a predefined number of clusters | part5_divide_graph_with_doublets_into_clusters.ipynb | |
| Identify cells to exclude | NA | Identify and flag for removal clusters enriched in doublets, Krt8 (non-immune), and hemoglobin genes (erythroid) | part6_decide_which_cells_to_exclude.ipynb | RZ |
| Repeat data visualization and spectral clustering after cleanup | Interactive explorer online | Visualization exclude cells flagged in the previous notebook. Eigenvectors for PCA calculated on Csf1Ri condition cells only (as a method for batch correction). | part7_sping_plot_main_iter1.ipynb, part8_divide_graph_into_clusters.ipynb | RZ |
| Identify more cells to exclude | NA | Repeating the visualization and clustering after initial clean up revealed more distinct doublet clusters | part8_divide_graph_into_clusters.ipynb | RZ |
| Data visualization and clustering after second round of cleanup | Interactive explorer online | Second iteration of plotting cleaned up data | part10_sping_plot_main_iter2.ipynb, part11_divide_graph_into_clusters.ipynb | RZ |
| Cluster annotation using Bayesian classifier results | Interactive explorer online, colortrack "*population" | Clusters are labeled after the dominant classification result obtained for individual cells. Ambiguous cases are reviewed manually. | part12_define_populations_based_on_classifier_results.ipynb | RZ |
| Visualize, cluster, and annotated T cells separately | Interactive explorer online | Resolving T cell sub-populations required subclustering T cells. Annotation is based on interactively exploring the resulting SPRING plot, and comparing cluster-enriched gene expression to known marker genes | part13_sping_plot_of_T_only.ipynb, part14_divide_T_cell_graph_into_clusters.ipynb, part15_define_T_subsets.ipynb | RZ |
| Final data visualization, clustering, and annotation after removing doublet clusters within T cells | Fig. 2B,C; 4E; 5D and more, Interactive explorer online | Visualization and annotation of T cells only in the previous step revealed two distinct doublet clusters. Visualization of all CD45+ cells was repeated with the further cleaned up data | part16_sping_plot_main_iter3.ipynb, part17_divide_graph_into_clusters.ipynb, part18_define_populations_based_on_classifier_results.ipynb | RZ |
| Final annotation of T cells | Interactive explorer online | Repeat T cell annotation after removing T cell doublets | part19_sping_plot_of_T_only.ipynb, part20_divide_T_cell_graph_into_clusters.ipynb, part21_define_T_subsets.ipynb | RZ |
| Clean up annotations | Fig. 2B | Tidy up cell annotations in adata.obs, merge minor populations to main types (e.g. DC1, DC2, DC3, pDC collectively are DCs) | part22_cleanup_labels.ipynb | RZ |
| Methods | Figure panel(s) | Comment | Relevant notebooks | Contributions |
|---|---|---|---|---|
| Example notebook for loading annotated data and xy coordinates | E.g. Fig. 2B | This notebooks uses a few examples to guide anyone interested through how the annotated data is organized | example_load_data_plot_something.ipynb | RZ |
| Plot a subset of cells from the main SPRING plot, color by gene expression | Multiple figures, the motif of Fig. 4E | Load xy coordinates, select a subset of cells, color by gene expression or population annotation | Colored_SPRING_plots.ipynb | RZ |
| Challenge annotations by plotting a heatmap of previously identified marker genes | Fig. 2D | Recreate marker gene heatmap from previous study [2] (same gene order) but using the newly defined cell populations | Annotation_challenging_marker_gene_heatmaps.ipynb | RZ |
| State %CD45 abundance, Arrow gene-expression change, and Differential Expression Analysis volcano Figures | 2E, 2F, S2A, S2C, S2D, S2F, S2G, 4G, S4I, S4J, 5F, S5D-H | Abundance_and_expression_change_analysis.ipynb | MM | |
| Make dot plots of relative gene expression and % cells expressing genes | Figs 2G, 3A, S3A | for-github_dotplot.ipynb | AZ | |
| Perform GSEA on GO:BP terms, make scatterplot of enriched immune activation-related terms | Fig. 2H | for-github_fgsea-scatterplot.ipynb | AZ | |
| Heatmap of scores for selected GO:BP terms in MoMac cells | Fig. 2I | for-github_scored-pathway-HM.ipynb | AZ | |
| Make circos plots for differentially expressed and immune activating/inhibitory interactions | Figs. 3B, 5B | for-github_cell-cell-comms_filter+circos.ipynb | AZ | |
| Make heatmaps depicting selected ligand-receptor interactions | Figs. 3C, S3C, 5C, S5B | for-github_cell-cell-comms_make-HMs.ipynb | AZ | |
| Fold change with respect to the median across states compared (relative expression); Pearson's r correlation; Linear regression | 3D, S3D, S5 | In this notebook, we compare the relative expression of highlighted ligands & receptors in DCs, NKs, T cells, and Monocyte/Macrophages in non-small cell lung cancer patients (Zilionis et al., 2019) and in vehicle-treated mice to support a cross-species analogy in the behavior of immunity. | heatmaps_scatter_human_mouse.ipynb | NAGF |
[1] To be updated after publication
[2] Zilionis R, Engblom C, Pfirschke C, et al. Single-Cell Transcriptomics of Human and Mouse Lung Cancers Reveals Conserved Myeloid Populations across Individuals and Species. Immunity. 2019;50(5):1317-1334.e10. doi:10.1016/j.immuni.2019.03.009
[3] Zemmour, D., Zilionis, R., Kiner, E., Klein, A.M., Mathis, D., and Benoist, C. (2018). Single-cell gene expression reveals a landscape of regulatory T cell phenotypes shaped by the TCR. Nat. Immunol. 19, 291–301
[4] Jaitin, D.A., Kenigsberg, E., Keren-Shaul, H., Elefant, N., Paul, F., Zaretsky, I., Mildner, A., Cohen, N., Jung, S., Tanay, A., and Amit, I. (2014). Massively par- allel single-cell RNA-seq for marker-free decomposition of tissues into cell types. Science 343, 776–779.
[5] Weinreb C, Wolock S, Klein AM. SPRING: a kinetic interface for visualizing high dimensional single-cell expression data. Bioinformatics. 2018;34(7):1246-1248. doi:10.1093/bioinformatics/btx792
[6] Wolock SL, Lopez R, Klein AM. Scrublet: Computational Identification of Cell Doublets in Single-Cell Transcriptomic Data. Cell Syst. 2019;8(4):281-291.e9. doi:10.1016/j.cels.2018.11.005