scpca-paper-figures

This repo contains the figures and tables included in the ScPCA manuscript.

Table of Contents

• Summary of figures and tables
• Generating figures and tables
  • Instructions to prepare data
  • Instructions for Data Lab members
• Additional repository contents
  • Sample info
  • Color palettes
  • Manuscript numbers
  • Nextflow logs
  • Analysis
• Renv
• Contributing

Summary of figures and tables

Below is a summary of all figures and tables in the paper.

Figure 1

• A. Bar chart summarizing the types of diagnoses found on the Portal.
• B. Bar chart summarizing the types of modalities found on the Portal.
• C. Example project card as shown on the Portal.

Figure 2

• A. Overview of main workflow.
• B-G. Simplified versions of plots in the main QC report.

Figure 3

• A. Illustration of individual sample download folder.
• B. Illustration of merged project download folder.
• C. Overview of merged workflow.
• D. Example UMAPs found in merged report.

Figure 4

• A. Overview of cell type annotation workflow.
• B. Comparison of cell type annotations obtained using SingleR and CellAssign.

Figure 5

• A. Dot plot summarizing marker gene expression in consensus cell types in Brain and CNS samples.
• B. Bar plot summarizing the percentage of cells annotated as each consensus cell type in High-grade and Low-grade glioma samples.
• C. Bar plot summarizing the percentage of cells annotated as each immune consensus cell type, with an emphasis on cells in the T-cell and myeloid lineages, in High-grade and Low-grade glioma samples.

Figure 6

• A. Scatterplots of the relationship between bulk and pseudobulk counts for relevant Brain and CNS projects.
• B. Bar plots of odds ratios from overrepresentation analysis of bulk expression data for relevant Brain and CNS projects.

Supplemental Figure 1

• A. Comparison of run time and peak memory usage between Alevin-fry and Cell Ranger.
• B. Total UMI/cell between Alevin-fry and Cell Ranger.
• C. Total genes detected/cell between Alevin-fry and Cell Ranger.
• D. Correlation of mean gene expression between Alevin-fry and Cell Ranger.

Supplemental Figure 2

• A. Overview of cell type annotation workflow.
• B-D. Simplified versions of plots found in the ADT section of the main QC report.
• E. Overview of multiplexed library workflow.

Supplemental Figure 3

• A. Overview of bulk RNA-seq workflow.
• B. Overview of spatial transcriptomics workflow.

Supplemental Figure 4

Comparison of delta median statistic obtained from running SingleR with different celldex references.

Supplemental Figure 5

• A. UMAP displaying cell type annotations from CellAssign for an example ScPCA library.
• B. Heatmap comparing submitter provided annotations to CellAssign and SingleR annotations for an example ScPCA library.

Supplemental Figure 6

• A. Dot plot summarizing marker gene expression in consensus cell types in Leukemia samples.
• B. Dot plot summarizing marker gene expression in consensus cell types in Sarcoma samples.
• C. Dot plot summarizing marker gene expression in consensus cell types in Other solid tumor samples.

Supplemental Figure 7

• A. Bar plot summarizing the percentage of cells annotated as each consensus cell type in Leukemia samples.
• B. Bar plot summarizing the percentage of cells annotated as each consensus cell type in Sarcoma samples.
• C. Bar plot summarizing the percentage of cells annotated as each consensus cell type in Other solid tumor samples.

Supplemental Figure 8

• A. Scatterplots of the relationship between bulk and pseudobulk counts for projects not shown in Figure 6A.
• B. Bar plots of odds ratios from overrepresentation analysis of bulk expression data for projects not shown in Figure 6B.

Table S1

Summary of libraries and types of libraries found on the Portal.

Table S2

List of references used for each project on the Portal with CellAssign, including the list of organs used to create the reference.

Generating figures and tables

The figures/ and tables/ folders contain the most up-to-date version of each of the figures and tables, respectively. The scripts/ folder contains all scripts used to create the figures and tables. See the README for the scripts folder for more information on figure and table scripts.

The generate-figures-tables.sh script can be used to prepare all figures and tables. The script assumes that a folder called s3_files has been populated with relevant data files. Therefore, before running this script you will first need to prepare these input data files, as described in the sections below.

Then, run the figure generation script as:

sh bash generate-figures-tables.sh

Instructions to prepare data

If you are not a member of the Data Lab, please follow the instructions provided in reproduce-figures/README.md to obtain and prepare additional data needed to regenerate figures and tables.

Instructions for Data Lab members

To prepare data for figure and table generation, you will need to run the figure setup scripts:

sh Rscript scripts/figure_setup/sync-metadata.R Rscript scripts/figure_setup/sync-data-files.R Rscript scripts/figure_setup/sync-reference-files.R Rscript scripts/figure_setup/sync-consensus-celltype-results.R

If you have setup 1Password to handle your AWS credentials, you will need to prefix scripts beginning with sync- with op run --, specifically:

sh op run -- Rscript scripts/figure_setup/sync-metadata.R op run -- Rscript scripts/figure_setup/sync-data-files.R op run -- Rscript scripts/figure_setup/sync-reference-files.R

Additional repository contents

Sample info

The sample-info/ folder contains metadata files used to create figures and tables.

• brain-classifications-no-multiplexed.tsv: This tsv file classifies brain-related diagnoses in the ScPCA Portal into "High-grade glioma" and "Low-grade glioma" for plotting.
• celltype-reference-metadata.tsv: This tsv file contains information about references used for CellAssign and SingleR cell type annotation on the ScPCA Portal.
• diagnosis-groupings.tsv: This tsv file contains one row per submitted_diagnosis associated with samples on the ScPCA Portal. For each submitted_diagnosis, a diagnosis_group is assigned.
• disease-timing.tsv: This tsv file contains one row per submitted_disease_timing associated with samples on the ScPCA Portal. For each submitted_disease_timing, a standardized_disease_timing is assigned.
• project-whitelist.txt: This file contains a list of all projects that are currently active on the ScPCA Portal.
• sample-whitelist.txt: This file contains a list of all samples that are currently active on the ScPCA Portal.
• scpca-project-celltype-metadata.tsv: This tsv file provides the specific CellAssign cell typing reference used for each ScPCA project.

Color palettes

The palettes/ folder contains any palettes used in generating the figures.

• diagnosis-group-palette.tsv: This is the palette used to color the diagnosis_group for each sample.
• disease-timing-palette.tsv: This is the palette used to color the disease_timing for each sample.
• immune-palette.tsv: This is the palette used to color certain immune cell types from the overall consensus cell types.
• suspension-palette.tsv: This is the palette used to color libraries by Single-cell or Single-nuclei.
• method-palette.tsv: This is the palette used to color by quantification method used, either Alevin-fry or Cell Ranger.
• validation-group-palette.tsv: This is the palette used to color broad cell type annotations used to assess consensus cell types.

Manuscript numbers

The manuscript-numbers folder contains tables with total sample counts referenced when writing the manuscript. These tables are not included in the final manuscript and were created as follows:

• bulk-analysis-counts.tsv was created by ../scripts/Fig6-FigS8_bulk-analysis.R
• diagnosis-group-counts.tsv and disease-timing-counts.tsv were created by ../scripts/Fig1A_sample-disease-barchart.R

Nextflow logs

The nextflow_logs folder contains text files with Nextflow log information from running the scpca-nf pipeline. These files are specifically used by scripts/FigS1A_memory-time-comparison.R to create Figure 1A.

Analysis

This analysis folder contains analyses comparing bulk and pseudobulk RNA-Seq data. Please refer to analysis/README.md for additional details.

Renv

Package dependencies for scripts used in this repo are managed using renv. For renv to work as intended, you'll need to work within the scpca-paper-figures.Rproj project in RStudio. You may need to run renv::restore() upon opening the project to ensure the renv.lock file is synced with the project library.

Contributing

When developing new scripts, you may need to install or use new R packages. Each time you install or use new packages, you will want to run renv::snapshot() to update the renv.lock file with any added package and dependencies necessary to run the analyses and scripts in this repo.

In addition, this repository uses the parsable-r pre-commit hook to ensure R scripts are parsable. To use this hook, first ensure that that the pre-commit package is installed on your system; you can install it with your favorite method (pip install pre-commit or conda install pre-commit). Then, run pre-commit install in the scpca-paper-figures folder to enable pre-commit hooks in this repository. This will install the hooks in the .git/hooks folder, and they will be run automatically when you commit changes. If the hook fails, the commit will be aborted, and you will need to fix the errors and re-commit.