shahcompbio / spectrum_tme

preprocessing and figures for SPECTRUM TME paper

SPECTRUM TME scRNA figure repository

The repository includes all analysis necessary to arrive at the figures for the SPECTRUM TME paper.

Scripts are separated into preprocessing scripts and plotting scripts.

STATUS

• OPEN
• DONE

Data dependencies

Processed expression data

(File paths on juno)

The output from Nick Ceglia's scRNA-seq pipeline serves as an input to the analysis.

• SPECTRUM v6 pre-Rx cohort expression object: /work/shah/isabl_data_lake/analyses/68/75/6875/RNASCP/outs/
• SPECTRUM v6 pre-Rx cohort embedding table: /work/shah/isabl_data_lake/analyses/68/75/6875/RNASCP/outs/cells.tsv
• v6 cell type objects (pipeline output, for annotated objects, see further below): /work/shah/isabl_data_lake/analyses/68/75/6875/RNASCP/outs/{cell_type}.rds
• v6 cell type cluster markers: /work/shah/isabl_data_lake/analyses/68/76/6875/RNASCP/outs/{cell_type}_markers.tsv
• SPECTRUM v7 pre-Rx cohort expression object: /work/shah/isabl_data_lake/analyses/84/78/8478/RNASCP/outs/
• SPECTRUM v7 pre-Rx cohort embedding table: /work/shah/isabl_data_lake/analyses/84/78/8478/RNASCP/outs/cells.tsv
• v7 cell type objects (pipeline output, for annotated objects, see further below): /work/shah/isabl_data_lake/analyses/84/78/8478/RNASCP/outs/{cell_type}.rds
• v7 cell type cluster markers: /work/shah/isabl_data_lake/analyses/84/78/8478/RNASCP/outs/{cell_type}_markers.tsv

Meta data

(File paths in this repo)

• scRNA-seq sample sheet:
  • google spreadsheet owned by Ignacio
  • downloaded copy, Dec 3rd, 2020: [_data/small/MSK\ SPECTRUM\ -\ Single\ cell\ RNA-seq_v6.xlsx](_data/small/MSK\ SPECTRUM\ -\ Single\ cell\ RNA-seq_v6.xlsx)
• color code: _data/small/signatures/hgsc_v7_colors.yaml
• mutational signature consensus labels: _data/small/mutational_signatures_summary_v3.tsv

Preprocessing scripts

Cohort

• compute patient specificity from SNN graph

  • _preprocessing/010_SPECTRUM_freeze_v7_patient_mixing_multicore.R

• run consensusOV for TCGA subtype analysis

  • helper script to start runs: _preprocessing/020_SPECTRUM_freeze_v7_TCGA_signatures.R
  • parameterized markdown to compute consensusOV for a given patient.rds input file _preprocessing/021_SPECTRUM_freeze_v7_TCGA_signatures_per_patient.Rmd
  • summarized report of TCGA analysis _preprocessing/022_SPECTRUM_freeze_v7_TCGA_summary.html

• output files and objects:

  • patient specificity per cell
  • TCGA consensusOV: /work/shah/uhlitzf/data/SPECTRUM/freeze/v7/consensusOV/SPECTRUM_freeze_v7_consensusOV.tsv

Cell type subsets

• common tasks

  • annotate sub type clusters and filter out doublet clusters
  • run progeny for pathway scoring
  • score gene signature modules

• individual scripts for common tasks:

  • B.super _preprocessing/031_SPECTRUM_freeze_v7_prepro_B.super.Rmd
  • Fibroblast.super _preprocessing/032_SPECTRUM_freeze_v7_prepro_Fibroblast.super.Rmd
  • T.super _preprocessing/033_SPECTRUM_freeze_v7_prepro_T.super.Rmd
  • Myeloid.super _preprocessing/034_SPECTRUM_freeze_v7_prepro_Myeloid.super.Rmd
  • Ovarian.cancer.cell.super _preprocessing/035_SPECTRUM_freeze_v7_prepro_Ovarian.cancer.cell.super.Rmd
  • Endothelial.super _preprocessing/036_SPECTRUM_freeze_v7_prepro_Endothelial.super.Rmd

• cell type specific tasks

  • Cancer cells: filter out low quality cells <1% mito reads and recluster
  • T cells: subcluster CD8 T cells and run diffusion map on CD8 subset
  • Myeloid cells: run diffusion map on macrophage subset

Output files and objects

• major cell types
  • /work/shah/uhlitzf/data/SPECTRUM/freeze/v7/Ovarian.cancer.super_processed_filtered_annotated.rds
  • /work/shah/uhlitzf/data/SPECTRUM/freeze/v7/T.super_processed_filtered_annotated.rds
  • /work/shah/uhlitzf/data/SPECTRUM/freeze/v7/Myeloid.super_processed_filtered_annotated.rds
  • /work/shah/uhlitzf/data/SPECTRUM/freeze/v7/Fibroblast.super_processed_filtered_annotated.rds
  • B.super.annotated.rds
  • Endothelial.cell.super.annotated.rds
• subset objects
  • /work/shah/uhlitzf/data/SPECTRUM/freeze/v7/CD8.T_processed_filtered.rds
  • /work/shah/uhlitzf/data/SPECTRUM/freeze/v7/work/shah/uhlitzf/data/SPECTRUM/freeze/v7/DCs_processed.rds
  • /work/shah/uhlitzf/data/SPECTRUM/freeze/v7/work/shah/uhlitzf/data/SPECTRUM/freeze/v7/Macrophages_processed.rds
• supplementary marker tables
  • /work/shah/uhlitzf/data/SPECTRUM/freeze/v7/supplementary_tables/Ovarian.cancer.super_marker_table_annotated.tsv
  • /work/shah/uhlitzf/data/SPECTRUM/freeze/v7/supplementary_tables/T.super_marker_table_annotated_full.tsv
  • /work/shah/uhlitzf/data/SPECTRUM/freeze/v7/supplementary_tables/Myeloid.super_marker_table_annotated.tsv
  • /work/shah/uhlitzf/data/SPECTRUM/freeze/v7/supplementary_tables/Fibroblast.super_marker_table_annotated.tsv

Plotting scripts

Cohort (Fig 2)

110_cohort_plotting_umap_v7.html

• UMAP embeddings
• Patient specificity
• TCGA subtypes

115_cohort_plotting_composition_v7.html

• Composition barplots scRNA
• Composition barplots mpIF

Cancer cell (Fig 3)

120_cancer_cell_plotting_umap_v7.html

• UMAP embeddings
• Marker heatmap
• Composition barplots
• Pathway heatmap and boxplots

T cell (Fig 5)

130_T_cell_plotting_umap_comp_heatmap_v7.html

• UMAP embeddings
• Marker heatmap
• Composition barplots

135_T_cell_CD8_diffusion_v7.html

• DC embeddings
• Pseudotime trajectories
• Module boxplots
• Correlation scatter

Myeloid cell (Fig 6)

140_Myeloid_cell_plotting_umap_comp_heatmap_v7.html

• UMAP embeddings
• Marker heatmap
• Composition barplots

145_Macrophage_diffusion_v7.html

• UMAP marker gene embeddings
• Module score boxplots