Appendix A — Data collation

In this section, we write various cell metadata to a csv for public sharing of the data.
A.1 Key metadata

We read in key metadata from various files, store it in one data frame, and write it to a csv.
Show the code
# if there's no seurat object, read it in
if (!("seu.obj" %in% ls())){
  seu.obj <- readRDS(results_list$seurat_file_postCellTyping)
}

#cell ID, major cell type, minor cell type, x/y positions from Seurat file
meta = seu.obj@meta.data[,c("cell_ID", "x_slide_mm", "y_slide_mm", "major_cell_type", "minor_cell_type")]

# Novae domain
novae_9 <- readRDS(paste0(niche_dir, "/Data/novae_domains_9_namedAssignments.rds"))
meta$novaeDomain <- novae_9[rownames(meta)]

# Tumor lobe designation
meta$tumorLobe <- NA
meta[results_list$upper_lobe_cells, "tumorLobe"] <- "Interdigitated"
meta[results_list$lower_lobe_cells, "tumorLobe"] <- "Dense"

# Smoothed pathway scores in tumor cells
key_pathways = c("JAK-STAT" = "REACTOME_GENE_AND_PROTEIN_EXPRESSION_BY_JAK_STAT_SIGNALING_AFTER_INTERLEUKIN_12_STIMULATION",
                 "MHC-1" = "REACTOME_ANTIGEN_PRESENTATION_FOLDING_ASSEMBLY_AND_PEPTIDE_LOADING_OF_CLASS_I_MHC",
                 "NF-kappa B" = "REACTOME_REGULATION_OF_NF_KAPPA_B_SIGNALING",
                 "TNF" = "REACTOME_TNF_SIGNALING"
)

idx = which(meta$major_cell_type == "Tumor")
  
# Gather score for each pathway
for(i in 1:4){
  pathway = key_pathways[i]
  
  # Read in smoothed pathway scores
  smoothed_path_score <- readRDS(file = paste0(immNeigh_dir, "/Data/xy_smoothed_tumor_", pathway, ".rds"))
  meta[,paste0(gsub("-", "_", names(key_pathways[i])), "SmoothScore_Tumor")] <- NA
  meta[idx,paste0(gsub("-", "_", names(key_pathways[i])), "SmoothScore_Tumor")] <- smoothed_path_score[1,rownames(meta[idx,])]
}

# Distance to vasculature
dist2vessel <-readRDS(file = paste0(vessel_dir, "/Data/dist2vessel.rds"))
meta$VesselDistance <- dist2vessel[,1]

# Neighborhood glycolysis score
smoothed_glyc <- readRDS(file = paste0(metabolism_dir, "/Data/xy_smoothed_glycolysis.rds"))
meta$neighborhoodGlycolysis <- smoothed_glyc[1,rownames(meta)]

# Cell cycle prediction
cellcycle <- readRDS(paste0(prolif_dir, "/Data/cell_cycle_predictions.rds"))
meta$cellCyclePrediction <- cellcycle[rownames(meta)]

# Patch assignment for proliferation DE
meta$PatchAssignment <- NA
patches <- readRDS(paste0(prolif_dir, "/Data/patchDE.RDS"))
meta[names(patches), "PatchAssignment"] <- patches


write.csv(meta, file = paste0(objects_dir, "/derivedMetadata.csv"))