differential_expression_blish_S100A8.Rmd

---
title: "Differential expression analysis for selected cells in the COVID S100A8 regression"
author: "Jan Schleicher"
output:
  github_document:
    toc: true
    toc_depth: 2
---

```{r, include=FALSE}
knitr::opts_chunk$set(
  collapse = FALSE,
  message=FALSE
)
```

```{r setup, message=FALSE, warning=FALSE}
library(Seurat)
library(dplyr)
library(reshape2)
library(ggplot2)
library(introdataviz)
library(future)
plan("multisession", workers = 4)
library(data.table)
```

```{r, include=FALSE}
setwd(dirname(rstudioapi::getActiveDocumentContext()$path))
```

## Loading the data

For this analysis, we load the Seurat object containing the COVID-19 data from [Wilk et al. 2021](https://doi.org/10.1084/jem.20210582) and the extracted filter response from the S100A8 regression.

```{r loading}
blish <- readRDS("../data/blish_seurat.rds")
filter_response <- read.csv("../output/filter_responses/blish_S100A8_filter_response.csv",
                            row.names = 1)
```

## Identifying selected cells

We add selected cells, i.e., cells with a filter response that is larger than $0.3 \times (\text{maximum filter response})$.

```{r selected_cells}
threshold = .3
filter_response$selected_filter_0 <- filter_response$response_filter_0 > threshold *
  max(filter_response$response_filter_0)
cat("Number of selected cells for filter 0:", sum(filter_response$selected_filter_0))
blish <- AddMetaData(blish, filter_response)

DimPlot(blish, cells.highlight = WhichCells(blish, expression = selected_filter_0), reduction = "umap")

coarse_cell_type_selected_0 <- paste(blish@meta.data$cell.type.coarse, blish@meta.data$selected_filter_0)
names(coarse_cell_type_selected_0) <- row.names(blish@meta.data)
blish <- AddMetaData(blish, coarse_cell_type_selected_0, col.name = "coarse_cell_type_selected_0")
```

## Analysing differential expression between selected and unselected cells

For a detailed characterization of selected cells, we determined differentially expressed genes between selected and unselected cells of the same type. We did this analysis for the most highly enriched selected cell types.

```{r functions}
extract_data_for_violin <- function(adata, genes, cell_type_field, cell_types) {
  selected_cells <- row.names(adata@meta.data[apply(adata[[cell_type_field]], 1, function(x) x %in% cell_types),])
  data <- FetchData(adata, vars = c(genes, cell_type_field), cells = selected_cells, slot = "data")
  return(reshape2::melt(data, id.vars = cell_type_field, variable.name = "gene", value.name = "expression"))
}

make_violin_plot <- function(adata, genes, cell_type_field, cell_types, title, base_size=12, palette="Set2") {
  p <- ggplot(extract_data_for_violin(adata, genes, cell_type_field, cell_types),
              aes(x = gene, y = expression, fill = get(cell_type_field))) +
    geom_split_violin(scale = "width",
                      trim = T) +
    scale_fill_brewer(name = "selected", labels = c("false", "true"), palette = palette) +
    theme_classic(base_size = base_size) +
    theme(plot.title = element_text(hjust = 0.5),
          axis.text.x = element_text(angle = 45, hjust = 1)) +
    labs(title=title)
  return(p)
}
```

```{r de_analysis}
filt <- "coarse_cell_type_selected_0"
cell_types <- c("Developing neutrophil", "PB")

cat("Identifying markers for", filt, "\n")
Idents(blish) <- filt
for (ct in cell_types) {
  print(ct)
  cat("...identifying markers for selected", ct, "cells \n")
  markers <- FindMarkers(blish, ident.1 = paste(ct, "TRUE", sep = " "),
                         ident.2 = paste(ct, "FALSE", sep = " "))
  markers <- markers[markers["p_val_adj"] < 0.05,]
  markers <- cbind(gene = row.names(markers), markers)
  fwrite(markers, file = paste("../output/tables/blish_S100A8", filt,
                               gsub(",*\\s|/", "_", ct), "deg.csv",
                               sep = "_"))
  p <- make_violin_plot(blish,
                        row.names(head(markers[-grep('^MT-', rownames(markers)),], n = 6)),
                        filt, c(paste(ct, "TRUE", sep = " "), paste(ct, "FALSE", sep = " ")),
                        ct, base_size = 14, palette = "Paired")
  print(p)
  svg(paste("../output/figures/blish_S100A8/blish_S100A8", filt,
               gsub(",*\\s|/", "_", ct), "deg.svg",  sep = "_"), width = 4, height = 4)
  print(p)
  dev.off()
}
```