propeller.Rmd

---
title: "Celltype composition with Propeller"
author: "Harvard Chan Bioinformatics Core"
date: "`r Sys.Date()`"
output:
   html_document:
      code_folding: hide
      df_print: paged
      highlights: pygments
      number_sections: true
      self_contained: true
      theme: default
      toc: true
      toc_float:
         collapsed: true
         smooth_scroll: true
params:
   project_file: ../information.R
---

```{r, message=FALSE, warning=FALSE}
# This set up the working directory to this file so all files can be found
library(rstudioapi)
library(tidyverse)
setwd(fs::path_dir(getSourceEditorContext()$path))
# NOTE: This code will check version, this is our recommendation, it may work
# .      other versions
stopifnot(R.version$major >= 4) # requires R4
if (compareVersion(R.version$minor, "3.1") < 0) warning("We recommend >= R4.3.1")
stopifnot(compareVersion(as.character(BiocManager::version()), "3.18") >= 0)
stopifnot(compareVersion(as.character(packageVersion("Seurat")), "5.0.0") >= 0)
```

This code is in this ![](https://img.shields.io/badge/status-draft-grey) revision.


```{r setup, cache=FALSE, message=FALSE, warning=FALSE, echo=FALSE}
# knitr::opts_chunk$set(echo = TRUE)
# Load libraries
library(knitr)
library(rmarkdown)

library(Seurat)
library(SummarizedExperiment)
library(SingleCellExperiment)
library(DEGreport)
library(pheatmap)

library(tidyverse)
library(data.table)
library(DT)
library(patchwork)

library(ggprism)
library(grafify)

library(future)
library(speckle)
# Set seed for reproducibility
set.seed(1454944673L)
opts_chunk[["set"]](
  audodep = TRUE,
  cache = FALSE,
  cache.lazy = FALSE,
  error = TRUE,
  echo = FALSE,
  fig.height = 5L,
  fig.retina = 2L,
  fig.width = 11,
  message = FALSE,
  tidy = TRUE,
  warning = FALSE
)

ggplot2::theme_set(theme_prism(base_size = 12))
# https://grafify-vignettes.netlify.app/colour_palettes.html
# NOTE change colors here if you wish
scale_colour_discrete <- function(...) {
  scale_colour_manual(..., values = as.vector(grafify:::graf_palettes[["kelly"]]))
}

if (future::supportsMulticore()) {
  future::plan(future::multicore)
} else {
  future::plan(future::multisession)
}
```

```{r sanitize_datatable}
sanitize_datatable <- function(df, ...) {
  # remove dashes which cause wrapping
  DT::datatable(df, ...,
    rownames = gsub("-", "_", rownames(df)),
    colnames = gsub("-", "_", colnames(df)),
    options = list(scrollX = TRUE, ...)
  )
}
```

## Load object and subset
Read in the full dataset Seurat object and subset to keep only cells from the conditions of interest (cold7 vs TN).

```{r load-data}
seurat <- readRDS("data.rds")

subset_seurat <- subset(x = seurat, subset = condition %in% c("Ctrl", "Treat"))

Idents(object = subset_seurat) <- "celltype"

DefaultAssay(subset_seurat) <- "RNA"

# Create metadata df and factor celltype
meta <- subset_seurat@meta.data
meta$celltype <- factor(meta$celltype)
```

## Check count numbers of cells

```{r cellnumbers}
meta$condition_sample <- paste0(meta$condition, "_", meta$sample)
table(meta$condition_sample, meta$celltype)
```

## Run propeller

```{r run-propeller}
# working code; takes into account cell numbers
# NOTE adapt with your variables of interest
propres <- propeller(subset_seurat,
  sample = subset_seurat$sample,
  clusters = subset_seurat$celltype,
  group = subset_seurat$condition
)

sanitize_datatable(propres)
```

## Proportions of cells table

To understand discordance with sccomp results, let's evaluate proportions and variability within group. 


```{r}
props <- getTransformedProps(meta$celltype,
  meta$condition_sample,
  transform = "logit"
)

props$Proportions
```

```{r extracode, eval=FALSE}
meta_propeller <- meta %>%
  group_by(condition, sample) %>%
  distinct(sample)

designAS <- model.matrix(~ meta_propeller$condition)
colnames(designAS) <- c("Ctrl", "Treat")

fit <- lmFit(props$TransformedProps, designAS)
fit <- eBayes(fit, robust = TRUE)

topTable(fit)
```
# Resources

* [Speckle vignette](http://127.0.0.1:29356/library/speckle/doc/speckle.html#finding-significant-differences-in-cell-type-proportions-using-propeller)
* [Application to mouse PBMC scRNA](https://phipsonlab.github.io/propeller-paper-analysis/RealDataAnalysis.html#Young_vs_Old)

# Conclusion

# R session

List and version of tools used for the QC report generation.

```{r}
sessionInfo()
```