3_pancreas_exploration.Rmd

---
title: "ABACBS EMCR Hackathon - 3. Some (unstructured) exploration of hackathon data bundle"
author: "Shila Ghazanfar"
date: "`r format(Sys.time(), '%d %B, %Y')`"
output:
       html_document:
                     toc: true
                     toc_float:
                           collapsed: false
                           smooth_scroll: false
                     code_folding: hide
                     fig_width: 10 
                     fig_height: 8
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE,
                      warning = FALSE,
                      message = FALSE,
                      cache = FALSE,
                      cache.lazy = FALSE)

set.seed(2024)
```

# Load packages and data
Load scripts and packages. Load the processed data we generated earlier.

```{r}
library(MoleculeExperiment) # for data structure
library(SpatialExperiment) # for data structure
library(randomForest) # for predictive model
library(scater) # for plotting
library(patchwork) # for plotting
```

```{r}
pancreas_colours = readRDS("../analysisOutput/pancreas_colours.Rds")
fit = readRDS("../analysisOutput/sc_cell_type_RF_fit.Rds")
feats = rownames(fit$importance)
me = readRDS("../processedData/xenium_pancreas_me.Rds")
me
spe_cells = readRDS("../processedData/xenium_pancreas_spe_cells.Rds")
spe_tiles = readRDS("../processedData/xenium_pancreas_spe_tiles.Rds")
```

# Plots

Generate plots of cells/tiles in space coloured by predicted cell type

```{r, fig.width = 15, fig.height = 20}
g = plotReducedDim(spe_cells, "spatial",
                   colour_by = "cell_type_pred", point_size = 0.1) + 
  ggtitle("cells") +
  scale_colour_manual(values = pancreas_colours) +
  coord_fixed() +
  plotReducedDim(spe_tiles, "spatial",
                 colour_by = "cell_type_pred", point_size = 0.1) + 
  ggtitle("tiles") + 
  scale_colour_manual(values = pancreas_colours) + 
  coord_fixed() +
  plot_layout(nrow = 2) +
  NULL
g


library(ggpubr)
g_0 = plotReducedDim(spe_cells, "spatial",
                   colour_by = "cell_type_pred", point_size = 0.1) + 
  ggtitle("cells") +
  scale_colour_manual(values = pancreas_colours) +
  coord_fixed() + 
  guides(colour = guide_legend(override.aes = list(size = 10), title = ""))

g_leg = as_ggplot(get_legend(g_0))
```

# Calculate distances between cells

```{r}
library(BiocNeighbors)

cells2tiles = queryKNN(reducedDim(spe_tiles, "spatial"),
                       reducedDim(spe_cells, "spatial"),
                       k = 1,
                       get.distance = TRUE)$distance

tiles2cells = queryKNN(reducedDim(spe_cells, "spatial"),
                       reducedDim(spe_tiles, "spatial"),
                       k = 1,
                       get.distance = TRUE)$distance

spe_tiles$dist2cell <- pmin(tiles2cells, 30)
spe_cells$dist2tile <- cells2tiles


g = plotReducedDim(spe_cells, "spatial",
                   colour_by = "dist2tile", point_size = 0.5) + 
  ggtitle("cells") +
  coord_fixed() +
  plotReducedDim(spe_tiles, "spatial",
                 colour_by = "dist2cell", point_size = 0.5) + 
  ggtitle("tiles") + 
  coord_fixed() +
  plot_layout(nrow = 2) +
  NULL
g

boxplot(spe_tiles$dist2cell ~ spe_tiles$cell_type_pred, las = 2)

```

Make a plot

```{r}
me_sub = subset_by_extent(me, c(xmin = 3000, xmax = 3200, ymin = 1500, ymax = 1700))
g2 = ggplot_me() +
  geom_polygon_me(me_sub, assayName = "cell", fill = NA, colour = "red") +
  geom_polygon_me(me_sub, assayName = "tiles", fill = NA, colour = "blue") +
  geom_point_me(me_sub, colour = "grey", size = 0.1)
g2

spe_cells_sub = spe_cells[, spe_cells$x_location >= 3000 & 
                            spe_cells$x_location <= 3200 & 
                            spe_cells$y_location >= 1500 & 
                            spe_cells$y_location <= 1700]



spe_tiles_sub = spe_tiles[, spe_tiles$x_location >= 3000 & 
                            spe_tiles$x_location <= 3200 & 
                            spe_tiles$y_location >= 1500 & 
                            spe_tiles$y_location <= 1700]



g2 + geom_point(aes(x = x_location, y = y_location, colour = cell_type_pred), 
               data = as.data.frame(colData(spe_cells_sub)),
               size = 2) + 
  scale_colour_manual(values = pancreas_colours) +
  
  g2 + geom_point(aes(x = x_location, y = y_location, colour = cell_type_pred), 
                 data = as.data.frame(colData(spe_tiles_sub)),
                 size = 2) + 
  scale_colour_manual(values = pancreas_colours) + 
  g_leg +
  plot_layout(nrow = 1)
  
```


# Finish

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