0_pancreas_get_MoleculeExperiment.Rmd

---
title: "ABACBS EMCR Hackathon - 0. Pancreas MoleculeExperiments"
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
editor_options: 
  markdown: 
    wrap: 72
---

```{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, including the `tileBoundaries` function from
Github.

```{r}
library(MoleculeExperiment) # for data structure
source(url("https://raw.githubusercontent.com/SydneyBioX/SpatialUtils/refs/heads/main/R/tileBoundaries.R"))
```

In this document we take the full data bundle from the 10x Genomics FFPE
Human Pancreas Xenium dataset, available
[here](https://www.10xgenomics.com/datasets/ffpe-human-pancreas-with-xenium-multimodal-cell-segmentation-1-standard).
Note that we download the "Xenium Output Bundle (full)" zip file. We
then place the unzipped folder into the relative location `../rawData/`.

We read the molecule-level data from this experiment using the
`readXenium()` function, and retain the `cell` boundary information. In
addition, we will generate new boundaries using square tiles at 15um
width. Generating the tile boundaries will take a few minutes to run.

```{r}
me = readXenium("../rawData/Xenium_V1_human_Pancreas_FFPE_outs",
                addBoundaries = c("cell"))
me <- tileBoundaries(me, moleculesAssay = "detected",
                     boundariesAssay = "tiles",
                     tile_width = 15)
me
```

The MoleculeExperiment object now contains information of both the
molecules and the cell and tile boundaries.

# Plot and check

We will use a small window of the entire dataset to visualise the
transcripts, cell boundaries and tile boundaries.

```{r, fig.width = 15, fig.height = 10}
me_sub = subset_by_extent(me, c(xmin = 3000, xmax = 3200, ymin = 1500, ymax = 1700))
g = 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)
g
```

# Save the object

We save the MoleculeExperiment object in the `../ProcessedData/` folder.
In addition, we save the feature names contained in the
MoleculeExperiment object.

```{r}
saveRDS(me, file = "../processedData/xenium_pancreas_me.Rds")
feats = MoleculeExperiment::features(me, "detected")[[1]]
saveRDS(feats, file = "../processedData/xenium_pancreas_features.Rds")
```

# Finish

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