Skip to content

Commit

Permalink
Updated vignettes.
Browse files Browse the repository at this point in the history
  • Loading branch information
@PratibhaPanwar
PratibhaPanwar committed Jul 27, 2024
1 parent dc17141 commit af7438e
Show file tree
Hide file tree
Showing 4 changed files with 52 additions and 32 deletions.
Binary file added man/figures/Fig1_schematic.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added man/figures/clustSIGNAL_hex.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
41 changes: 25 additions & 16 deletions vignettes/seqFISH_mouseEmbryo.Rmd → vignettes/1_seqFISH_mouseEmbryo.Rmd
View file
Original file line number Diff line number Diff line change
@@ -1,9 +1,15 @@
---
title: "Single sample analysis"
author:
- Pratibha Panwar
date: "`r Sys.Date()`"
output: rmarkdown::html_vignette
vignette: >
%\VignetteIndexEntry{Getting started: clustSIGNAL}
output:
BiocStyle::html_document:
toc_float: true
BiocStyle::pdf_document: default
package: clustSIGNAL
vignette: |
%\VignetteIndexEntry{clustSIGNAL: Single sample analysis}
%\VignetteEngine{knitr::knitr}
%\VignetteEncoding{UTF-8}
---
Expand Down Expand Up @@ -41,14 +47,13 @@ names(colData(spe))

To run clustSIGNAL, we need the column names of sample and cell labels in the colData dataframe of the spatial experiment object. Here, the cell labels are in the column 'uniqueID' and sample labels are in 'embryo' column.

# Run clustSIGNAL
# Running clustSIGNAL

```{r}
set.seed(100)
samples = "embryo"
cells = "uniqueID"
dimRed = "PCA"
res_emb = clustSIGNAL(spe, samples, cells, dimRed, cluster.fun = "leiden", outputs = "a")
res_emb = clustSIGNAL(spe, samples, cells, outputs = "a")
```

The output variable is a list that can contain dataframe of cluster names, matrix of NN neighbours of each cell, final spe object, or a combination of these, depending on the choice of 'outputs' selected.
Expand All @@ -66,7 +71,7 @@ spe = res_emb$spe_final
spe
```

# Calculate clustering metrics
# Calculating clustering metrics

```{r}
# calculating silhouette width
Expand All @@ -87,7 +92,7 @@ as.data.frame(colData(spe)) %>%
mean_Entropy = mean(entropy))
```

# Visualise clustSIGNAL outputs
# Visualising clustSIGNAL outputs

```{r}
colors = c("#635547", "#8EC792", "#9e6762", "#FACB12", "#3F84AA", "#0F4A9C",
Expand All @@ -100,14 +105,14 @@ colors = c("#635547", "#8EC792", "#9e6762", "#FACB12", "#3F84AA", "#0F4A9C",
"#DA5921", "#E1C239", "#9DD84A")
```

The entropy measures can be very useful in assessing the tissue composition of samples - homogeneous (low entropy) with domain-like regions, or heterogeneous (high entropy) with uniform distribution of cells.
## Entropy spread and distribution

```{r}
# Histogram of entropy spread
hst_ent <- as.data.frame(colData(spe)) %>%
ggplot(aes(entropy)) +
geom_histogram(binwidth = 0.05) +
ggtitle("Entropy spread of regions") +
ggtitle("A") +
labs(x = "Entropy", y = "Number of regions") +
theme_grey() +
theme(text = element_text(size = 12))
Expand All @@ -120,15 +125,17 @@ spt_ent <- as.data.frame(colData(spe)) %>%
aes(colour = entropy)) +
scale_colour_gradient2("Entropy", low = "grey", high = "blue") +
scale_size_continuous(range = c(0, max(spe$entropy))) +
ggtitle("Spatial distribution of region entropy") +
ggtitle("B") +
labs(x = "x-coordinate", y = "y-coordinate") +
theme_classic() +
theme(text = element_text(size = 12))
hst_ent + spt_ent
```

We can also visualize the spatial plot and entropy distribution of the clusters, which provide spatial context of the cells and their neighbourhoods, as well as the compositions of the neighbouhoods.
The spread (A) and spatial distribution (B) of region entropy measures can be very useful in assessing the tissue composition of samples - low entropy regions are more homogeneous with domain-like structure, whereas high entropy regions are heterogeneous with more uniform distribution of cells.

## clustSIGNAL clusters visualisation

```{r}
df_ent = as.data.frame(colData(spe))
Expand All @@ -139,7 +146,7 @@ spt_clust <- df_ent %>%
y = -spatialCoords(spe)[, 2])) +
geom_point(size = 0.5, aes(colour = reCluster)) +
scale_color_manual(values = colors) +
ggtitle("Spatial distribution of clusters") +
ggtitle("A") +
labs(x = "x-coordinate", y = "y-coordinate") +
guides(color = guide_legend(title = "Clusters",
override.aes = list(size = 3))) +
Expand All @@ -162,12 +169,14 @@ box_clust <- df_ent %>%
ggplot(aes(x = reCluster, y = entropy, fill = reCluster)) +
geom_boxplot() +
scale_fill_manual(values = colors_ent) +
ggtitle("Entropy ditribution of clusters") +
ggtitle("B") +
labs(x = "clustSIGNAL clusters", y = "Entropy") +
theme_classic() +
theme(legend.position = "none",
text = element_text(size = 15),
axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
text = element_text(size = 12),
axis.text.x = element_text(angle = 90, vjust = 0.5, hjust = 1))
spt_clust + box_clust + patchwork::plot_layout(guides = "collect", widths = c(1, 2))
```

The spatial location (A) and entropy distribution (B) of the clusters provide spatial context of the cluster cells and their neighbourhoods, as well as the compositions of the neighbouhoods.
43 changes: 27 additions & 16 deletions vignettes/MERFISH_mouseHypothalamus.Rmd → vignettes/2_MERFISH_mouseHypothalamus.Rmd
View file
Original file line number Diff line number Diff line change
@@ -1,9 +1,15 @@
---
title: "Multisample analysis"
author:
- Pratibha Panwar
date: "`r Sys.Date()`"
output: rmarkdown::html_vignette
vignette: >
%\VignetteIndexEntry{Getting started: clustSIGNAL}
output:
BiocStyle::html_document:
toc_float: true
BiocStyle::pdf_document: default
package: clustSIGNAL
vignette: |
%\VignetteIndexEntry{clustSIGNAL: Multisample analysis}
%\VignetteEngine{knitr::knitr}
%\VignetteEncoding{UTF-8}
---
Expand Down Expand Up @@ -41,22 +47,21 @@ names(colData(spe2))

To run clustSIGNAL, we need the column names of sample and cell IDs in the colData dataframe of the spatial experiment object. Here, the cell IDs are in the column 'Cell_ID' and sample IDs are in 'samples' column.

# Run clustSIGNAL
# Running clustSIGNAL

```{r}
set.seed(101)
samples = "samples"
cells = "Cell_ID"
dimRed = "PCA"
res_hyp = clustSIGNAL(spe2, samples, cells, dimRed, outputs = "a")
res_hyp = clustSIGNAL(spe2, samples, cells, outputs = "a")
```

```{r}
spe2 = res_hyp$spe_final
spe2
```

# Calculate clustering metrics
# Calculating clustering metrics

```{r}
samplesList <- levels(spe2[[samples]])
Expand Down Expand Up @@ -89,7 +94,7 @@ as.data.frame(colData(spe2)) %>%
mean_Entropy = mean(entropy))
```

# Visualise clustSIGNAL outputs
# Visualising clustSIGNAL outputs

```{r}
colors = c("#635547", "#8EC792", "#9e6762", "#FACB12", "#3F84AA", "#0F4A9C",
Expand All @@ -102,14 +107,14 @@ colors = c("#635547", "#8EC792", "#9e6762", "#FACB12", "#3F84AA", "#0F4A9C",
"#DA5921", "#E1C239", "#9DD84A")
```

The entropy measures can be very useful in assessing the tissue composition of samples - homogeneous (low entropy) with domain-like regions, or heterogeneous (high entropy) with uniform distribution of cells.
## Entropy spread and distribution

```{r}
# Histogram of entropy spread
hst_ent <- as.data.frame(colData(spe2)) %>%
ggplot(aes(entropy)) +
geom_histogram(binwidth = 0.05) +
ggtitle("Entropy spread of regions") +
ggtitle("A") +
facet_wrap(vars(samples), nrow = 1) +
labs(x = "Entropy", y = "Number of regions") +
theme_grey() +
Expand All @@ -124,15 +129,18 @@ spt_ent <- as.data.frame(colData(spe2)) %>%
scale_colour_gradient2("Entropy", low = "grey", high = "blue") +
scale_size_continuous(range = c(0, max(spe2$entropy))) +
facet_wrap(vars(samples), scales = "free", nrow = 1) +
ggtitle("Spatial distribution of region entropy") +
ggtitle("B") +
labs(x = "x-coordinate", y = "y-coordinate") +
theme_classic() +
theme(text = element_text(size = 12))
theme(text = element_text(size = 12),
axis.text.x = element_text(angle = 90, vjust = 0.5))
hst_ent / spt_ent
```

We can also visualize the spatial plot and entropy distribution of the clusters, which provide spatial context of the cells and their neighbourhoods, as well as the compositions of the neighbouhoods.
In multisample analysis, the spread (A) and spatial distribution (B) of region entropy measures can be useful in assessing and comparing the tissue structure in the samples.

## clustSIGNAL clusters visualisation

```{r}
df_ent = as.data.frame(colData(spe2))
Expand All @@ -148,7 +156,8 @@ spt_clust <- df_ent %>%
guides(color = guide_legend(title = "Clusters",
override.aes = list(size = 3))) +
theme_classic() +
theme(text = element_text(size = 12))
theme(text = element_text(size = 12),
axis.text.x = element_text(angle = 90, vjust = 0.5))
box_clust = list()
for (s in samplesList) {
Expand Down Expand Up @@ -176,11 +185,13 @@ for (s in samplesList) {
theme_classic() +
theme(legend.position = "none",
text = element_text(size = 12),
axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1))
axis.text.x = element_text(angle = 90, vjust = 0.5))
}
spt_clust / (patchwork::wrap_plots(box_clust[1:4], nrow = 1) +
spt_clust / (patchwork::wrap_plots(box_clust[1:3], nrow = 1) +
plot_layout(axes = "collect")) +
plot_layout(guides = "collect") +
plot_annotation(title = "Spatial and entropy distributions of clusters")
```

The spatial location (top) and entropy distribution (bottom) of the clusters can be compared in a multisample analysis, providing spatial context of the cluster cells and their neighbourhood compositions in the different samples.

0 comments on commit af7438e

Please sign in to comment.