# load required packages
library(simpleSeg)
library(ggplot2)
library(EBImage)
library(cytomapper)
# Install the package from Bioconductor
if (!requireNamespace("BiocManager", quietly = TRUE)) {
install.packages("BiocManager")
}
BiocManager::install("simpleSeg")
The simpleSeg
package extends existing bioconductor packages such as cytomapper
and EBImage
by providing a structured pipeline for creating segmentation masks from multiplexed cellular images in the form of tiff stacks. This allows for the single cell information of these images to be extracted in R, without the need for external segmentation programs. simpleSeg
also facilitates the normalisation of cellular features after these features have been extracted from the image, priming cells for classification / clustering. These functions leverage the functionality of the EBImage
package on Bioconductor. For more flexibility when performing your segmentation in R we recommend learning to use the EBimage
package. A key strength of simpleSeg
is that we have coded multiple ways to perform some simple segmentation operations as well as incorporating multiple automatic procedures to optimise key parameters when these aren’t specified.
In the following we will reanalyse two MIBI-TOF images from (Risom et al., 2022) profiling the spatial landscape of ductal carcinoma in situ (DCIS), which is a pre-invasive lesion that is thought to be a precursor to invasive breast cancer (IBC). These images are stored in the “extdata” folder in the package. When the path to this folder is identified, we can read these images into R using readImage
from EBImage
and store these as a CytoImageList
using the cytomapper
package.
# Get path to image directory
pathToImages <- system.file("extdata", package = "simpleSeg")
# Get directories of images
imageDirs <- dir(pathToImages, "Point", full.names = TRUE)
names(imageDirs) <- dir(pathToImages, "Point", full.names = FALSE)
# Get files in each directory
files <- files <- lapply(
imageDirs,
list.files,
pattern = "tif",
full.names = TRUE
)
# Read files with readImage from EBImage
images <- lapply(files, EBImage::readImage, as.is = TRUE)
# Convert to cytoImageList
images <- cytomapper::CytoImageList(images)
mcols(images)$imageID <- names(images)
simpleSeg
accepts an Image
, list
of Image
’s, or CytoImageList
as input and generates a CytoImageList
of masks as output. Here we will use the histone H3 channel in the image as a nuclei marker for segmentation. By default, simpleseg
will isolate individual nuclei by watershedding using a combination of the intensity of this marker and a distance map. Nuclei are dilated out by 3 pixels to capture the cytoplasm. The user may also specify simple image transformations using the transform
argument.
masks <- simpleSeg::simpleSeg(images,
nucleus = "HH3",
transform = "sqrt"
)
The display
and colorLabels
functions in EBImage
make it very easy to examine the performance of the cell segmentation. The great thing about display
is that if used in an interactive session it is very easy to zoom in and out of the image.
# Visualise segmentation performance one way.
EBImage::display(colorLabels(masks[[1]]))
The plotPixels
function in cytomapper
make it easy to overlay the masks on top of the intensities of 6 markers. Here we can see that the segmentation appears to be performing reasonably.
# Visualise segmentation performance another way.
cytomapper::plotPixels(
image = images[1],
mask = masks[1],
img_id = "imageID",
colour_by = c("PanKRT", "GLUT1", "HH3", "CD3", "CD20"),
display = "single",
colour = list(
HH3 = c("black", "blue"),
CD3 = c("black", "purple"),
CD20 = c("black", "green"),
GLUT1 = c("black", "red"),
PanKRT = c("black", "yellow")
),
bcg = list(
HH3 = c(0, 1, 1.5),
CD3 = c(0, 1, 1.5),
CD20 = c(0, 1, 1.5),
GLUT1 = c(0, 1, 1.5),
PanKRT = c(0, 1, 1.5)
),
legend = NULL
)