SubClusterTool

Overview

SubClusterTool is an R package designed to facilitate subclustering and integration of subclusters back into Seurat objects. It allows users to extract a specific cluster from a Seurat object, perform subclustering with custom resolutions and dimensions, and merge the refined subclusters back into the original Seurat object with user-defined labels. This tool is particularly useful for detailed single-cell RNA sequencing analysis, where deeper insights into specific clusters are needed.

Installation

First, install the devtools package if you haven't already:

install.packages("devtools")

Then, install the SubClusterTool package from GitHub:

devtools::install_github("WilsonWukz/SubClusterTool")

Usage

To use the SubClusterTool package, follow these steps:

1. Load the package and required dependencies:

library(Seurat)
library(SubClusterTool)

2. Subcluster a specific cluster within a Seurat object:

Example: Subcluster cluster 0 with resolution 0.5 and using the first 10 PCA dimensions

cluster_subset <- subcluster_view(seurat_obj = your_seurat_object, cluster_id = 0, resolution = 0.5, dims = 1:10)

# Visualize the subclustered result
DimPlot(cluster_subset, reduction = "umap", label = TRUE, pt.size = 0.5)

3. Merge subclusters back into the original Seurat object:

Define new labels for the subclusters

new_labels <- c("Subcluster_1", "Subcluster_2", "Subcluster_3")

Merge the subclusters back into the original object

your_seurat_object <- merge_subclusters(
  seurat_obj = your_seurat_object,
  subcluster_obj = cluster_subset,
  new_labels = new_labels,
  new_id_column = "refined_clusters"
)

Visualize the updated Seurat object with refined clusters

DimPlot(your_seurat_object, reduction = "umap", group.by = "refined_clusters", label = TRUE, pt.size = 0.5)

Example

Then the code should be

B_subset <- subcluster_view(seurat_obj = Cell.integrated, cluster_id = "B cells", resolution = 0.6, dims = 1:10)

Then we will get Then we try to merge them back to the Cell.intergrated, and if I want to devide the B_subsets into 2 kinds, B0 and B1

Cell.integrated <- merge_subclusters(
  seurat_obj = Cell.integrated,
  subcluster_obj = B_subset,
  new_labels = c("B0", "B0", "B0", "B0", "B1", "B1","B1"),
  new_id_column = "refined_clusters"
)

Functions

Built-in Functions

• subcluster_view(seurat_obj, cluster_id, resolution, dims)
  Extracts a specified cluster from a Seurat object, performs subclustering, and returns the subclustered Seurat object.

• merge_subclusters(seurat_obj, subcluster_obj, new_labels, new_id_column)
  Merges the refined subclusters back into the original Seurat object with user-defined labels. The merged labels are stored in a new column.

Author

Kezhao Wu
Maintainer: Kezhao Wu

Contributor

• wenxian-lin
• XiaminZhang
• xiulimao

License

MIT

Bug Reports

If you encounter any issues or have suggestions for improvements, please report them on the GitHub issues page.

URL

GitHub Repository