This repository contains Rmarkdown files that detail the workflow for performing CellChat analysis on single-cell RNA sequencing data. The analysis is aimed at comparing gene expression patterns across different cell types and samples, particularly focusing on the carotid and femoral cell types.
CellChatBasics.Rmd This file provides the foundational steps for CellChat analysis. It encompasses the primary methods and functions required for the overall analysis.
CellChat_CarotidvsFemoral.Rmd This file dives deeper into a comparative analysis between carotid and femoral samples. It builds upon the basics introduced in CellChatBasics.Rmd and adds specific comparisons between the two sample types.
1. Replicating the Environments & Dependencies
All the necessary libraries should be automatically installed and loaded when you run "renv::restore()" function.
2. Data Checks and Recall
This script checks the unique cell types in the merged dataset and recalls the CellChat object to be used.
3. Sample Subsetting
This section then subsets the data into two categories: carotid and femoral samples.
4. Carotid Sample Analysis
This section kicks off the CellChat analysis focused on carotid samples. It begins by processing the expression data, mapping gene expression onto protein-protein interactions (PPI), determining communication probabilities, filtering out specific cell-cell communications, delving into communication within the framework of signaling pathways, and finally aggregating the overall cell-cell communication network.
5. Femoral Sample Analysis
Similarly, this section covers the analysis of femoral samples.
6. Network Plot Visualization of Carotid vs. Femoral Samples
It plots network visualizations for both the carotid and femoral samples showcasing both interaction counts and weights/strengths.
7. Pathway Analysis
In this section, a detailed pathway analysis is conducted. It calculates the core importance or centrality in the cell-cell communication networks, highlights the strength of interactions through scatter plots, visualizes these interactions with chord diagrams, and illustrates the information using bubble plots.
8. Violin Plots & Ridgeline Plots
For the curated list that describes genes of interest, this section generates violin plots and ridgeline plots to deeply understand the gene expression trends among different cell types or samples.
Make sure the following R packages are installed: renv, CellChat, Seurat. You may use the renv.lock file provided for reconstructing the dependencies needed to run the codes.
Please ensure that the working directories set in the script are adjusted according to the location of your datasets and where you intend to save the visual outputs.
Daniel Lee
August 11, 2023