Dataset DOI: 10.5061/dryad.m0cfxppjn

Description of the data and file structure

README for Single cell RNAseq generated by the Chromium Single Cell Platform (10× Genomics). The gz files should include barcodes, features, and matrix.

This README provides a guide to the contents of the R command files for violinplot, dotplot, heatmap, UMAP, and Chordplot. RStudio 2026.04.0+526 or later versions could be used for analyses. They could be run independent from each other, as each one contains the full procedure starting from package installations and file inputs.

Package functions and purpose of use in each command are described below:

• Violinplot_code.R

1. Seurat

This is the core package used for almost the entire workflow. It is a specialized toolkit for single-cell genomics. Functions used: Read10X(), CreateSeuratObject(), NormalizeData(), FindVariableFeatures(), ScaleData(), RunPCA(), FindIntegrationAnchors(), IntegrateData(), DefaultAssay(), RunUMAP(), FindNeighbors(), FindClusters(), DimPlot(), NoLegend(), FindAllMarkers(), FeaturePlot(), and VlnPlot().

Purpose: Loading data, QC, normalization, dataset integration, clustering, dimensional reduction (PCA/UMAP), and genomic plotting.

2. dplyr

This package is part of the tidyverse ecosystem and is used for data manipulation. Functions used: The pipe operator %>%, group_by(), top_n(), summarise(), mutate(), and n().Purpose: Filtering the top 10 marker genes for each cluster and calculating cell count proportions from the data matrix.

3. ggplot2

This is a popular data visualization package, also part of the tidyverse. Functions used: ggplot(), aes(), geom_bar(), and labs().

Purpose: Generating the custom bar chart at the bottom of your script to compare cell cluster proportions between your samples.

4. BiocManager (Installation Only)

While it is explicitly mentioned at the top of your script, it is not used to run any data processing commands. It is only included as a package manager utility tool to download and install the Bioconductor-dependent version of Seurat.

• dotplot_code.R

1. Seurat

This remains the core package powering the data structures, manipulation, and genomics-specific plotting for the script.Functions used: Read10X(), CreateSeuratObject(), merge(), NormalizeData(), FindVariableFeatures(), ScaleData(), subset(), and DotPlot().

Purpose: Loading 10X data matrices, creating and combining sample-specific data objects, executing basic normalization and scaling, filtering for Myh11-positive cells, and constructing the comparative dot plot.

2. ggplot2: This visualization package is explicitly used to modify and refine the appearance of the generated genomic plots.

Functions used: scale_color_gradient(), theme_minimal(), labs(), theme(), element_text(), and ggtitle().Purpose: Customizing the plot aesthetics by mapping expression scales (from light blue to red), adjusting title/axis labels, and rotating X-axis gene text labels for clean readability.

• Heatmap_code.R

1. Seurat

This is the core toolkit used to manage the single-cell expression matrices and perform initial data transformations.Functions used: Read10X(), CreateSeuratObject(), NormalizeData(), FindVariableFeatures(), ScaleData(), and FetchData().

Purpose: Loading 10X sparse matrix directories, creating separate project data structures, standardizing gene expression scales, and cleanly pulling out numerical matrix values for your 52 specific genes of interest.

2. pheatmap

This package specializes in creating highly customizable, publication-quality aesthetic cluster heatmaps. Functions used: pheatmap().

Purpose: Rendering the individual expression profiles for both projects with custom clustering behaviors (clustering the cell columns while locking the gene rows in their specified order) and suppressing immediate drawing to save them as programmatic objects.

3. gridExtra

This package provides extensions to the grid graphics system to handle the positioning of arbitrary graphical objects on a canvas. Functions used: grid.arrange().

Purpose: Capturing the raw graphical elements from your two standalone heatmaps (using the [[4]] grob extraction subset) and drawing them side by side in a clean, two-column layout.

• UMAP_code.R

1. Seurat

This is the core package used for almost the entire data processing workflow, spatial/dimensional modeling, and genomic visualization.

Functions used: Read10X(), CreateSeuratObject(), NormalizeData(), FindVariableFeatures(), ScaleData(), RunPCA(), FindIntegrationAnchors(), IntegrateData(), DefaultAssay(), RunUMAP(), FindNeighbors(), FindClusters(), DimPlot(), NoLegend(), FindAllMarkers(), and FeaturePlot().

Purpose: Ingesting raw expression data, normalizing and scaling individual datasets, identifying anchors to correct for batch effects between conditions, executing PCA and UMAP, partitioning cells into clusters, and rendering cell distribution maps.

2. dplyr

This package is part of the tidyverse ecosystem and is used for fast, declarative data frame manipulation. Functions used: The pipe operator %>%, group_by(), and top_n().Purpose: Slicing the large table generated by FindAllMarkers() to group the data by cell cluster and extract only the top 10 most statistically significant marker genes per cluster based on log fold-change.

• Chordplot_code.R

1. Seurat

This is the core toolkit used to ingest, normalize, and extract highly variable biological signals from the dataset.Functions used: Read10X(), CreateSeuratObject(), NormalizeData(), FindVariableFeatures(), ScaleData(), and VariableFeatures().

Purpose: Loading the 10X sparse matrix files, creating the structured dataset object, standardizing expression levels, and extracting the names of the top 50 most variable genes to use for downstream interaction simulation.

2. dplyr

This package is part of the tidyverse ecosystem and handles data frame subsetting and sequencing layout operations. Functions used: The pipe operator %>%, arrange(), and desc().

Purpose: Sorting the generated table of simulated gene pairings in descending order based on their connection scores and slicing the top 75 strongest interactions for the visualization matrix.

3. circlize

This package provides a comprehensive layout engine for rendering multi-dimensional data structures into circular tracks and connection links . Functions used: circos.clear() and chordDiagram().

Purpose: Resetting the graphics device canvas to prevent layout overlap and mapping the tabular gene-to-gene cross-references into an interconnected chord plot.

Code/software

README for Single cell RNAseq generated by Chromium Single Cell Platform (10× Genomics). The gz files should include barcodes, features, and matrix.

This README provides a guide to the contents of the R command files for violinplot, dotplot, heatmap, UMAP, and Chordplot. RStudio 2026.04.0+526 was used for analyses. They could be run independently from each other, as each one contains the full procedure starting from package installations and file inputs.