biocmtools

A collection of tools for the BioCM

This is a package for tools used by the BioCM.

Installation

Biocmtools currently requires the following packages:

Seurat
ggplot2
patchwork
dplyr
stringr
scales
presto

To install biocmtools, please do so via devtools:

devtools::install_github("BioinformaticsMUSC/biocmtools")

Usage

There are currently two main tools.

BarProp

This function seamlessly creates a bar proportion plot from a Seurat object. It's parameters are:

• seurat = seurat object
• x_axis = the X axis of the plot, usually cell type or Seurat clusters
• split_by = the category to split each bar by, usually sample, condition, or any other categorical metadata column

BarProp(seurat, x_axis = 'Cell', split_by = 'sample')

MarkerDimPlot

This function creates a plot that includes a DimPlot from Seurat along with a table of markers. This is useful to quickly visualize markers per cluster or cell type. It requires the following parameters:

• seurat = seurat object
• group_by = the category to show in the DimPlot, usually 'Cell', 'Seurat Clusters', or another resolution
• n = the number of markers to show for each cluster or cell type. The default is 20.

This function uses presto to generate markers and patchwork to combine the plot with the table. The resulting plot may need a large saved image size (width = 18 inches, height = 18 inches for example).

MarkerDimPlot(seurat, group_by = 'Cell', n = 20)

CleanDimPlot

This function creates a clean DimPlot from a Seurat object, without axes or a legend. It takes the following parameters:

• seurat = seurat object
• group_by = category in Seurat meta data to group the cells by

CleanDimPlot(seurat, group_by = 'Cell')

MultiDimPlot

This function creates an array of Clean DimPlots split by a specified category in the Seurat meta data. It uses the following parameters:

• seurat = seurat object
• group_by = category in Seurat meta data to group the cells by
• split_by = category in Seurat meta data to split the plots by
• ncol = number of columns for the final figure (ignored for 3 or fewer plots)
• split_order = order that split plots will appear

MultiDimPlot(seurat, group_by = 'Cell', split_by = 'condition', ncol = 4)

RunSCtype

This function runs sc-type on a Seurat objecgt. It uses the following parameters:

• seurat = seurat object
• tissue = the type of tissue you are annotating (must be available in sc-type)
• idents = the clustering column to use (e.g. "seurat_clusters")
• assay = should be RNA or SC (default is SCT)
• new_col_name = the title of the new cell type annotation column (default is sctype_{tissue}_prediction)

RunSCtype(seurat, tissue = 'Brain', idents = 'seurat_clusters', assay = 'SCT')