README.md

In CostaLab/scMarkerDetect: Detect marker panel from scRNA-seq profile

sc2marker

sc2marker is a toolkit for identification of cell specific markers for clusters of cells in a single cell RNA sequencing data set. In particular, sc2marker focus on markers that have valid antibodies to be used for either imaging or flow analyis.

Install

You can install sc2marker with the following commands:

install.packages("devtools")
devtools::install_github("https://github.com/CostaLab/sc2marker", build_vignettes = TRUE)

Detect markers for given cell cluster

To run sc2marker you need to execute the following command, providing a clustered single cell data sets (as Seurat object), the cell type of interest and the antibody databased (IHC, ICC or Flow).

require(sc2marker)
require(Seurat)
nk.markers <- Detect_single_marker(mca.spleen, id = "NK cell(Spleen)", category = "Flow", org = "mouse")

You can display the results as a table with the command:

get_antibody(nk.markers, org = "mouse")

and you can generate ridge plot with the following command:

plot_ridge(mca.spleen, id = "NK cell(Spleen)", genes = nk.markers[1:9,]$gene, ncol = 3, assay = "RNA", aggr.other = F)

To use customized gene set, you can run the following command. sc2marker will recognise genes and ignore the cases.

nk.markers <- Detect_single_marker(mca.spleen, id = "NK cell(Spleen)", category = "Flow", geneset = c("CD19", "GeneA", "welcome2022"), org = "mouse")
get_antibody(nk.markers, rm.noab = F)

To include your own antibody database

Example of self made antibody database can be found in Self-made database. Be sure to save the database within your R path and update the code above to reflect the file location in your system.

nk.markers <- Detect_single_marker(mca.spleen, id = "NK cell(Spleen)", category = "Flow", org = "mouse", self.db = "/Path to Self_DB.csv/")
get_antibody(nk.markers, self.db = "/Path to Self_DB.csv/", org = "mouse")

calculate markers for all cell clusters

To calculate markers for all cell clusters, you can do by following command:

all.markers <- Detect_single_marker_all(mca.spleen, category = "Flow", org = "mouse")

To Check T cell markers from results of all clusters, and get the antibody information, you can do following.

t.markers <- all.markers[["T cell(Spleen)"]]
get_antibody(t.markers, org = "mouse")

Generate report

To automatically generate the sc2marker report of all cell clusters, you can run following command:

generate_report(mca.spleen, all.markers, fpath = ".", fname = "mca.spleen")

Or you can only analysis subset of cell clusters (B cell and NK cell) and generate the report as following:

all.markers <- Detect_single_marker_all(mca.spleen, category = "Flow", clusters_to_detect = c("Marginal zone B cell(Spleen)", "NK cell(Spleen)"), org = "mouse")
generate_report(mca.spleen, all.markers, fpath = ".", fname = "mca.spleen")

Reproduction

All code and R object needed to generate the results of sc2marker manuscript can be found in zenodo.

The scripts below reproduce individual analysis. Please note that some packages might require manual installation ( 'seurat-data' and 'ggsci'). Also, configure your R paths before running these scripts.

Code for MCA Spleen (Mouse)

Example for MCA-Spleen in Rmarkdown format and in html format.

Code for Human Bone marrow (human-BM)

SeuratData to load the R object. Example_human_BM.Rmd

Code for Human PBMC&lung (human-PBMC&lung)

Example_human_PBMC&lung.Rmd

Code for Human Bone marrow (human-PBMC)

Example_human_PBMC.Rmd

Also, you can find the original data of MCA, human-PBMC and human-PBMC&lung and Stromal data as following: MCA, human-PBMC&lung, human-PBMC, human-BM, niche stromal

CostaLab/scMarkerDetect documentation built on May 12, 2022, 6:05 a.m.