rcellmarker provides method to identify cell type based on single cell sequencing data. Since most methods try to annotate cell types manually after clustering the single-cell RNA-seq data. Such methods are labor-intensive and heavily rely on user expertise, which may lead to inconsistent results. Here, we present rcellmarker package -- an automatic tool to annotate cell types from single-cell RNA-seq data. Now only support human, mouse and rat .
library(devtools)
install_github("guokai8/rcellmarker")
set.seed(123)
library(rcellmarker)
gene=sample(unique(human$SYMBOL),20)
res<-cells(gene,species = "human",keytype ="SYMBOL")
head(res)
#if you have the result include the cluster information, padj value and avg_logFC named as "single"
head(getdetail(res,single))
## to show the markers for the cluster
marker(res)
## if yuo have the results from Seurat of cellranger named as 'df'
## default only show the top 3 cell type for each cluster and you can also set the minSize which stand for minimal genes should be include in the celltype
## You can use paramater topn to change it
res <- cellMarker(df,type='seurat',species='human',keytype='SYMBOL',weight=1) #weight for avg_logFC
# or
res <- cellMarker(df,type='cellranger',species='human',keytype='SYMBOL',weight=100)
## only do the assign for cluster 1
res <- cellMarker(df,type='cellranger',species='human',keytype='SYMBOL', cluster=1, weight=100)
# or you just have two columns include cluster and gene name
res <- cellMarker(df,type='custom',species='human',keytype='SYMBOL')
## to show the markers for each cluster
marker(res)
The rcellmarker package use the CellMarker and PanglaoDB database as the reference with ID cleaning and merging. We also add information from the newest public papers related with single cell field from NCBI. The package is still under development. Other species will be supported soon.
For any questions please contact guokai8@gmail.com