AutomaticCellTypeIdentification

AutomaticCellTypeIdentification is a wrapper of published automatic cell type identification methods which contains supervised methods, unsupervised methods and semi-supervised methods.

Installation

Install from R

You can install AutomaticCellTypeIdentification from github with:

devtools::install_github('xiebb123456/AutomaticCellTypeIdentification')

Note: AutomaticCellTypeIdentification is a wrapper of published methods, the needed package is in Description file.

Install from docker

sudo docker pull registry.cn-hangzhou.aliyuncs.com/xiebb123456/automaticcelltypeidentification

Running AutomaticCellTypeIdentification methods

Now, three interface of eagersupervised, lazysupervised, markersupervised methods supports the available automatic methods.
eagersupervised methods include ACTINN, CaSTLe, CHETAH, clustifyr, Garnett, Markercount, MARS, scClassifR, scHPL, SciBet, scID, scLearn, scmapcluster, scPred, scVI, Seurat, SingleCellNet and SingleR.
lazysupervised methods include CELLBLAST and scmapcell.
markersupervised methods include scTyper, Markercount, SCSA, DigitalCellSorter and SCINA.

Prepare input data

The input of training and testing data is count matrix, the row is gene and the column is cell.

Example with running eagersupervised methods Seurat

eagersupervised(train,test,label_train,method='Seurat')

Example with running lazysupervised methods CELLBLAST

lazysupervised(train,test,label_train,method='CELLBLAST')

Example with running markersupervised methods SCSA

markersupervised(test,marker,method='SCSA')

Note: the conda environment of python-based method is needed to load at the beginning in R.

Tutorial

For more details and basic usage see following tutorials: Guided Tutorial

Contact

Feel free to submit an issue or contact us at xiebb7@mail.sysu.edu.cn for problems about the package installation and usage.