Cell type purification by single-cell transcriptome-trained sorting
File description:
*Functions_gate_design.R: functions to source to run the code.
*Code_gate_design.R contains the code to perform GateID gate design and normalization.
To test the code, you can download the following test dataset: "gateID_training_dataset_test.csv" The code designs gates for cluster 1.
The following files are the gate solutions for the test dataset if the code ran correctly: *gateID_solutions_test.csv : this csv file is the equivalent of the gate_sol dataframe that should be obtain after running the code. *gateID_solutions_test.pdf: are the plots of the gate solutions if the code ran correctly. They are rank by decreasing purity (first page of the pdf= higher purity solution).
You can input your own dataset instead of the test dataset. Your dataset should have: *single cells as rows *the first column with a cell type identification: ideally a cluster number *all subsequent column are FACS index values for all available channels
Happy gate design!