This project is for the development of a supervised classifier of cell types based on the single-cell RNAseq digital expression profiles.
You may run the standalone command as follow:
python pred_types.py -i <input_dge.csv> -o <cell_type_output.csv> -p -m <model.h5> -s <human/mouse> -t <id2type.csv>
To prepare input_dge.csv from 10xGenomics output files for the above prediction command, you may run the command as follow:
python transform_10x_to_dge.py -m <matrix.mtx> -g <genes.tsv> -b <barcodes.tsv> -o <transformed_dge.csv>
The R version of SuperCT is now availabe in https://github.com/weilin-genomics/rSuperCT