We upload the conda environment file (It contains the versions of required packages and other information). To install it, please use:
conda env create --name cpsc488 --file=cpsc488_env.yml
To run our codes, please first pre-process the given dataset based on different tasks.
For cell-type annotation task, please use:
python preprocessing_cta.py
For deconvolution task, please use:
python preprocessing_deconv.py
Then for different tasks, we also have different main files.
For cell-type annotation task, please use:
python cta_main.py
For deconvolution task, please use:
python deconv_main.py
For the usage of baseline models or evaluation, please refer the corresponding folders (baseline and evaluation)
Please refer the notebook under the tutorial folder for running the codes of cell-type annotation task and deconvolution task, also the evaluation.
For the demo datasets, please use this link to download.
If you have questions, please contact Tianyu Liu or Zhiyuan Cao.