sciPENN (single cell imputation Protein Embedding Neural Network) is a deep learning computational tool that is useful for analyses of CITE-seq data. sciPENN can be used to:
- Predict proteins in a query scRNA-seq dataset from a reference CITE-seq dataset.
- Integrate the scRNA-seq and CITE-seq data into a shared latent space.
- Combine multiple CITE-seq datasets with different protein panels by imputing missing proteins for each CITE-seq dataset.
- Transfer cell-type labels from a reference CITE-seq dataset to a query scRNA-seq dataset.
To find code to reproduce the results we generated in that paper, please visit this separate github repository, which provides all code (including that for other methods) necessary to reproduce our results.
Recomended installation procedure is as follows.
- Install Anaconda if you do not already have it.
- Create a conda environment, and then activate it as follows in terminal.
$ conda create -n scipennenv
$ conda activate scipennenv
- Install an appropriate version of python.
$ conda install python==3.7
- Install nb_conda_kernels so that you can change python kernels in jupyter notebook.
$ conda install nb_conda_kernels
- Finally, install sciPENN.
$ pip install sciPENN
Now, to use sciPENN, always make sure you activate the environment in terminal first ("conda activate scipennenv"). And then run jupyter notebook. When you create a notebook to run sciPENN, make sure the active kernel is switched to "scipennenv"
A tutorial jupyter notebook, together with a dataset, is publicly downloadable.
- Python >= 3.7
- torch >= 1.6.1
- scanpy >= 1.7.1
- pandas >= 1.1.5
- numpy >= 1.20.1
- scipy >= 1.6.1
- tqdm >= 4.59.0
- anndata >= 0.7.5
- numba <= 0.50.0