cellplm 0.1.0

Package of CellPLM: A pretrain-ed cell language model beyond single cells. Paper link: https://www.biorxiv.org/content/10.1101/2023.10.03.560734

CellPLM

This is the official codebase for CellPLM: Pre-training of Cell Language Model Beyond Single Cells.

CellPLM is the first single-Cell Pre-trained Language Model that encodes cell-cell relations and it consistently outperforms existing pre-trained and non-pre-trained models in diverse downstream tasks, with 100x higher inference speed compared to existing pre-trained models.

Installation

We plan to release our project on PyPI soon. For now, please follow the instruction below to set up the environment for CellPLM:

Quick Installaton

git clone [e-mail]:OmicsML/cellplm.git && cd cellplm
pip install -r requirements.txt

Complete Installation (recommended for HPC users)

conda create -n cellplm python=3.9 -y && conda activate cellplm
conda install cudatoolkit=11.7 -c pytorch -c nvidia
pip install torch==1.13.0+cu117 --extra-index-url https://download.pytorch.org/whl/cu117
pip install einops ipdb pydance torchmetrics wandb hdf5plugin dgl mygene

We recommend using python 3.9 and cuda 11.7 but they are adjustable.

Tutorials

We offer several notebooks for various downstream tasks as introductory tutorials.

We are also working on developing more streamlined protocols for supported tasks and a comprehensive documentation. We aim to release these by the end of the year.

Pretrained CellPLM Model Checkpoints

The checkpoint can be acquired from our dropbox. We might update our checkpoints from time to time.

[10/10/2023] The latest version is 20230926_85M.

Citation

@article{wen2023cellplm,
  title={CellPLM: Pre-training of Cell Language Model Beyond Single Cells},
  author={Wen, Hongzhi and Tang, Wenzhuo and Dai, Xinnan and Ding, Jiayuan and Jin, Wei and Xie, Yuying and Tang, Jiliang},
  journal={bioRxiv},
  pages={2023--10},
  year={2023},
  publisher={Cold Spring Harbor Laboratory}
}