DeepCCI (Deep learning framework for Cell-Cell Interactions inference from scRNA-seq data)

DeepCCI is a graph convolutional network (GCN)-based deep learning framework for Cell-Cell Interactions inference from scRNA-seq data.

Installation:

From Source:

Start by grabbing this source codes:

git clone https://github.com/JiangBioLab/DeepCCI.git
cd DeepCCI

(Recommended) Use python virutal environment with conda（https://anaconda.org/）

conda create -n deepcciEnv python=3.7.4 pip
conda activate deepcciEnv
pip install -r requirements.txt

because data processing and result visualization require R and then also install the following in R:

conda install r-base
R
install.packages(‘Seurat’)
install.packages("igraph")
install.packages('NMF')
install.packages("devtools")
devtools::install_github("jokergoo/circlize")
devtools::install_github("jokergoo/ComplexHeatmap")
devtools::install_github("sqjin/CellChat")
devtools::install_github('satijalab/seurat-data')

Quick Start

1. Cell Cluster Model

(1) Preprocess input files

The cluster model of DeepCCI accepts scRNA-seq data format: CSV and h5. The processed feature file of scRNA-seq data will be provided. Depending on the size of the scRNA-seq file，the process will take about 5-10 minutes.

CSV format

Take an example of Yan 's (GSE36552).

cd Cluster_model
python preprocess.py --name Yan --file_format csv

h5 format

Take an example Qx Limb Muscle (GSE109774 ).

cd Cluster_model
python preprocess.py --name Quake_10x_Limb_Muscle --file_format h5

(2) Cell Clustering

The clustering results of scRNA-seq data will be output.

With pre-train:

It will take about 25 minutes.

python Cluster.py --name Yan --pretain True --pretrain_epoch 50 --device cuda

Without pre-train: The pretrained model files are in the pretain_model folder. It will take about 5 minutes.

python Cluster.py --name Yan --pretain False --device cuda
python Cluster.py --name Quake_10x_Limb_Muscle --pretain False --device cuda

2. Cell Interaction Model

(1) Preprocess input files

The example test file can be download from http://jianglab.org.cn/deepcci_download/. The processed feature file will be provided. Depending on the size of the scRNA-seq file，the process will take about 10-20 minutes.

With cell-label:

cd Interaction_model
python Feature.py --label_mode True --species Human

Without cell-label

python Feature.py --label_mode False --species Human

(2) Interaction Inference

The predicted interaction outfile will be provided. The predicted process will take about 1-2 minutes.

python Interaction_inference.py --device cuda 

3. Visualization

To show the CCI output intuitively, several visualization methods are provided. The process will take about 1 minutes.

cd Plot
python Plot.py

Contact

Feel free to submit an issue or contact us at wenyiyang22@163.com for problems about the package.