DeepGRNCS: Deep learning-based framework for jointly inferring Gene Regulatory Networks across Cell Subpopulations

Dependencies

• Python == 3.9
• Pytorch == 2.0.0
• scikit-learn == 1.3.0
• numpy == 1.22.3
• pandas == 2.0.3

Usage

Dataset

1. For simulated gene expression datasets, regulatory networks between genes are inferred. Prepare the gene expression dataset in the following format:

	Cell1	Cell2	……
Gene1	0.046	1.720	……
Gene2	1.817	0.019	……
……	……	……	……

2. For real gene expression datasets, regulatory networks between tfs and genes are inferred. Prepare the scRNA-seq dataset (as above), along with information on TFs and genes in the following format:

TF	index
TF1	1
TF2	3
……	……

Gene	index
Gene1	0
Gene2	2
……	……

Where 'index' indicates the column index position of the TF (gene) in the scRNA-seq dataset.

Command to run DeepGRNSC

1. Inferring GRNs based on Gaussian simulated datasets:

python main.py --task "DeepGRNCS_demo" --data_file "data/ExpressionData" --save_name "output" --net_number 3

2. Inferring GRN based on BoolODE simulated dataset:

python main.py --task "DeepGRNCS" --data_file "data/ExpressionData" --save_name "output" --net_number 3

3. Inferring GRNs based on real datasets:

python main.py --task "DeepGRNCS_TF" --data_file "data/ExpressionData" --tf_file "data/TF.csv" --gene_file "data/Target.csv"  --save_name "output" --net_number 3

4. Inferring GRN based on a single real dataset:

python main.py --task "DeepGRNCS_TF_Single" --data_file "data/ExpressionData.csv" --tf_file "data/TF.csv" --gene_file "data/Target.csv"  --save_name "output"