Graph attention network for link prediction of gene regulations from single cell RNA-sequencing data

Dependencies

• Python == 3.8
• Pytorch == 1.6.0
• scikit-learn==1.0.2
• numpy==1.20.3
• scanpy==1.7.2
• gseapy==0.10.8

Usage

1. Preparing for gene expression profiles and gene-gene adjacent matrix

   GENELink integrates gene expression matrix (N×M) with prior gene topology (N×N) to learn low-dimensional vertorized representations with supervision.

2. Command to run GENElink

   To train an ab initio model, simply uses the script 'main.py'.

   python main.py

   To apply dot product as score metric:

   Type == 'dot', flag== False

   To apply causal inference:

   Type == 'MLP', flag==True

License

GENELink is licensed under the MIT License.