THis is the code for our work "Multi-view graph convolutional network for predicting cancer cell-specific synthetic lethality". Our model, MVGCN-iSL, comprises three parts. In the first, the GCN processes multiple biological networks independently as cell-specific and cell-independent input graphs to obtain graph-specific representations that provide diverse information for SL prediction. In the second part, a max pooling operation integrates several graph-specific representations into one, and in the third part, a multi-layer deep neural network (DNN) model utilizes these integrated representations as input to predict SL.
We use torch as the architecture to build our deep learning model, and torch-geometric to implement graph neural network models. Here are a list of packages required to run our model:
- numpy
- pandas
- scipy
- scikit-learn
- networkx
- argparse
- tqdm
- torch
- torch-geometric
python main.py
By default, the model runs on the 'K562' cell line using all five graph features and four types of omics features as the input.
Please refer to "main.py" for a list of parameters to be adjusted.