search

masashitsubaki / CPI_prediction

This is a code for compound-protein interaction (CPI) prediction based on a graph neural network (GNN) for compounds and a convolutional neural network (CNN) for proteins.
Apache License 2.0
160 stars 36 forks source link

how to make predictions? #7

Open ghost opened 4 years ago

ghost commented 4 years ago

Thanks for this excellent method for predicting protein-compound interactions. I am just wondering how to make predictions for some compounds? I have trained a model using my own data, the training curve seems to be right. So, how can I make predictions for some other data using this trained model?

bioliuxiang commented 4 years ago

It is recommended that you do not do this. I have made predictions using this code and the results are very bad. A compound is either bound to all proteins or not bound at all.

ghost commented 4 years ago

Oh I see. Thanks.