A hierarchical docking method: XDZ_2
MGLTools/Smina/Tensorflow
Gasteiger charges
Protein-ligand complexes were used as the training set for binding mode prediction.
Preparation for docking with Smina was done using MGLTools.
Exhaustiveness=50 #exhaustiveness of global search (default=8)
PL_CNNMode model + PL_CNNScore model
Num_modes=100 #max number of poses to generate
Energy_range=10 #energy difference (kcal/mol) between the best and worst binding mode
For a query ligand,the bound protein structure provided by D3R was as the receptor structure for Smina docking.
The binding modes were classified into near-native and non-near-native binding modes with PL_CNNMode model,
a convolutional neural networks (CNN) model for pose prediction that was developed using a training set of the protein-ligand structures.
Then, the near-native models were ranked with the PL_CNNScore model,
a CNN-based scoring function that was developed using a training set of the protein-ligand structures and their corresponding binding affinities.
No
No