DLIGAND2 is a knowledge-based method to predict protein-ligand binding affinity based on a distance-scaled, finite, ideal-gas reference (DFIRE) state.
Virtual Screening can simulate the interaction between target and drug candidate, calculate the affinity between the two, reduce the actual number of compounds screened, and improve the efficiency of lead compound discovery.
Structure-aware protein-protein interacting site prediction using deep graph convolutional network.
Structure-aware protein solubility prediction from sequence through graph convolutional network and predicted contact map.
Predicting secondary structure, local backbone angles, ASA, HSE, and Expected Errors of proteins.
Improved prediction on secondary structure and other structural properties of proteins by LSTM.
Predicting protein functions from sequence which combines deep convolutional neural network (CNN) model with sequence similarity based predictions.
Copyright © NSCC-GZ