Enhancing protein–ligand pose prediction via dual-branch deep learning with docking-oriented multi-component loss and iterative refinement
KTransPose is presented, a GNN-based protein-ligand pose-refinement framework that combines a dual-branch architecture, a docking-oriented multi-component loss, and iterative refinement, and finds that a TransformerConv-based model provides the strongest overall performance.
Md. Khorshed Alam, Julia Rahman, M. A. Hakim Newton et al.
· Applied intelligence (Boston... · 0 citations