MSIGN: A deep learning framework based on multi-scale interaction graph neural networks for predicting binding of synthetic cannabinoids to receptors

Key Points

• The central aim is to enhance the prediction of binding affinities for synthetic cannabinoids to their receptors using deep learning techniques.
• Developed a multi-scale interaction framework using graph neural networks (GNNs)
• Focused on improving 3D ligand-complex representation
• Evaluated model performance against existing methods
• Achieved better accuracy in predicting binding affinities compared to traditional models
• Showed improved generalization across different receptor types
• Identified key structural features influencing binding interactions