AI Model Reads Molecules Atom by Atom to Predict Dangerous Drug Combinations
A new multimodal transformer-CNN model called InfoMedex predicts 168 types of drug-drug interactions from molecular structure and uses atom-level attribution ...
A new multimodal transformer-CNN model called InfoMedex predicts 168 types of drug-drug interactions from molecular structure and uses atom-level attribution ...
A new multimodal AI framework called MEGH-DDI combines chemical language models with graph neural networks to predict drug–drug interactions with ...
A new multiscale heterogeneous graph transformer called HierHGT-DTI achieves large gains in cold-start drug-target interaction prediction, especially for protein targets ...
© 2025 Scienmag - Science Magazine
© 2025 Scienmag - Science Magazine