MGATAF: Multi-channel Graph AttentionNetwork with Adaptive Fusion forCancer-Drug Response Prediction

Drug response prediction is critical in personalized medicine, aiming todetermine the most effective and safe treatments for individual patients.Traditional prediction methods relying on demographic and genetic dataoften fall short in accuracy and robustness. Recent graph-based models,while promising, frequently neglect the critical role of atomic interactionsand fail to integrate drug fingerprints with SMILES for comprehensivemolecular graph construction. We introduces MGATAF (MultimodalMulti-channel Graph Attention Network and Adaptive Fusion), a frame-work designed to enhance drug response predictions by capturing bothlocal and global interactions among graph nodes. MGATAF improvesdrug representation by integrating SMILES and fingerprints, resulting inmore precise predictions of drug effects. The methodology involves con-structing multimodal molecular graphs, employing multi-channel graph attention networks to capture diverse interactions, and using adap-tive fusion to integrate these interactions at multiple abstraction levels.Empirical results demonstrate MGATAF’s superior performance com-pared to traditional and other graph-based techniques. For example,on the GDSC dataset, MGATAF achieved a 5.12% improvement in thePearson correlation coefficient (PCC), reaching 0.9312 with an RMSE of0.0225.Similary, in new cell-line tests, MGATAF outperformed baselineswith a PCC of 0.8536 and an RMSE of 0.0321 on the GDSC dataset,and a PCC of 0.7364 with an RMSE of 0.0531 on the CCLE dataset