Separate XAI: Independent Training Framework for Cancer Drug Sensitivity Prediction Using GDSC and CCLE with Explainable AI-Driven Drug Repositioning

ProphDR: An Interpretable Deep Learning Model for Predicting Cancer Drug Response via Multi-Omics and Cross-Attention Mechanisms

ProphDR is an interpretable deep learning framework that integrates multiomics data and drug structural information using a hierarchical attention mechanism, and generates biologically interpretable attention maps that highlight key pharmacophores and resistance-related genes consistent with established mechanisms in NSCLC and BRCA.

Essentiality-driven prediction of anticancer drug responses in preclinical and clinical contexts

DrGee is presented, an essentiality-centered platform that infers drug sensitivity solely from gene expression profiles, and the built-in DeepEEAA model integrates gene expression, gene essentiality, drug-protein affinity, and drug-gene associations to quantitatively predict IC50 values.

Drug sensitivity prediction across cancer types using graph isomorphism networks and biological pathway features: A dual-branch deep learning approach

An innovative dual-branch approach based on Graph Isomorphism Network drug representations coupled with a Multilayer Perceptron (MLP) for 50-dimensional ssGSEA pathway activities calculated from CCLE gene expression is proposed, proving the importance of biological features in the two-branch model.

[Prediction and verification of therapeutic drugs for triple-negative breast cancer using a knowledge graph-based drug repurposing model].

DR.DEGMON: self-explainable deep neural network for drug-induced cell viability prediction incorporating differentially expressed genes and gene ontology

The integration of GO and LRP enabled the model to suggest the underlying biological processes involved in drug responses, making it a valuable tool for predicting outcomes and discovering new biomedical knowledge in cancer pharmacogenomics.

Next-Generation Artificial Intelligence for Drug Repurposing: Integrating Machine Learning, Deep Learning, Multi-Omics, Knowledge Graphs, and Clinical Translation: A Review

Drug repurposing has emerged as a promising strategy to accelerate drug discovery by identifying new therapeutic indications for existing approved or investigational drugs, thereby reducing development time, cost, and clinical risk compared with traditional de novo drug development. The rapid expansion of biomedical big data, together with advances in artificial intelligence (AI) and machine learning (ML), has transformed computational drug repurposing into a data-driven and highly efficient discipline. Conventional machine learning algorithms, including Support Vector Machines, Random Forests, and gradient boosting methods, have demonstrated significant utility in predicting drug-target and drug-disease associations. More recently, deep learning architectures such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Autoencoders, Transformer-based models, and Graph Neural Networks (GNNs) have enabled the integration of heterogeneous datasets, including chemical structures, transcriptomics, proteomics, metabolomics, pharmacogenomics, protein-protein interaction networks, and electronic health records, substantially improving prediction accuracy. This review provides a comprehensive overview of AI-driven drug repurposing, covering computational strategies, publicly available biomedical databases, feature representation methods, machine learning and deep learning algorithms, and their applications in cancer, infectious diseases, neurological disorders, cardiovascular diseases, and rare diseases. Furthermore, recent advances in knowledge graphs, explainable artificial intelligence (XAI), federated learning, foundation models, and large language models (LLMs) are discussed as emerging technologies capable of improving prediction reliability, interpretability, and clinical applicability. Current challenges, including data heterogeneity, limited external validation, algorithmic bias, model interpretability, and regulatory barriers, are critically evaluated. Finally, future perspectives focusing on multimodal multi-omics integration, digital twins, real-world evidence, and precision medicine are presented.