Deep Learning-Driven Anticancer Drug Discovery: Emodepside as a Potential Therapeutic Candidate for Triple-Negative Breast Cancer

Artificial Intelligence-Guided Prioritization and Experimental Evaluation of Synergistic Target Combinations for Breast Cancer Therapy

Background/Objectives: Breast cancer exhibits substantial molecular heterogeneity, resulting in diverse therapeutic vulnerabilities and limiting the efficacy of single-agent therapies. Although multi-target strategies may offer improved therapeutic benefit, the systematic identification of synergistic higher-order target combinations remains challenging. This study aimed to identify and validate effective higher-order target combinations for heterogeneous breast cancer. Methods: DeepMDS, a previously developed deep learning-based multi-compound synergy prediction model, was used to prioritize candidate target combinations for breast cancer. Exhaustive two-target and three-target combinations were ranked in the gene-expression contexts of ER-positive luminal-like MCF-7 and triple-negative MDA-MB-231 breast cancer cells. The top-ranked target combinations were evaluated using single, pairwise, and triple small interfering RNA (siRNA) perturbations. Representative inhibitors were subsequently assessed in fixed-ratio combinations in MCF-7, MDA-MB-231, and 4T1 cells and in a 4T1 syngeneic mouse experiment. Results: BIRC5-NAMPT-TOP1 ranked first in both cell lines. Triple siRNA co-transfection targeting BIRC5, NAMPT, and TOP1 produced the strongest antiproliferative effects, with inhibition rates of 62.94% in MCF-7 cells and 55.62% in MDA-MB-231 cells. The corresponding inhibitors, LQZ-7I, FK866, and topotecan, exhibited synergistic antiproliferative activity at multiple molar ratios, with combination index values below 1. The optimized 2.5:10:1 molar ratio showed strong synergy in MCF-7, MDA-MB-231, and 4T1 cells, with combination index values of 0.26, 0.19, and 0.15, respectively. In tumor-bearing mice model, the triple-inhibitor regimen achieved a tumor inhibition rate of 62.05%. Topotecan-containing groups showed hematological alterations, whereas no statistically detectable elevations were observed in the measured terminal serum hepatic or renal biomarkers. Conclusions: The BIRC5-NAMPT-TOP1 combination showed reproducible phenotypic activity in the tested genetic and pharmacological models. These findings support the use of DeepMDS as a hypothesis-generation tool for higher-order target prioritization.

Metabolic signatures of triple-negative breast cancer

A Combined Chemoinformatics- and Machine Learning-Based Approach Identifies Chlormidazole as a Drug Repurposing Candidate against Aggressive Prostate Cancer

Despite recent therapeutic advances, treatment options for advanced, therapy-resistant, and metastatic prostate cancer (PCa) remain limited. Here, we developed and prospectively validated an integrated chemoinformatics and machine learning (ML) workflow with ligand-based similarity filtering. Validation on independent external data sets showed that this applicability-domain-guided integration strategy can reduce false positives and improve virtual screening performance. Screening of DrugBank identified five repurposing candidates with confirmed antiproliferative activity in both 2D and 3D PCa models. Among them, the antifungal agent chlormidazole emerged as the most promising candidate, displaying tumor-selective and predominantly cytostatic activity associated with p57 upregulation, reduced Rb phosphorylation, and G1 arrest. Chlormidazole also enhanced the antiproliferative activity of docetaxel in both models, achieving comparable efficacy at substantially lower docetaxel concentrations. These findings identify chlormidazole as a promising repurposing candidate for PCa and demonstrate the value of integrating chemoinformatics with ML for drug repurposing and virtual screening.

Advancing cancer drug discovery through the integration of machine learning and high-throughput screening.

This review highlights the synergy between AI and HTS, emphasizing DL techniques such as convolutional neural networks for bioactivity prediction, recurrent neural networks for de novo design, and reinforcement learning for property optimization.

Deep Learning-Based Molecular Generation for Lung Cancer Therapeutics

Background: The leading cause of cancer-related deaths globally is lung cancer, and the P2X7 receptor (P2X7R) is a promising therapeutic target due to its role in the disease progression. Methods: A deep learning-based molecular generation framework that integrates a fragment-based drug method with Relational Graph Convolutional Networks (RGCNs) and a Wasserstein Generative Adversarial Network (WGAN) was employed. Known P2X7R targeting drugs were fragmented to construct a fragment library, which was used to generate new candidate molecules. The generated molecules from the model were evaluated for chemical validity, novelty, Lipinski’s Rule of Five compliance, quantitative estimate of drug-likeness (QED), lipophilicity (LogP), similarity using the Tanimoto coefficient, and binding affinity through molecular docking. Results: The model generated 4498 chemically valid molecules, including 968 unique and 384 novel molecules. Approximately 97% satisfied standard drug-likeness criteria, with QED values predominantly above 0.6 and LogP values within acceptable pharmacokinetic ranges. The novel molecules demonstrated an improved docking score against P2X7R compared to the seed molecules. Conclusions: Despite training on 5000 SMILES due to limited computational resources, the model achieved high validity, strong molecular diversity, and drug-like physicochemical properties, demonstrating the feasibility of a scalable, target-specific AI pipeline for lung cancer drug discovery using fragment-based molecular generation, RGCN and WGAN. Nevertheless, the biological activity of the generated molecules remains experimentally unvalidated, and the findings are based solely on computational analyses.

A GPT-Driven De Novo Discovery of SIRT2 Inhibitors Against Breast Cancer