Deregulated protein kinases are crucial in promoting cancer cell proliferation and driving malignant cell signaling. Although these kinases are essential targets for cancer therapy due to their involvement in cell development and proliferation, only a small part of the human kinome has been targeted by drugs. A comprehensive scoring system is needed to evaluate and prioritize clinically relevant kinases. We recently developed CancerOmicsNet, an artificial intelligence model employing graph-based algorithms to predict the cancer cell response to treatment with kinase inhibitors. The performance of this approach has been evaluated in large-scale benchmarking calculations, followed by the experimental validation of selected predictions against several cancer types. To shed light on the decision-making process of CancerOmicsNet and to better understand the role of each kinase in the model, we employed a customized saliency map with adjustable channel weights. The saliency map, functioning as an explainable AI tool, allows for the analysis of input contributions to the output of a trained deep-learning model and facilitates the identification of essential kinases involved in tumor progression. The comprehensive survey of biomedical literature for essential kinases selected by CancerOmicsNet demonstrated that it could help pinpoint potential druggable targets for further investigation in diverse cancer types.

The binding of small organic molecules to protein targets is fundamental to a wide array of cellular functions. It is also routinely exploited to develop new therapeutic strategies against a variety of diseases. On that account, the ability to effectively detect and classify ligand binding sites in proteins is of paramount importance to modern structure-based drug discovery. These complex and non-trivial tasks require sophisticated algorithms from the field of artificial intelligence to achieve a high prediction accuracy. In this communication, we describe GraphSite, a deep learning-based method utilizing a graph representation of local protein structures and a state-of-the-art graph neural network to classify ligand binding sites. Using neural weighted message passing layers to effectively capture the structural, physicochemical, and evolutionary characteristics of binding pockets mitigates model overfitting and improves the classification accuracy. Indeed, comprehensive cross-validation benchmarks against a large dataset of binding pockets belonging to 14 diverse functional classes demonstrate that GraphSite yields the class-weighted F1-score of 81.7%, outperforming other approaches such as molecular docking and binding site matching. Further, it also generalizes well to unseen data with the F1-score of 70.7%, which is the expected performance in real-world applications. We also discuss new directions to improve and extend GraphSite in the future.

Drug combination therapy shows promise in cancer treatment by addressing drug resistance, reducing toxicity, and enhancing therapeutic efficacy. However, the intricate and dynamic nature of biological systems makes identifying potential synergistic drugs a costly and time-consuming endeavor. To facilitate the development of combination therapy, techniques employing artificial intelligence have emerged as a transformative solution, providing a sophisticated avenue for advancing existing therapeutic approaches. In this study, we developed SynerGNet, a graph neural network model designed to accurately predict the synergistic effect of drug pairs against cancer cell lines. SynerGNet utilizes cancer-specific featured graphs created by integrating heterogeneous biological features into the human protein–protein interaction network, followed by a reduction process to enhance topological diversity. Leveraging synergy data provided by AZ-DREAM Challenges, the model yields a balanced accuracy of 0.68, significantly outperforming traditional machine learning. Encouragingly, augmenting the training data with carefully constructed synthetic instances improved the balanced accuracy of SynerGNet to 0.73. Finally, the results of an independent validation conducted against DrugCombDB demonstrated that it exhibits a strong performance when applied to unseen data. SynerGNet shows a great potential in detecting drug synergy, positioning itself as a valuable tool that could contribute to the advancement of combination therapy for cancer treatment.

SynerGNet is a novel approach to predicting drug synergy against cancer cell lines. In this study, we discuss in detail the construction process of SynerGNet, emphasizing its comprehensive design tailored to handle complex data patterns. Additionally, we investigate a counterintuitive phenomenon when integrating more augmented data into the training set results in an increase in testing loss alongside improved predictive accuracy. This sheds light on the nuanced dynamics of model learning. Further, we demonstrate the effectiveness of strong regularization techniques in mitigating overfitting, ensuring the robustness and generalization ability of SynerGNet. Finally, the continuous performance enhancements achieved through the integration of augmented data are highlighted. By gradually increasing the amount of augmented data in the training set, we observe substantial improvements in model performance. For instance, compared to models trained exclusively on the original data, the integration of the augmented data can lead to a 5.5% increase in the balanced accuracy and a 7.8% decrease in the false positive rate. Through rigorous benchmarks and analyses, our study contributes valuable insights into the development and optimization of predictive models in biomedical research.

Background: The human gut microbiome is critical for host health by facilitating essential metabolic processes. Our study presents a data-driven analysis across 312 bacterial species and 154 unique metabolites to enhance the understanding of underlying metabolic processes in gut bacteria. The focus of the study was to create a strategy to generate a theoretical (negative) set for binary classification models to predict the consumption and production of metabolites in the human gut microbiome. Results: Our models achieved median balanced accuracies of 0.74 for consumption predictions and 0.95 for production predictions, highlighting the effectiveness of this approach in generating reliable negative sets. Additionally, we applied a kernel principal component analysis for dimensionality reduction. The consumption model with a polynomial kernel, and the production model with a radial basis function with 32 reduced features, showed median accuracies of 0.58 and 0.67, respectively. This demonstrates that biological information can still be captured, albeit with some loss, even after reducing the number of features. Furthermore, our models were validated on six previously unseen cases, achieving five correct predictions for consumption and four for production, demonstrating alignment with known biological outcomes. Conclusions: These findings highlight the potential of integrating data-driven approaches with machine learning techniques to enhance our understanding of gut microbiome metabolism. This work provides a foundation for creating bacteria–metabolite datasets to enhance machine learning-based predictive tools, with potential applications in developing therapeutic methods targeting gut microbes.

Background: Invasive fungal infections (IFIs) represent a pressing global health threat, particularly for immunocompromised individuals, yet the development of antifungal drugs continues to lag behind antibacterial therapeutics. In this study, we present a data-driven machine learning framework to predict antifungal compound activity, leveraging cheminformatics and supervised learning. Method: A curated dataset of 3,748 positive (antifungal) and 4,096 negative (non-antifungal) compounds was constructed using ChEMBL, ChemDiv, and HMDB. Chemical class assignment via NPClassifier and Tanimoto similarity filtering ensured non-overlapping, structurally meaningful training data. We extracted 217 molecular descriptors per compound and evaluated physicochemical differences between positive and negative sets, confirming statistically significant divergence in Lipinski parameters (p < 0.001). Results: Feature selection using model-specific importance metrics identified key descriptors such as molecular weight, van der Waals surface area, and nitrogen group counts. Multiple supervised learning models were trained—Random Forest (RF), Extreme Gradient Boosting (XGBoost), Support Vector Machines (SVMs with RBF, polynomial, and sigmoid kernels), and Multi-Layer Perceptron (MLP)—and evaluated using five-fold cross-validation. RF and MLP achieved the highest AUCs of 0.996, with SVM-RBF and XGBoost performing comparably well. To assess generalizability, we introduced chemical class-based cross-validation, wherein compounds were partitioned by their chemical class to reduce information leakage. Despite a slight drop in metrics compared to random splits, all models retained balanced accuracies above 0.91. These results demonstrate the promise of integrating molecular informatics with machine learning for antifungal drug discovery and highlight the importance of rigorous validation strategies aligned with chemical diversity.