Application of Artificial Intelligence-Based Approaches in the Discovery and Development of Protein Kinase Inhibitors (PKIs) Targeting Anticancer Activity

Herein, we present an in-depth review focused on the application of different artificial intelligence (AI) approaches for developing protein kinase inhibitors (PKIs) targeting anticancer activity, focusing on how the AI-based tools are making promising advances in drug design and development, by predicting active compounds for essential targets involved in cancer. In this context, the machine learning (ML) approach performs a critical role by promoting a fast analysis of a thousand potential inhibitors within a small gap of time by processing large datasets of chemical data, putting it at a higher level than other traditionally used methods for screening molecules. In general, AI-based screening of compounds reduces the time of the work and increases the chances of success in the end. Additionally, we have covered recent studies focused on the application of deep neural networks (DNNs) and quantitative structure-activity relationships (QSAR) to identify PKIs. Furthermore, the paper covers new AI-based methodologies for filtering or improving datasets of potential compounds or even targets, such as generative models for the creation of novel compounds and ML-based strategies to collect information from different databases, increasing the efficiency in drug design and development. Finally, this review highlights how AI-based tools are increasing and improving the field of PKIs targeting cancer, making them an alternative for the future in the medicinal chemistry field.