The discovery and development of novel therapeutic agents is a complex, time-consuming, and expensive process, often characterized by high failure rates in clinical trials. In recent years, the integration of computational intelligence and machine learning into cheminformatics and computational chemistry has emerged as a transformative force in drug design. By leveraging the vast amount of chemical and biological data available, computational methods can significantly accelerate the identification of promising drug candidates, optimize lead compounds, and reduce the reliance on costly wet-lab experiments. This special session aims to explore the state-of-art computational methods applied to the chemical domain. We focus on the application of advanced statistical models, deep learning architectures, and data mining techniques to systematically analyze large-scale molecular datasets. A key challenge in this field is accurate property prediction, ranging from physicochemical characteristics (QSPR) to biological activity (QSAR) and toxicity profiles (predictive toxicology). Furthermore, the rise of geometric deep learning and graph neural networks has revolutionized how molecules are represented and analyzed, allowing for the capture of complex structural dependencies that traditional descriptors might miss.

Our special session on “Cheminformatics and computational chemistry for drug discovery and prediction of chemical properties and activities” invites researchers to present novel algorithms, software tools, and application-oriented studies. We aim to bridge the gap between computer science and chemistry, fostering a multidisciplinary dialogue on how computational intelligence can solve fundamental challenges in predicting molecular behavior, generating novel chemical entities, and understanding structure-activity relationships.