A group of scientists from the Sbarro Health Research Organization (SHRO), the University of Pisa, and Temple University created KinasePred, an AI-powered tool to predict kinase interactions for potential disease treatments. The study was published in the International Journal of Molecular Sciences (IJMS).
Image Credit: Maryna Olyak/Shutterstock.com
Research on protein kinases offers a chance to investigate molecular targets within the body for the treatment of autoimmune illnesses and cancer. These enzymes may attach to specific locations within cells and prevent abnormal processes like the growth of tumors and the overproduction of malignant cells. Scientists are using artificial intelligence (AI) to forecast and develop a model of which pairing could have a therapeutic effect because of the enormous variety of potential kinase and cell architectures.
The authors include scientists from the University of Pisa and other research institutions in Italy, as well as researchers working with the Sbarro Health Research Organization (SHRO), which is led by Antonio Giordano, M.D., Ph.D., Professor at Temple University. This AI-powered process can forecast kinase activity, learn about the relationships between molecular targets, and find combinations that may be used to treat cancer.
Under the direction of Dr. Miriana de Stefano of the University of Pisa's Department of Pharmacy, the study offers a sophisticated computational tool intended to improve the prediction of kinase interactions with small compounds. One data-dependent computational method designed to address the specific issue of kinase inhibitor selection is KinasePred. This is accomplished by using a prediction model that takes advantage of the molecular underpinnings of kinase binding and selectivity.
KinasePred explains the chemical properties that enable the interactions and makes precise predictions using machine learning (ML) and artificial intelligence (AI). With the use of novel molecular representations and several machine learning techniques, the researchers anticipate that the tool will produce more accurate predictions and offer a more thorough understanding of kinase interactions. These developments are essential for detecting and reducing off-target effects, which will ultimately improve the therapeutic medicines' safety and selectivity.
Journal Reference:
Stefano, D, M., et al. (2025) KinasePred: A Computational Tool for Small-Molecule Kinase Target Prediction. International Journal of Molecular Sciences. doi.org/10.3390/ijms26052157