New artificial intelligence method predicts drug-drug interactions with greater accuracy
GENNIUS, developed by researchers from the University of Navarra and Stanford University, could revolutionize the process of finding new drugs.
Researchers from the Cima University of Navarra, the School of Engineering-Tecnun and the data Science and Artificial Intelligence Institute (DATAI) of the University of Navarra, in partnership with the research center in Biomedical Informatics at Stanford University, have developed GENNIUS, an innovative artificial intelligence system that could revolutionize the finding of new drugs.
GENNIUS uses graph neural networks (GNN) and deep learning to predict drug-drug interactions. Using nodes (representing drugs and proteins) and their connections, it generates graphical representations that preserve and disseminate core topic biological information. This has enabled researchers to identify hitherto hidden patterns and discover new, previously unknown drug-protein interactions, a decisive step in drug repositioning. In addition, this new system, described in the scientific journal Bioinformaticshas outperformed other state-of-the-art models by evaluating different data instructions with unprecedented accuracy and speed.
"Our system innovatively combines the structure of drug-protein interaction networks with detailed molecular features, enabling it to make more accurate and generalizable predictions. This breakthrough not only improves interaction prediction, but also provides a useful and rapid tool for researchers. Furthermore, by facilitating the finding new uses for already approved drugs, GENNIUS could identify therapeutic alternatives for patients and significantly reduce the time and costs associated with the development of new treatments," explains Dr. Mikel Hernáez, researcher at Cima University of Navarra and DATAI, lead author of the article.
The scientists plan to continue improving GENNIUS and hope that it will soon be applied in real drug finding projects, accelerating the path to clinical trials and the development of innovative treatments. To further advance this technology and foster new programs of study, the researchers have made GENNIUS available to the academic community. This innovation reinforces the potential of artificial intelligence in biomedical research , opening new doors in the fight against complex diseases.
award for the best methodological contribution
This research has been distinguished as "Best Methodological Contribution" in the III DATAI Awards, an award given by the data Science and Artificial Intelligence Institute (DATAI) to its researchers. The goal of these awards is to highlight the impact and innovation in the use of data science and artificial intelligence of pioneering scientific contributions, fostering excellence inmultidisciplinary research and strengthening its influence in the international academic community .
Idoia Ochoa, researcher at Tecnun and DATAI, also author of the study, collected the award on behalf of the group researcher who developed GENNIUS.
