Researchers at the University of Navarra in Spain, have developed RNACOREX, a new open-source software tool that reveals hidden genetic regulatory networks involved in cancer and helps predict patient survival. Tested across 13 different tumor types using data from The Cancer Genome Atlas (TCGA), RNACOREX identifies key interactions between microRNAs and messenger RNAs—molecular relationships that are often missed by conventional analyses.

Unlike many artificial intelligence models that function as “black boxes,” RNACOREX produces interpretable molecular maps that show how genes interact within tumors. These networks can stratify patients according to survival probability with accuracy comparable to advanced AI approaches, while clearly explaining the biological mechanisms behind the predictions.

Published in PLOS Computational Biology, the study demonstrates how RNACOREX can uncover shared molecular patterns across cancers, highlight individual molecules of biomedical interest, and generate new hypotheses about tumor progression. Freely available via GitHub and PyPI, the tool is designed to be accessible for research laboratories worldwide and represents a step forward in explainable AI for precision oncology.

C. Ola Landgren, M.D., Ph.D., received HealthTree Foundation’s prestigious 2025 Innovation Award for his work in developing CORAL, a new research tool that leverages AI to predict individual outcomes and guide treatment decisions in patients with multiple myeloma.

At the core of each of our cells, the DNA molecule is constantly changing its configuration. When it twists or unfolds, it determines which genes are activated at each moment.

"We have shown that the way DNA twists is a layer of gene expression regulation that had previously gone unnoticed," says CNIO researcher Felipe Cortés.

Many breast cancers are treated by blocking oestrogens. Understanding on a molecular scale how this hormone works could help design better therapies, the authors suggest.

Shown in a study by researchers from CNIO and CABIMER published in 'Science Advances'.