University of Navarra researchers have created RNACOREX, an open-source software tool that reconstructs gene regulatory networks from RNA data and applies them to cancer survival analysis. The team at the Institute of Data Science and Artificial Intelligence (DATAI) and the Cancer Center Clinica Universidad de Navarra validated the method with data from 13 tumor types compiled by The Cancer Genome Atlas (TCGA) consortium. The study, published in PLOS Computational Biology, shows that RNACOREX can process thousands of molecules at once to detect key interactions that conventional analyses often miss, generating molecular network maps that clarify the mechanisms driving tumor behavior.
Inside cells, microRNAs (miRNAs) and messenger RNAs (mRNAs) interact within complex regulatory networks, and disruptions in these connections can lead to cancer and other diseases. Ruben Armananzas, head of the Digital Medicine Laboratory at DATAI and one of the lead authors, explained that "understanding the architecture of these networks is crucial for detecting, studying, and classifying different tumor types.
However, reliably identifying these networks is a challenge due to the vast amount of available data, the presence of many false signals, and the lack of accessible and precise tools capable of distinguishing which molecular interactions are truly associated with each disease". RNACOREX tackles this challenge by combining curated information from international databases with real gene-expression data analysis to rank miRNA-mRNA interactions by their biological relevance. Using this information, the software derives increasingly complex regulatory networks that can also serve as probabilistic models.
To assess its performance, the researchers evaluated RNACOREX on 13 cancer types, including breast, colon, lung, stomach, melanoma, and head and neck tumors, using data from The Cancer Genome Atlas (TCGA) consortium.
"The software predicted patient survival with accuracy on par with sophisticated AI models, but with something many of those systems lack: clear, interpretable explanations of the molecular interactions behind the results", adds Aitor Oviedo-Madrid, a researcher at the Digital Medicine Laboratory of DATAI and first author of the study. RNACOREX not only identifies regulatory networks associated with clinical outcomes, but also uncovers molecular patterns shared across tumor types and highlights individual molecules of particular biomedical interest. These findings support new hypotheses about the mechanisms that regulate tumor growth and suggest possible diagnostic or therapeutic targets. "Our tool provides a reliable molecular 'map' that helps prioritize new biological targets, speeding up cancer research", he concludes.
RNACOREX is an open-source program available on GitHub and through the Python Package Index (PyPI), and it incorporates automated database downloads to streamline its use in laboratories and research centers. The project has been partially funded by the Government of Navarra through the ANDIA 2021 program and the ERA PerMed JTC2022 PORTRAIT initiative.
As artificial intelligence methods spread through genomics, Armananzas noted that "as artificial intelligence in genomics accelerates, RNACOREX positions itself as an explainable, easy-to-interpret solution and an alternative to 'black-box' models, helping bring omics data into biomedical practice". The University of Navarra team is adding features such as pathway analysis and additional interaction layers to build models that better capture the mechanisms that drive tumor growth and progression, with the aim of supporting precision oncology and connecting biomedicine, AI, and data science.
Research Report:RNACOREX – RNA coregulatory network explorer and classifier
