New Computational Tool Reveals Patterns in Complex Biological Datasets

Researchers at EMBL-EBI, working with collaborators from Heidelberg University, have created CORNETO, a computational tool that applies machine learning to biological data to uncover how genes, proteins, and signaling pathways interact. By integrating prior biological knowledge with experimental data from diverse samples and conditions, CORNETO builds molecular networks that show connections within cells under various states. These insights aim to clarify the processes that contribute to health or disease. 

As biological datasets become increasingly large and complex, finding meaningful patterns within them presents a growing challenge. CORNETO, which stands for Constrained Optimisation for the Recovery of NETworks from Omics, addresses this issue by jointly analyzing multiple types of omics data—such as transcriptomics, proteomics, and metabolomics—alongside curated biological information.

“We wanted to solve a common challenge in systems biology: how to make sense of omics data when you have so much complex data available all at once,” said Julio Saez-Rodriguez, senior author of the paper published in Nature Machine Intelligence. “CORNETO helps by combining these complex data with prior information coming from biological databases to find patterns that are consistent, interpretable, and biologically meaningful.”

While traditional approaches often compare healthy and diseased cells separately, CORNETO analyzes multiple conditions at once. This broader approach helps identify shared and differing biological processes across datasets. According to first author Pablo Rodríguez-Mier, “Using CORNETO is like finding the common threads in a tangled web. It helps researchers pull out the key biological processes that are happening across many samples and understand what’s different or the same in each one.”

CORNETO has been used in cancer research to analyze gene expression across patients, identifying deregulated kinases using only transcriptomic data—results confirmed by phosphoproteomics. It also contributes to the DECIDER project, studying resistance to chemotherapy in ovarian cancer. In yeast, CORNETO identified core metabolic functions linked to survival, supporting its application in industrial biotechnology, such as developing strains for biofuel production.

CORNETO is available as open-source software on GitHub.