Scientists at the University of South Australia have developed a new artificial intelligence-based method to uncover genetic interactions that contribute to cancer progression. The findings, published in Royal Society Open Science, suggest that groups of genes working together are responsible for driving tumour development, rather than individual mutated genes acting alone.
Dr. Andres Cifuentes-Bernal, the lead researcher on the project, explained that his team used AI tools to identify gene groups that collectively influence cancer advancement. Associate Professor Thuc Le, co-author of the study, emphasized the significance of this approach for understanding cancer biology. "Cancer is not static," he said. "It develops through a cascade of dynamic changes. Many genes act together to disrupt normal cell behaviour, but existing methods can struggle to detect that. Our approach is designed to capture that complexity."
The researchers tested their method using large breast cancer datasets and found that it could identify both well-known and previously hidden cancer-related genes. Some of these newly discovered genes were not mutated but still played a role in influencing other genes and contributing to tumour growth.
Their system successfully recognized many established cancer drivers listed in the Cancer Gene Census, an international reference source, which supports the accuracy of their approach. It also identified new candidate genes involved in cell signalling, immune response, and metastasis.
Assoc Prof Le noted that their technique focuses on detecting cooperative networks among genes instead of looking at isolated genetic factors. "These networks highlight how genes collaborate to collectively push cancer into more aggressive states," he said.
The team believes this AI-driven framework could help identify new therapeutic targets for patients whose tumours do not have common mutations typically associated with cancer progression. Dr Cifuentes-Bernal added: "Understanding these dynamics gives us a richer view of how tumours evolve. It moves us beyond thinking about single-cell mutations and towards a better understanding of the broader biological systems at play."
The researchers also suggested their method could be adapted for use in studying other diseases where gene regulation changes over time, such as neurodegenerative diseases, autoimmune disorders, and chronic inflammatory conditions.
'Identifying cooperative genes causing cancer progression with dynamic causal inference' is published in Royal Society Open Science (DOI: 10.1098/rsos.250442).
