In a world first, Dr. Shai Rosenberg, Hadassah senior neuro-oncologist and head of the Laboratory for Computational Biology of Cancer at the Hadassah Cancer Research Institute (HCRI), has developed an AI-based algorithm that can identify all possible harmful point mutations of the TP53 gene, which is associated with half of all cancerous tumours, with an unprecedented 96.5% level of accuracy.
“Every tumour and every patient has a unique mutation profile,” Dr. Rosenberg explains. Using artificial intelligence and “big data,” he is creating additional models that will reveal which biological processes are active in individual tumours. This breakthrough will provide a platform for improved genetic screening and the development of personalized cancer treatment for carriers of these mutations.
Dr. Rosenberg, who led the international research collaboration, notes, “Identifying the true driver mutations in a patient’s tumour is a major challenge in precision oncology. Our new gene-specific machine learning model can predict the functional consequences of every possible point mutation in TP53. The model we created has an extremely high accuracy of 96.5 percent, which allows it to become a tool for clinical applications.”