AI-Generated Cancer Genomes to Support Precision Medicine

Synthetic data for real patients.

STARTER STATS

• Approximately 2 in 5 Canadians are expected to be diagnosed with cancer in their lifetime, with cancer claiming the lives of 1 in 4.

• Among men, prostate cancer is the most common, followed by lung and colorectal.

• Among women, breast cancer is the most common, followed by lung and colorectal cancer.

---

Scientists at the University of Toronto and the Ontario Institute for Cancer Research have developed an AI system that generates realistic synthetic cancer genomes. By training algorithms on more diverse and accurate data, they can detect cancer earlier and improve tailored cancer treatments.

Current analysis tools often rely on outdated training data that fails to capture the full biological diversity of cancer. While newer sequencing data exists, access is restricted due to patient confidentiality concerns. The realistic GenAI genomes help improve diagnostic algorithms without compromising patient privacy.

"With OncoGAN, we are creating realistic genomes out of nothing, with no connection to any real person, yet a huge amount of value scientifically. These synthetic genomes don't contain any personal health information, and so they can be shared without limitation."

— Dr. Lincoln Stein

The OncoGAN platform uses GenAI to simulate tumour genomes across eight cancer types, including breast, prostate, and pancreatic cancers.

Read more →