Grant: Doctoral Research Award: Canada Graduate Scholarships

Cancer continues to be a leading cause of death globally, accounting for nearly 10 million deaths in 2020, or nearly one in six deaths. An estimated 2 in 5 Canadians will be diagnosed with cancer in their lifetime and about 1 in 4 will die of it. A projected 233,900 new cancer cases and 85,100 cancer deaths are expected in Canada in 2022. These numbers indicate that cancer remains a cause of high morbidity and mortality worldwide and the current treatments are still inadequate. An urgent need for new, selective, and more efficient drugs is evident. Peptides are a notable class of anticancer drugs that have shown promise in killing cancerous cells. Deep generative models are artificial intelligence (AI) algorithms capable of generating novel data with desired properties. Such algorithms have been well studied for generating images, texts, and even drug-like chemical compounds. However, little exploration has been done to study the effectiveness of these generative algorithms to generate novel peptide-based drugs. Furthermore, careful analysis of the potential of such AI algorithms to specifically generate anticancer peptides is even more critical. In this work, we offer to investigate peptide chemical space and develop a new AI method to sample novel peptides that have anti-cancer properties. To achieve this objective, we utilize prominent AI algorithms such as deep learning, reinforcement learning, and genetic algorithms. Further, using in-house peptide synthesis technologies, we experimentally validate peptides generated from our developed deep generative model against breast cancer and leukemia cells.