Overview
Brendan Frey is chief innovation officer, founder, and board member of Deep Genomics, an internationally acclaimed entrepreneur, engineer, and scientist per the company team page. Frey made fundamental contributions in deep learning, genomic medicine, and information technology, co-authoring over 200 papers including more than a dozen in Nature, Science, and Cell.
His work with Geoffrey Hinton on the wake-sleep algorithm helped launch the field of deep learning (Science 1995) per Deep Genomics biography. Frey was a pioneer in developing AI systems that could accurately predict normal and pathological cell and genome biology.
Role and company
As Founder and Chief Innovation Officer, Frey guides Deep Genomics’ AI platform for predicting pathogenic mutations and identifying therapeutic targets while Brian O’Callaghan, MBA, serves as chief executive officer per the team page. Deep Genomics develops therapeutic candidates for patients with genetic disorders using AI foundation models integrated with experimental biology.
Frey co-founded Deep Genomics in 2015, creating the first AI system for predicting pathogenic mutations and identifying therapeutic targets per company biography — a founding claim tied to the company’s scientific identity.
Background and career context
Frey is Professor of Engineering and Medicine at the University of Toronto and Fellow of the American Association for the Advancement of Science, IEEE, Canadian Institute for Advanced Research, and Royal Society of Canada per Deep Genomics team biography. He co-founded the Vector Institute for Artificial Intelligence.
Frey received his Ph.D. from the University of Toronto in 1997 and has held researcher and professor positions at University of Illinois, Cambridge University, Isaac Newton Institute, and University of Waterloo per the cited biography.
Company-building or leadership context
Frey’s founding vision connected breakthrough AI biology predictions to therapeutic development pipelines — moving Deep Genomics from target discovery into drug candidates for genetic disorders. He advised Microsoft Research leadership on machine learning and served on its Technical Advisory Board per team biography.
Deep Genomics’ executive team includes CSO Greg Hoffman, PhD, and platform biology leaders generating proprietary training data for AI foundation models — reflecting Frey’s emphasis on closed-loop biology-AI integration.
Public milestones
2015: Deep Genomics founded by Brendan Frey per team biography. Frey’s prior deep learning contributions include wake-sleep algorithm (Science 1995) and factor graph work forming foundations of thousands of worldwide products including WiFi telecommunications standards per biography.
Frey remains Chief Innovation Officer and Founder on deepgenomics.com/team reviewed July 10, 2026.
Why their work matters
Deep Genomics under Frey’s scientific founding helped establish that genomic medicine could be computationally predicted before exhaustive wet-lab screening — a thesis now contested and refined across dozens of AI biotech competitors. Frey’s academic credentials and Vector Institute co-founding give Deep Genomics credibility in Toronto’s AI research cluster.
For healthcare AI readers, Frey represents the genomics-drug-discovery frontier distinct from clinical workflow AI but foundational to precision medicine narratives.
Sources and references
Primary sourcing: Deep Genomics team page Brendan Frey biography, reviewed July 10, 2026.
- Deep Genomics — Team (deepgenomics.com)
Academic appointments and publications count from official team biography only.
Frey’s co-authored wake-sleep algorithm paper (Science 1995) with Geoffrey Hinton and factor graph work cited on Deep Genomics team biography form foundations of thousands of worldwide products including WiFi telecommunications standards — scientific lineage connecting Frey’s academic contributions to Deep Genomics’ therapeutic AI platform under his Chief Innovation Officer founder role.
Deep Genomics team page lists CSO Greg Hoffman, PhD, and platform biology leaders generating proprietary training data for AI foundation models — experimental biology integration Frey’s founding vision requires to move from pathogenic mutation prediction to therapeutic candidates for genetic disorders.