New Toronto Venture Set To Enter Genome Sequencing Market
Studies investigating the human genome have become a somewhat contentious issue over the past couple decades. However, many of these studies provide insights that can help us identify health issues and design effective treatments before conditions progress past the point of no return.

Brendan Frey, a professor in the University of Toronto’s department of computer and electrical engineering, has developed a method that combines artificial intelligence with big data genomics in an effort to create a clearer “roadmap for DNA-based therapy.”

Deep Genomics

Frey’s company, which he has aptly named Deep Genomics, will be entering a fast-growing market for diagnostics and personalized medicine that has its’ roots firmly based in whole genome sequencing. While researchers have already begun identifying genetic mutations that link to various diseases, Frey’s technology takes things one step further.

The technology that supports Deep Genomics’ entry into the genome sequencing market uses a carefully designed computer algorithm capable of “[teasing] out cause and effect relationships.” “My approach,” says Frey, the company’s president and CEO, “was let’s train a neural network to figure out why a mutation leads to a disease. That’s what makes our technology unique.”

Drawing on ‘Deep Learning’

Frey’s method builds upon a relatively new discipline within the field of computer science. This discipline is commonly known as ‘deep learning’, and it has recently led to a number of interesting developments in visual identification and speech recognition, two areas in which context plays a vital role in arriving at the correct answer.

The field has also used artificial intelligence extensively in the past. Deep learning takes computer programs capable of adjusting to become better and better at a complex task and further removes them from human guidance. In essence, this has led to programs that can make complicated associations, often beyond those of a human expert.

A New Application

Frey has taken the principles of deep learning and applied them to studies of the human genome. He has created a set of computer algorithms that observe patterns of mutation in an individual’s DNA. From those observations, the algorithm makes predictions as to “how those mutations collectively affect the operation of different types of cells in the body.”

Although the methods behind deep learning have been applied to a number of other fields, its’ potential uses in the medical field are just beginning to be explored. “By linking the DNA sequence to cellular function, [Frey’s] objective is to help determine not only what may be the source of a health problem but what treatment may be more likely to succeed [as well].”

While Frey had initially hoped his findings would quickly interest entrepreneurs in the biomedical sector, the identification of more tangible applications has since dampened his expectations ever so slightly. Nevertheless, according to Yann LeCun, the New York-based director of artificial intelligence at Facebook and an advisor to Frey’s new company, “The potential applications are really huge.”

With the help of initial angel investors, Frey’s company currently employs eight full-time staff members, but with LeCun’s help, and their newfound niche in genetic diagnostics, Frey anticipates doubling his staff by the year’s end.


Be sure to check out OnCorp Direct's service offering for Healthcare professionals
Share |
Posted by: DSR
Tuesday, September 1, 2015
Tag: Healthcare
Business Start Ups
Subscribe to the OnCorp Direct RSS feed and stay up to date with what's going on within the blog.
Address: 1033 Bay Street, Suite 313
Toronto, Ontario M5S 3A5 Canada
Phone: (416) 964-CORP (2677)
Toll Free: 1-800-461-7772
Fax: (416) 964-3975
To Comment on this blog, email:
OnCorp Direct Blog est. 2010 | Site Maintained By: