Big Questions, Six Billion Tiny Answers

Sequencing the genome is easier and cheaper than ever; but what does it all mean? The answers could profoundly change society.

 |  Transcript [PDF]

Behind every cell in your body is a copy of your personal DNA: your genome. Written in that genome is the code for life in six billion letters strung together in a unique combination of A’s, C’s, G’s, and T’s.

The technology is here today to sequence that genome, to read out that letter sequence, for about $1,000. And it’s getting cheaper all the time.

Brendan Frey, CIFAR senior fellow and professor of computer engineering at the University of Toronto, believes that in a few years genome sequencing will be so affordable that it will cost less than a trip to the grocery store to get this information.

But right now, we wouldn’t know what to do with it.

Frey calls this the phenotype-genotype gap: we can see a mutation in the text of a genome, but we don’t necessarily know how this will manifest in a patient, whether it will cause disease, or how to treat a health problem that might arise from it.

This is a big issue for modern healthcare. Over 60 percent of people will be diagnosed with a serious genetic disorder in their lifetime, says Frey. Chances are, we will all be personally touched by a diagnosis like this in our circles of family and friends.

In fact, Frey’s interest in genomics stems from his personal experience with this phenotype-genotype disconnect.

“Around 2002, my wife and I discovered that the baby she was carrying had a genetic problem,” says Frey. “We went and saw a counsellor and the counsellor told us that it could be nothing, or it could be a disaster. That was a very difficult time for me, but after that, I made a decision which was I didn’t want to work on trivial problems of detecting cats and YouTube videos anymore (using machine learning).

“I wanted to work on something that would profoundly change society. That’s what really changed my direction to focus on genomics, understanding the connections between mutations in your DNA and disease.”

Gathering more data will allow researchers to probe more deeply into health without even having to form questions to ask. In the past, studying biology has been hypothesis-driven, says Frey. The first step is to ask a question, which leads to a study to gather the required data, and then getting a yes or no answer.

By contrast, with the exponential growth in data available at hand, biology is shifting to an informatics-based approach. The data are all there at the start, and the question becomes how to comb through it.

“We have an enormous amount of data telling us what’s going on inside of your body, and the question is how to cultivate that data, combine that data and interpret that data to figure out what’s going wrong,” says Frey. “There’s a massive paradigm shift. Putting together the right resources to make this happen is very crucial.”

Frey credits networking and collaboration as a critical component for success. Through CIFAR, he is part of the Genetic Networks program and the Neuro-Computation program, both of which bring world-leading experts together and foster new interdisciplinary research. A better understanding of the genome and its role in health and disease is expected to bring down the cost of drug development while producing more effective medicines.

Distilling those six billion letters and closing the phenotype-genotype gap has the potential to be a game changer in human health.

‹ Previous post
Next post ›

In addition to founding Deep Genomics, Brendan Frey co-founded the Vector Institute for Artificial Intelligence and was a professor in engineering and medicine at the University of Toronto. He has made fundamental contributions to the fields of machine learning and genome biology, both in research and in industry. He led the team that developed a deep learning method for identifying the splicing-related genetic determinants of disease, which was published in the January 9, 2015 edition of Science Magazine. In the past twenty years, he has co-authored over 12 papers in Science, Nature and Cell, including one of the first papers on deep learning (Science, 1995). Frey is a co-inventor of the affinity propagation algorithm and of the factor graph notation for graphical models. He has consulted for over a dozen machine learning-powered companies, has served on the technical advisory board of Microsoft Research, holds seven patents, and has served as an expert witness in patent litigation. Frey’s former team members include entrepreneurs, industrial researchers, and professors at highly recognized centers in Canada, the United States, England and Europe.