Brendan Frey / en ֱ's Deep Genomics inks partnership with U.S. biotech firm /news/u-t-s-deep-genomics-inks-partnership-us-biotech-firm <span class="field field--name-title field--type-string field--label-hidden">ֱ's Deep Genomics inks partnership with U.S. biotech firm</span> <div class="field field--name-field-featured-picture field--type-image field--label-hidden field__item"> <img loading="eager" srcset="/sites/default/files/styles/news_banner_370/public/UofT12852_20170320_BrendanFrey%28weblead%29.jpg?h=afdc3185&amp;itok=i_2mILhT 370w, /sites/default/files/styles/news_banner_740/public/UofT12852_20170320_BrendanFrey%28weblead%29.jpg?h=afdc3185&amp;itok=Np_jEUmz 740w, /sites/default/files/styles/news_banner_1110/public/UofT12852_20170320_BrendanFrey%28weblead%29.jpg?h=afdc3185&amp;itok=1A6N-WR4 1110w" sizes="(min-width:1200px) 1110px, (max-width: 1199px) 80vw, (max-width: 767px) 90vw, (max-width: 575px) 95vw" width="740" height="494" src="/sites/default/files/styles/news_banner_370/public/UofT12852_20170320_BrendanFrey%28weblead%29.jpg?h=afdc3185&amp;itok=i_2mILhT" alt="Brendan Frey"> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span>Christopher.Sorensen</span></span> <span class="field field--name-created field--type-created field--label-hidden"><time datetime="2018-04-11T17:18:49-04:00" title="Wednesday, April 11, 2018 - 17:18" class="datetime">Wed, 04/11/2018 - 17:18</time> </span> <div class="clearfix text-formatted field field--name-field-cutline-long field--type-text-long field--label-above"> <div class="field__label">Cutline</div> <div class="field__item">Professor Brendan Frey co-founded Deep Genomics in 2015. The startup uses AI to find treatments for genetic illnesses (photo by Johnny Guatto)</div> </div> <div class="field field--name-field-author-reporters field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/authors-reporters/chris-sorensen" hreflang="en">Chris Sorensen</a></div> </div> <div class="field field--name-field-topic field--type-entity-reference field--label-above"> <div class="field__label">Topic</div> <div class="field__item"><a href="/news/topics/global-lens" hreflang="en">Global Lens</a></div> </div> <div class="field field--name-field-story-tags field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/tags/artificial-intelligence" hreflang="en">Artificial Intelligence</a></div> <div class="field__item"><a href="/news/tags/brendan-frey" hreflang="en">Brendan Frey</a></div> <div class="field__item"><a href="/news/tags/faculty-applied-science-engineering" hreflang="en">Faculty of Applied Science &amp; Engineering</a></div> <div class="field__item"><a href="/news/tags/international-partnerships" hreflang="en">International partnerships</a></div> <div class="field__item"><a href="/news/tags/jlabs" hreflang="en">JLabs</a></div> </div> <div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>A University of Toronto startup that uses artificial intelligence to develop treatments for genetic diseases has forged a partnership with a U.S. biotechnology firm.</p> <p>Deep Genomics, spun out of research done by Professor&nbsp;<strong>Brendan Frey</strong>&nbsp;in the Faculty of Applied Science &amp; Engineering, said this week it will be working with Cambridge, Mass.-based, Wave Life Sciences to identify novel therapies for neuromuscular disorders – a category of illness that includes diseases like Duchenne muscular dystrophy.</p> <p>Frey, Deep Genomics’ co-founder and CEO, said Wave’s neuromuscular research is tailor-made for the startup’s machine learning technology and targets a group of disorders for which treatments are badly needed.</p> <p>“Neuromuscular diseases are a tragic category of diseases for children that impacts them at a very young age, and impacts their lives severely and can be fatal,” he said, adding Deep Genomics and Wave have yet to disclose the particular neuromuscular diseases they intend to focus on.&nbsp;</p> <p>“But there’s a significant portion of these disorders that have a clear genetic basis, meaning there are mutations in their DNA that they’ve inherited at birth that cause something to go wrong.&nbsp;</p> <p>“That makes it a perfect match for our platform, which is all about genetics – RNA, DNA and linking the genetics to what’s going to go wrong inside the cell.”</p> <p>Founded in 2015, Deep Genomics has built an “artificial intelligence-powered discovery platform” that&nbsp;<a href="/news/u-t-s-deep-genomics-applies-ai-accelerate-drug-development-genetic-conditions">combines machine learning with genomics research to develop genetic medicines that can potentially treat a myriad of genetic illnesses, from autism to cancer</a>.&nbsp;</p> <p>The startup, which raised US$13 million in financing last year, received support from ֱ's Innovations &amp; Partnerships Office, as well as UTEST and the Creative Destruction Lab at the Rotman School of Management – all part of the expansive ֱ entrepreneurship ecosystem. It is now working with Johnson &amp; Johnson’s JLABS life sciences incubator in Toronto.</p> <p>Wave, meanwhile, is a publicly listed company that uses a chemistry-based platform to find “transformational therapies for patients with serious, genetically-defined diseases.”&nbsp;</p> <p>While Deep Genomics’ main research focus continues to be on metabolic and neurodegenerative disorders – it’s investing $10 million to develop its pre-clinical platform in these areas – Frey said working with Wave on neuromuscular diseases offers a number of additional benefits. They include “a way to rapidly deploy the output of our platform within a drug development context,” and the opportunity to further fine-tune Deep Genomics’ machine learning technologies.</p> <p>Frey said Deep Genomics has held talks with most of the world’s major pharmaceutical companies, as well as several smaller ones, but decided to partner with Wave because it’s a young, research-focused firm that shares his startup’s vision.&nbsp;</p> <p>And what is that outlook? If the traditional approach to medical research is mostly trial and error at the lab bench, Frey believes machine learning can be used to crunch through billions of data points, including those created by the sequencing of the human genome, to determine not only the cause of a particular genetic illness, but to find a cure – all “in silico,” or on the computer.</p> <p>“The AI systems that we’ve built aren’t meant to speculatively offer up potential solutions,” Frey says.&nbsp;</p> <p>“They’re highly precise and very intentional.”&nbsp;</p> <h3><a href="http://entrepreneurs.utoronto.ca/">Learn more about ֱ Entrepreneurship</a></h3> <h3>&nbsp;</h3> <p>&nbsp;</p> </div> <div class="field field--name-field-news-home-page-banner field--type-boolean field--label-above"> <div class="field__label">News home page banner</div> <div class="field__item">Off</div> </div> Wed, 11 Apr 2018 21:18:49 +0000 Christopher.Sorensen 133211 at Six ֱ researchers join Vector Institute /news/six-u-t-researchers-join-vector-institute <span class="field field--name-title field--type-string field--label-hidden">Six ֱ researchers join Vector Institute</span> <div class="field field--name-field-featured-picture field--type-image field--label-hidden field__item"> <img loading="eager" srcset="/sites/default/files/styles/news_banner_370/public/2017-09-26-vector-logo-resized_0.jpg?h=afdc3185&amp;itok=_PZ-ZvK4 370w, /sites/default/files/styles/news_banner_740/public/2017-09-26-vector-logo-resized_0.jpg?h=afdc3185&amp;itok=Voh7uoRG 740w, /sites/default/files/styles/news_banner_1110/public/2017-09-26-vector-logo-resized_0.jpg?h=afdc3185&amp;itok=hgN5nSlM 1110w" sizes="(min-width:1200px) 1110px, (max-width: 1199px) 80vw, (max-width: 767px) 90vw, (max-width: 575px) 95vw" width="740" height="494" src="/sites/default/files/styles/news_banner_370/public/2017-09-26-vector-logo-resized_0.jpg?h=afdc3185&amp;itok=_PZ-ZvK4" alt="Vector Institute logo"> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span>rasbachn</span></span> <span class="field field--name-created field--type-created field--label-hidden"><time datetime="2017-10-31T11:16:00-04:00" title="Tuesday, October 31, 2017 - 11:16" class="datetime">Tue, 10/31/2017 - 11:16</time> </span> <div class="field field--name-field-author-reporters field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/authors-reporters/jennifer-robinson" hreflang="en">Jennifer Robinson</a></div> </div> <div class="field field--name-field-topic field--type-entity-reference field--label-above"> <div class="field__label">Topic</div> <div class="field__item"><a href="/news/topics/our-community" hreflang="en">Our Community</a></div> </div> <div class="field field--name-field-story-tags field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/tags/artificial-intelligence" hreflang="en">Artificial Intelligence</a></div> <div class="field__item"><a href="/news/tags/brendan-frey" hreflang="en">Brendan Frey</a></div> <div class="field__item"><a href="/news/tags/computer-science" hreflang="en">Computer Science</a></div> <div class="field__item"><a href="/news/tags/faculty-applied-science-engineering" hreflang="en">Faculty of Applied Science &amp; Engineering</a></div> <div class="field__item"><a href="/news/tags/faculty-medicine" hreflang="en">Faculty of Medicine</a></div> <div class="field__item"><a href="/news/tags/geoffrey-hinton" hreflang="en">Geoffrey Hinton</a></div> <div class="field__item"><a href="/news/tags/raquel-urtasun" hreflang="en">Raquel Urtasun</a></div> <div class="field__item"><a href="/news/tags/university-toronto-scarborough" hreflang="en">University of Toronto Scarborough</a></div> <div class="field__item"><a href="/news/tags/vector-institute" hreflang="en">Vector Institute</a></div> </div> <div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>Six researchers at the University of Toronto with expertise in areas such as machine learning, medicine, engineering and statistics are joining the Vector Institute.</p> <p>They’re joining a highly accomplished team of world-class researchers who helped launch the Vector Institute in March, including ֱ's&nbsp;<a href="http://www.provost.utoronto.ca/awards/uprofessors.htm">University Professor</a> Emeritus <strong>Geoffrey Hinton</strong>; Associate Professor and leader of Uber’s Advanced Technologies Group <strong>Raquel Urtasun; </strong>and Deep Genomics founder and electrical and computing engineering Professor <strong>Brendan Frey</strong>.</p> <p>“My top priority since Vector’s launch has and continues to be to build a collaborative, talented team,” said one of the original Vector Institute co-founders, <strong>Richard Zemel</strong>, a ֱ computer science professor and the research director for Vector. “I am thrilled that these individuals will join a diverse and growing team of faculty committed to making Vector a global leader in AI.”</p> <p>The new members are:</p> <ul> <li><strong>David J. Fleet</strong>, ֱ professor of computer science,&nbsp;department of mathematical sciences at the University of Toronto Scarborough</li> <li><strong>Anna Goldenberg, </strong>ֱ assistant professor of computer science (computational biology group) and scientist in the Genetics and Genome Biology Lab at SickKids Research Institute</li> <li><strong>Frank Rudzicz, </strong>ֱ assistant professor of computer science (status) and scientist at the Toronto Rehabilitation Institute-UHN</li> <li><strong>Jimmy Ba</strong>, assistant professor of ֱ computer science, joining ֱ in fall of 2018</li> <li><strong>Murat Erdogdu</strong>, joint appointment as ֱ assistant professor of&nbsp; computer science and statistical sciences, joining ֱ in fall of 2018</li> <li><strong>Marzyeh Ghassemi, </strong>joint appointment as ֱ assistant professor of computer science and medicine, joining ֱ in fall of 2018</li> </ul> <p><img alt class="media-image attr__typeof__foaf:Image img__fid__6587 img__view_mode__media_large attr__format__media_large" src="/sites/default/files/2017-10-31-Erdogdu-resized_0.jpg" style="width: 350px; height: 386px; margin: 10px; float: left;" typeof="foaf:Image">Originally from Turkey, Erdogdu (left) is coming to ֱ from Microsoft Research New England, where he is a postdoctoral researcher. He has a PhD in statistics and a master’s degree in computer science from Stanford University, as well as bachelor degrees in electrical engineering and mathematics from Boğaziҫi University.</p> <p>His specialty is the design of optimization algorithms for machine learning models, such as deep learning models and recommender systems. By using more efficient algorithms, model training times can be significantly reduced from weeks (depending on the size of the datasets) to just hours, enabling researchers to efficiently test and select the best model for the problem at hand.</p> <p>“I’m really looking forward to designing efficient algorithms for real-world problems,” Erdogdu said, adding he’s an avid hiker and camper and is looking forward to exploring the city and surrounding countryside when he’s not in the lab.</p> <p>“When I visited Canada, I found it extremely friendly and it was really diverse,” he said.</p> <p>Invited to speak on campus last March, he said he was blown away by the “energetic” and collaborative nature of the computer science department and all of the activity on the downtown Toronto campus. He was also invited to attend the official launch of the Vector Institute.</p> <p>“The University of Toronto is one of the best universities in my area of research, which is artificial intelligence, machine learning and optimization,” he continued. “It will be a pleasure to be around experts like Geoff Hinton and <strong>Radford Neal</strong> – I think it’s safe to say they changed the world. There are also lots of young faculty that I’d be very excited to work with and collaborate with.</p> <p>“And that’s why I chose the University of Toronto and Vector.”</p> <p><img alt class="media-image attr__typeof__foaf:Image img__fid__6585 img__view_mode__media_large attr__format__media_large" src="/sites/default/files/2017-10-31-Jimmy_Ba-resized.jpg" style="width: 340px; height: 453px; margin: 10px; float: left;" typeof="foaf:Image">Ba (left), whose research focuses on the development of learning algorithms for deep neural networks, is returning to ֱ – this time in computer science – after completing his undergraduate, master’s and PhD here in the Edward S. Rogers Sr. Department of Electrical &amp; Computer Engineering.</p> <p>He was supervised at ֱ by some leading researchers in machine learning, including Hinton, Frey and Ruslan Salakhutdinov, who is now a computer science professor at Carnegie Mellon University.</p> <p>Among his many accomplishments, Ba developed the Adam Optimizer, one of the go-to&nbsp;algorithms to train deep learning models. He was also one of the first students from a Canadian institution to win a Facebook PhD Fellowship, and in 2015 his team achieved the highest place among academic labs in the image caption generation competition at the Conference on Computer Vision and Pattern Recognition. He is currently a postdoctoral researcher at MIT.</p> <p><img alt class="media-image attr__typeof__foaf:Image img__fid__6586 img__view_mode__media_large attr__format__media_large" src="/sites/default/files/2017-10-31-Marzyeh_Ghassemi-resized.jpg" style="width: 350px; height: 391px; margin: 10px; float: left;" typeof="foaf:Image">Ghassemi (left), a&nbsp;recent PhD graduate from MIT,&nbsp; is a visiting researcher with Alphabet’s Verily and a part-time postdoc at MIT. Her hiring at ֱ signals a new partnership between computer science and medicine: She is the first joint hire in computational medicine.</p> <p>“The University of Toronto computer science department has established experts in machine learning,” said Ghassemi, “as well as really exciting junior faculty doing work in important areas of machine learning like Raquel [Urtasun] and driverless vehicles."</p> <p>She has focused her research on machine learning applications in health care. This means using machine learning as a tool in her work exploring clinical data with algorithms to “predict interesting and important human risks” and anticipate patient needs and decrease mortality rates, she explained.</p> <p>For example, clinical data can be mined to determine what kind of patients in the intensive care unit will require a ventilator or a blood cell transfusion. It can also predict a patient's length of stay and their risk of death within a year of leaving care.</p> <p>Ghassemi is also interested in using machine learning in non-invasive patient monitoring. She’s worked on detecting voice disorders that impact a subject's ability to speak using an electromechanical device called an accelerometer, coupled with a machine learning algorithm.</p> <p>Instead of placing a camera in a patient’s throat that detects polyps and other damage to strained vocal chords, the accelerometer is simply taped to their neck.</p> <p>“We were able to non-invasively detect which patients had nodules, which was very exciting,” she said. Her clinical collaborators at the Massachusetts&nbsp;General Hospital Voice Center are currently exploring the use of a cellphone prompting system based on findings&nbsp;of the non-invasive accelerometer. When subjects are in danger of straining their chords, they are sent an alert.</p> <p>In addition to its machine learning expertise, the ability to work with “world-class clinical collaborators” at ֱ’s nine partner hospitals was another important draw for Ghassemi.</p> <p>“All of the people I met at the Faculty of Medicine were excited to work on clinically meaningful problems,” she said. “They really wanted to understand how we could apply machine learning techniques and develop new algorithms to be useful in a clinical setting. That combination of a fantastic technical school and a collaborative, world-class clinical environment is pretty rare.”</p> <p align="center">&nbsp;</p> </div> <div class="field field--name-field-news-home-page-banner field--type-boolean field--label-above"> <div class="field__label">News home page banner</div> <div class="field__item">Off</div> </div> Tue, 31 Oct 2017 15:16:00 +0000 rasbachn 120492 at ֱ's Deep Genomics raises US$13 million to fund expansion /news/u-t-s-deep-genomics-raises-us13-million-fund-expansion <span class="field field--name-title field--type-string field--label-hidden">ֱ's Deep Genomics raises US$13 million to fund expansion</span> <div class="field field--name-field-featured-picture field--type-image field--label-hidden field__item"> <img loading="eager" srcset="/sites/default/files/styles/news_banner_370/public/UofT12854_20170320_BrendanFrey-%28web-lead%29.jpg?h=afdc3185&amp;itok=ETOEk9MG 370w, /sites/default/files/styles/news_banner_740/public/UofT12854_20170320_BrendanFrey-%28web-lead%29.jpg?h=afdc3185&amp;itok=8anVspQg 740w, /sites/default/files/styles/news_banner_1110/public/UofT12854_20170320_BrendanFrey-%28web-lead%29.jpg?h=afdc3185&amp;itok=wa4CXXjW 1110w" sizes="(min-width:1200px) 1110px, (max-width: 1199px) 80vw, (max-width: 767px) 90vw, (max-width: 575px) 95vw" width="740" height="494" src="/sites/default/files/styles/news_banner_370/public/UofT12854_20170320_BrendanFrey-%28web-lead%29.jpg?h=afdc3185&amp;itok=ETOEk9MG" alt> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span>Christopher.Sorensen</span></span> <span class="field field--name-created field--type-created field--label-hidden"><time datetime="2017-09-26T11:17:47-04:00" title="Tuesday, September 26, 2017 - 11:17" class="datetime">Tue, 09/26/2017 - 11:17</time> </span> <div class="clearfix text-formatted field field--name-field-cutline-long field--type-text-long field--label-above"> <div class="field__label">Cutline</div> <div class="field__item">Brendan Frey, a professor in ֱ's Faculty of Applied Science &amp; Engineering, married machine learning and genomic science to develop genetic medicines (photo by Johnny Guatto)</div> </div> <div class="field field--name-field-author-reporters field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/authors-reporters/chris-sorensen" hreflang="en">Chris Sorensen</a></div> </div> <div class="field field--name-field-author-legacy field--type-string field--label-above"> <div class="field__label">Author legacy</div> <div class="field__item">Chris Sorensen</div> </div> <div class="field field--name-field-topic field--type-entity-reference field--label-above"> <div class="field__label">Topic</div> <div class="field__item"><a href="/news/topics/our-community" hreflang="en">Our Community</a></div> </div> <div class="field field--name-field-story-tags field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/tags/artificial-intelligence" hreflang="en">Artificial Intelligence</a></div> <div class="field__item"><a href="/news/tags/brendan-frey" hreflang="en">Brendan Frey</a></div> <div class="field__item"><a href="/news/tags/faculty-applied-science-engineering" hreflang="en">Faculty of Applied Science &amp; Engineering</a></div> <div class="field__item"><a href="/news/tags/thisistheplace" hreflang="en">ThisIsThePlace</a></div> </div> <div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>Deep Genomics, an artificial intelligence-powered health startup co-founded by University of Toronto's<strong>&nbsp;Brendan Frey</strong>, <a href="https://www.deepgenomics.com/series-a">has raised</a> US$13 million from a Silicon Valley venture capital firm.</p> <p>The startup, launched in 2015, combines artificial intelligence, or AI, and genomics research to help develop genetic medicines to treat a myriad&nbsp;of disorders – everything from autism to cancer.&nbsp;</p> <p>Frey, a professor in ֱ's Faculty of Applied Science &amp; Engineering, told the<em> Globe and Mail</em> the cash raised through an equity investment by Khosla Ventures of Menlo Park, Calif. will be used to hire a range of scientific experts, doubling the firm's current 20-person staff.</p> <p>“We are aiming to flip medicine on its back and do something completely different,” said Frey, who is also Deep Genomics's CEO.</p> <h3><a href="https://beta.theglobeandmail.com/report-on-business/toronto-ai-startup-deep-genomics-lands-financing/article36383250/?ref=http://www.theglobeandmail.com&amp;">Read more about Deep Genomics in the <em>Globe and Mail</em></a></h3> <p>The company’s <a href="/news/u-t-s-deep-genomics-applies-ai-accelerate-drug-development-genetic-conditions">platform </a>is used to search through 69 billion molecules to generate 1,000 compounds that can be used to manipulate cell biology and develop new therapies. The hope is the technology will eventually help drug companies create new pharmaceuticals to treat genetic ailments.</p> <p>Frey,&nbsp;a founding member of the&nbsp;<a href="/news/toronto-s-vector-institute-officially-launched">Vector Institute for artificial intelligence research</a>, said in a statement that Deep Genomics is “a company whose founding principle is that the future of medicine will rely on artificial intelligence because biology is too complex for humans to understand.”</p> <p>Deep Genomics previously raised $3.7 million in a seed round in 2015. It is&nbsp;currently a resident of Johnson &amp; Johnson’s <a href="/news/jlabs-startup-incubator-selects-university-toronto-mars-first-international-expansion">JLABS life science incubator in Toronto</a>, which was launched last year in partnership with ֱ, MaRS and pharmaceutical company Janssen.&nbsp;</p> <p>The startup also recieved support from ֱ's Innovations &amp; Partnerships Office, as well as UTEST and the&nbsp;Creative Destruction Lab&nbsp;at the Rotman School of Management – all part of the expansive ֱ entrepreneurship ecosystem.&nbsp;</p> <p>“We believe that the technology developed by Deep Genomics puts them in a unique position to identify new therapies,” said Khosla founding partner Vinod Khosla in a statement.&nbsp;</p> <p>“Because of the quality of their science and engineering team and the deep integration of their AI technology into their preclinical drug development pipeline, we are confident a very large potential exists here.”</p> <h3><a href="http://entrepreneurs.utoronto.ca/">Learn more about Entrepreneurship at ֱ</a></h3> </div> <div class="field field--name-field-news-home-page-banner field--type-boolean field--label-above"> <div class="field__label">News home page banner</div> <div class="field__item">Off</div> </div> Tue, 26 Sep 2017 15:17:47 +0000 Christopher.Sorensen 117220 at The AI revolution spreads its wings in Toronto, and at ֱ /news/ai-revolution-spreads-its-wings-toronto-and-u-t <span class="field field--name-title field--type-string field--label-hidden">The AI revolution spreads its wings in Toronto, and at ֱ</span> <div class="field field--name-field-featured-picture field--type-image field--label-hidden field__item"> <img loading="eager" srcset="/sites/default/files/styles/news_banner_370/public/UofT3395_20130312_GeoffreyHinton%20%28web%20lead%29.jpg?h=afdc3185&amp;itok=n8Y5M_J4 370w, /sites/default/files/styles/news_banner_740/public/UofT3395_20130312_GeoffreyHinton%20%28web%20lead%29.jpg?h=afdc3185&amp;itok=wq7ujwxY 740w, /sites/default/files/styles/news_banner_1110/public/UofT3395_20130312_GeoffreyHinton%20%28web%20lead%29.jpg?h=afdc3185&amp;itok=jjBpAKTN 1110w" sizes="(min-width:1200px) 1110px, (max-width: 1199px) 80vw, (max-width: 767px) 90vw, (max-width: 575px) 95vw" width="740" height="494" src="/sites/default/files/styles/news_banner_370/public/UofT3395_20130312_GeoffreyHinton%20%28web%20lead%29.jpg?h=afdc3185&amp;itok=n8Y5M_J4" alt="Geoffrey Hinton"> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span>Christopher.Sorensen</span></span> <span class="field field--name-created field--type-created field--label-hidden"><time datetime="2017-08-31T15:13:39-04:00" title="Thursday, August 31, 2017 - 15:13" class="datetime">Thu, 08/31/2017 - 15:13</time> </span> <div class="clearfix text-formatted field field--name-field-cutline-long field--type-text-long field--label-above"> <div class="field__label">Cutline</div> <div class="field__item">ֱ's Geoffrey Hinton is optimistic about the future of deep learning, an AI field he helped pioneer (photo by Johnny Guatto)</div> </div> <div class="field field--name-field-author-reporters field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/authors-reporters/chris-sorensen" hreflang="en">Chris Sorensen</a></div> </div> <div class="field field--name-field-author-legacy field--type-string field--label-above"> <div class="field__label">Author legacy</div> <div class="field__item">Chris Sorensen</div> </div> <div class="field field--name-field-topic field--type-entity-reference field--label-above"> <div class="field__label">Topic</div> <div class="field__item"><a href="/news/topics/global-lens" hreflang="en">Global Lens</a></div> </div> <div class="field field--name-field-story-tags field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/tags/computer-science" hreflang="en">Computer Science</a></div> <div class="field__item"><a href="/news/tags/artificial-intelligence" hreflang="en">Artificial Intelligence</a></div> <div class="field__item"><a href="/news/tags/faculty-applied-science-engineering" hreflang="en">Faculty of Applied Science &amp; Engineering</a></div> <div class="field__item"><a href="/news/tags/geoffrey-hinton" hreflang="en">Geoffrey Hinton</a></div> <div class="field__item"><a href="/news/tags/brendan-frey" hreflang="en">Brendan Frey</a></div> </div> <div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p><strong>Geoffrey Hinton</strong> may feel a bit sheepish about his “godfather” of deep learning moniker, but there’s little doubt the <a href="http://www.provost.utoronto.ca/awards/uprofessors.htm">University Professor</a> Emeritus at the University of Toronto <a href="/news/u-t-geoffrey-hinton-ai-will-eventually-surpass-human-brain-getting-jokes-could-take-time">helped spark a revolution in artificial intelligence</a>, or AI, that could eventually touch everyone on the planet.&nbsp;</p> <p>“I feel slightly embarrassed by being called the godfather,” Hinton admits in a recent profile published by London’s <em>The Daily Telegraph</em>, the first in a three-part series on the explosion of AI research and startups in Toronto.</p> <h3><a href="http://www.telegraph.co.uk/technology/2017/08/26/godfather-ai-making-machines-clever-whether-robots-really-will/">Read <em>The Daily Telegraph</em> story on Hinton</a></h3> <p>Hinton’s early life – raised in Great Britain, a brief stint as a carpenter, then into academia with a focus on AI – is recounted. So are the years toiling in relative obscurity as he worked on a machine learning approach at ֱ that mimics the way human toddlers learn. &nbsp;</p> <p>Hinton, who is an engineering&nbsp;fellow at Google and the chief scientific adviser at the <a href="/news/toronto-s-vector-institute-officially-launched">newly created Vector Institute</a>, also voices optimism about the impact AI will have on fields like medicine and why the automation of the economy shouldn’t be feared.</p> <p>“In a sensibly organized society, if you improve productivity, there is room for everybody to benefit,” he says in the article.&nbsp;“The problem isn’t the technology, but the way the benefits are shared out.”</p> <p>That’s to say nothing of the benefits provided by AI technologies themselves. In the series’ second instalment, the paper details how <strong>Frank Rudzicz</strong>, a scientist at the Toronto Rehabilitation Institute and an assistant professor at&nbsp;ֱ’s department of computer science, and his computer science colleague&nbsp;<strong>Liam Kaufman</strong> are using AI algorithms to examine human speech to better track the health of patients suffering from neurological disorders.&nbsp;</p> <p>Their startup, <a href="/news/natural-language-processing-dementia">WinterLight Labs</a>, was co-founded with <strong>Katie Fraser</strong>, who recently received a PhD from the Faculty of Arts &amp;&nbsp;Science, and <strong>Maria Yancheva</strong>, who completed a master’s degree&nbsp;in computer science at ֱ.</p> <p>Rudzicz also talks about the robot he built – named Ludwig – that uses machine learning algorithms to talk to patients and assess their mental health.</p> <h3><a href="http://www.telegraph.co.uk/technology/2017/08/28/inside-ai-healthcare-revolution-meeting-robots-can-detect-alzheimers/">Read <em>The Daily Telegraph</em>&nbsp;story on WinterLight Labs and Deep Genomics</a></h3> <p><img alt class="media-image attr__typeof__foaf:Image img__fid__5844 img__view_mode__media_original attr__format__media_original" height="500" src="/sites/default/files/UofT12853_20170320_BrendanFrey%20%28web%20embed%29.jpg" typeof="foaf:Image" width="750" loading="lazy"></p> <p><em>Brendan Frey of ֱ's department of electrical &amp; computer engineering, is a co-founder and CEO of Deep Genomics (photo by Johnny Guatto)</em></p> <p>The paper also talks to&nbsp;<strong>Brendan Frey&nbsp;</strong>about his&nbsp;startup Deep Genomics, which uses machine learning techniques to help develop drugs&nbsp;that target the root causes of diseases and disorders, from cystic fibrosis to cancer.&nbsp;</p> <p>“The basic fact is no human or group of humans will ever understand how the genome works,” says Frey, an engineering professor at ֱ with cross-appointments in the department of computer science and the Donnelly Centre for Cellular and Biomolecular Research. He is also a co-founder of the Vector Institute.</p> <p>“We have an exponentially growing set of data to allow us to peer into cells and read out what is changing,” he says. “There is only one solution: artificial intelligence. It’s the best technology we have in our systems to understand complex data.”&nbsp;</p> </div> <div class="field field--name-field-news-home-page-banner field--type-boolean field--label-above"> <div class="field__label">News home page banner</div> <div class="field__item">Off</div> </div> Thu, 31 Aug 2017 19:13:39 +0000 Christopher.Sorensen 114106 at ֱ's Deep Genomics applies AI to accelerate drug development for genetic conditions /news/u-t-s-deep-genomics-applies-ai-accelerate-drug-development-genetic-conditions <span class="field field--name-title field--type-string field--label-hidden">ֱ's Deep Genomics applies AI to accelerate drug development for genetic conditions</span> <div class="field field--name-field-featured-picture field--type-image field--label-hidden field__item"> <img loading="eager" srcset="/sites/default/files/styles/news_banner_370/public/2017-05-02-brendan-frey.jpg?h=afdc3185&amp;itok=wxZvw5DC 370w, /sites/default/files/styles/news_banner_740/public/2017-05-02-brendan-frey.jpg?h=afdc3185&amp;itok=uk1xPfaY 740w, /sites/default/files/styles/news_banner_1110/public/2017-05-02-brendan-frey.jpg?h=afdc3185&amp;itok=TjNY6ZnI 1110w" sizes="(min-width:1200px) 1110px, (max-width: 1199px) 80vw, (max-width: 767px) 90vw, (max-width: 575px) 95vw" width="740" height="494" src="/sites/default/files/styles/news_banner_370/public/2017-05-02-brendan-frey.jpg?h=afdc3185&amp;itok=wxZvw5DC" alt="Brendan Frey"> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span>ullahnor</span></span> <span class="field field--name-created field--type-created field--label-hidden"><time datetime="2017-05-03T17:10:58-04:00" title="Wednesday, May 3, 2017 - 17:10" class="datetime">Wed, 05/03/2017 - 17:10</time> </span> <div class="clearfix text-formatted field field--name-field-cutline-long field--type-text-long field--label-above"> <div class="field__label">Cutline</div> <div class="field__item">ֱ Engineering Professor Brendan Frey is the founder and CEO of Deep Genomics, a startup company applying deep-learning techniques to revolutionize genomic medicine (photo courtesy of Deep Genomics) </div> </div> <div class="field field--name-field-author-reporters field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/authors-reporters/marit-mitchell" hreflang="en">Marit Mitchell</a></div> </div> <div class="field field--name-field-author-legacy field--type-string field--label-above"> <div class="field__label">Author legacy</div> <div class="field__item">Marit Mitchell</div> </div> <div class="field field--name-field-topic field--type-entity-reference field--label-above"> <div class="field__label">Topic</div> <div class="field__item"><a href="/news/topics/global-lens" hreflang="en">Global Lens</a></div> </div> <div class="field field--name-field-story-tags field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/tags/artificial-intelligence" hreflang="en">Artificial Intelligence</a></div> <div class="field__item"><a href="/news/tags/vector" hreflang="en">Vector</a></div> <div class="field__item"><a href="/news/tags/brendan-frey" hreflang="en">Brendan Frey</a></div> <div class="field__item"><a href="/news/tags/entrepreneurship" hreflang="en">Entrepreneurship</a></div> <div class="field__item"><a href="/news/tags/faculty-applied-science-engineering" hreflang="en">Faculty of Applied Science &amp; Engineering</a></div> <div class="field__item"><a href="/news/tags/startup" hreflang="en">Startup</a></div> <div class="field__item"><a href="/news/tags/drugs" hreflang="en">Drugs</a></div> <div class="field__item"><a href="/news/tags/faculty-arts-science" hreflang="en">Faculty of Arts &amp; Science</a></div> <div class="field__item"><a href="/news/tags/donnelly" hreflang="en">Donnelly</a></div> <div class="field__item"><a href="/news/tags/faculty-medicine" hreflang="en">Faculty of Medicine</a></div> </div> <div class="field field--name-field-subheadline field--type-string-long field--label-above"> <div class="field__label">Subheadline</div> <div class="field__item">ֱ spinoff company combines leading research in both machine learning and genomic science to accelerate development of highly tailored medical treatments</div> </div> <div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>Genetic mutations are the cause of countless diseases and disorders, from cancer to autism to cystic fibrosis.</p> <p>Now, startup company <a href="https://www.deepgenomics.com/">Deep Genomics</a> is applying decades of research into machine learning and genomic science to develop genetic medicines –&nbsp;accelerating treatments that address the root causes of these conditions.</p> <p>“If you have smoke billowing out of the tailpipe of your car, you don’t just put a filter on the tailpipe –&nbsp;you have to look under the hood and address the original problem,” says <strong>Brendan Frey</strong>, the co-founder and CEO of Deep Genomics, and a ֱ engineering professor with cross-appointments in the department of computer science and the Donnelly Centre for Cellular and Biomolecular Research. “That’s what we’re doing: applying our platform for the discovery-phase development of medicines that address genetic problems.”</p> <p>Developing new drugs is expensive, slow and inefficient –&nbsp;when researchers identify a protein involved in a disease, pharmaceutical companies often use a ‘guess-and-test’ approach to see whether any of the known drug molecules in their arsenal is&nbsp;a match to the protein’s unique shape. Often, thousands of molecules need to be screened in order to generate a match.</p> <p>Frey’s team at Deep Genomics is looking at the first biological step in the process: at the genes that contain the blueprints for proteins and instructions on how and when to produce them.</p> <p>“There are many ways a protein could be causing a problem, resulting from different changes to the genome. We can see those changes at the level of individual genes,” says Frey. “Instead of focusing on proteins, we’re focusing on the genetic mutations that are the source of the problem.”</p> <h3><a href="/news/tracking-proteins-using-ai-u-t-scientists-develop-deep-learning-algorithm">Read more about startup by Frey's student using AI to analyze protein data</a></h3> <p>Most new drugs fail in clinical trials, and <a href="https://www.scientificamerican.com/article/cost-to-develop-new-pharmaceutical-drug-now-exceeds-2-5b/">the cost of developing a new drug is over $2.5 billion</a>.</p> <p>Frey hopes that by harnessing the massive amount of genetic data that has become available since the human genome was sequenced in 2001, Deep Genomics can help pharmaceutical companies significantly cut down on the number of failures, and pinpoint the winners earlier. The company plans to collaborate with pharmaceutical companies to develop compounds.&nbsp;</p> <p><a href="/news/vector-institute-points-toronto-global-hot-spot-ai-research">Frey is also a co-founder of the recently formed&nbsp;</a><a href="/news/vector-institute-points-toronto-global-hot-spot-ai-research">Vector Institute</a>, an academia-industry-government centre that solidifies Toronto’s position as a global hub for artificial intelligence research and development. With over $200 million in funding, the institute builds on ֱ’s long-standing strength in branches of AI such as deep learning, machine learning, neural networks, augmented reality, self-driving and autonomous vehicles and robotics.</p> <p>“I think in the next 10 to 20 years, almost all aspects of Canadian society will be impacted by artificial intelligence, from farming to medicine to education,” says Frey. “Artificial intelligence, and deep learning in particular, is the best way to interpret data and then make rational, good choices. As the amount of data grows in all areas of society, AI will play a crucial role in making that happen.”</p> <p>In medicine, Deep Genomics has identified the most promising ways to tackle rare Mendelian disorders, a class of genetic conditions caused by mutations in a single gene.&nbsp;Over 350 million people worldwide are affected by rare Mendelian disorders. Frey says the first three conditions they’ll explore will be disorders of the central nervous system, eye and liver.</p> <p>“So far, we’ve been focusing on our core technology: using machine learning to gain new insights into how mutations anywhere in the genome contribute to disease conditions,” says Frey. “Now it’s time to use that platform to help pharmaceutical companies develop genetic medicines for some of these conditions that affect millions of people.”</p> <h3><a href="http://entrepreneurs.utoronto.ca/">Learn more about entrepreneurship and startups at ֱ</a></h3> </div> <div class="field field--name-field-news-home-page-banner field--type-boolean field--label-above"> <div class="field__label">News home page banner</div> <div class="field__item">Off</div> </div> Wed, 03 May 2017 21:10:58 +0000 ullahnor 107226 at Tracking proteins using AI: ֱ scientists develop deep learning algorithm /news/tracking-proteins-using-ai-u-t-scientists-develop-deep-learning-algorithm <span class="field field--name-title field--type-string field--label-hidden">Tracking proteins using AI: ֱ scientists develop deep learning algorithm</span> <div class="field field--name-field-featured-picture field--type-image field--label-hidden field__item"> <img loading="eager" srcset="/sites/default/files/styles/news_banner_370/public/2017-05-02-cells.jpg?h=afdc3185&amp;itok=f2mDbcWg 370w, /sites/default/files/styles/news_banner_740/public/2017-05-02-cells.jpg?h=afdc3185&amp;itok=gTrrYz0l 740w, /sites/default/files/styles/news_banner_1110/public/2017-05-02-cells.jpg?h=afdc3185&amp;itok=JO-rdYLD 1110w" sizes="(min-width:1200px) 1110px, (max-width: 1199px) 80vw, (max-width: 767px) 90vw, (max-width: 575px) 95vw" width="740" height="494" src="/sites/default/files/styles/news_banner_370/public/2017-05-02-cells.jpg?h=afdc3185&amp;itok=f2mDbcWg" alt="photo of protein"> </div> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span>ullahnor</span></span> <span class="field field--name-created field--type-created field--label-hidden"><time datetime="2017-05-03T11:56:25-04:00" title="Wednesday, May 3, 2017 - 11:56" class="datetime">Wed, 05/03/2017 - 11:56</time> </span> <div class="clearfix text-formatted field field--name-field-cutline-long field--type-text-long field--label-above"> <div class="field__label">Cutline</div> <div class="field__item">Yeast cells (purple) with DNA-containing nuclei (pink) and protein (green) residing in a cell's waste compartment </div> </div> <div class="field field--name-field-author-reporters field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/authors-reporters/jovana-drinjakovic" hreflang="en">Jovana Drinjakovic</a></div> </div> <div class="field field--name-field-author-legacy field--type-string field--label-above"> <div class="field__label">Author legacy</div> <div class="field__item">Jovana Drinjakovic</div> </div> <div class="field field--name-field-topic field--type-entity-reference field--label-above"> <div class="field__label">Topic</div> <div class="field__item"><a href="/news/topics/global-lens" hreflang="en">Global Lens</a></div> </div> <div class="field field--name-field-story-tags field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/tags/deep-learning" hreflang="en">Deep Learning</a></div> <div class="field__item"><a href="/news/tags/artificial-intelligence" hreflang="en">Artificial Intelligence</a></div> <div class="field__item"><a href="/news/tags/vector" hreflang="en">Vector</a></div> <div class="field__item"><a href="/news/tags/donnelly-centre" hreflang="en">Donnelly Centre</a></div> <div class="field__item"><a href="/news/tags/brendan-frey" hreflang="en">Brendan Frey</a></div> <div class="field__item"><a href="/news/tags/molecular-genetics" hreflang="en">Molecular Genetics</a></div> <div class="field__item"><a href="/news/tags/faculty-applied-science-engineering" hreflang="en">Faculty of Applied Science &amp; Engineering</a></div> <div class="field__item"><a href="/news/tags/faculty-medicine" hreflang="en">Faculty of Medicine</a></div> <div class="field__item"><a href="/news/tags/machine-learning" hreflang="en">machine learning</a></div> </div> <div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>ֱ researchers have developed a deep learning algorithm that can track proteins to help&nbsp;reveal what makes cells healthy and what goes wrong in disease.</p> <p>From self-driving cars to computers that can diagnose cancer, artificial intelligence (AI) is shaping the world in ways that are hard to predict, but for cell biologists, the change could not have come soon enough. With new and fully automated microscopes, scientists collect reams of data faster than they can analyze it. &nbsp;</p> <p>The protein-tracking algorithm, Dubbed DeepLoc,&nbsp;can recognize patterns in the cell made by proteins better and much faster than the human eye or previous computer vision-based approaches.</p> <p>“We can learn so much by looking at images of cells. How does the protein look under normal conditions? Do they look different in cells that carry genetic mutations or when we expose cells to drugs or other chemical reagents?” says <strong>Benjamin Grys</strong>, a graduate student in molecular genetics who recently co-authored a&nbsp;paper on the research.&nbsp;“People have tried to manually assess what’s going on with their data, but that takes a lot of time.”</p> <p><a href="http://msb.embopress.org/content/13/4/924">In the cover story of the latest issue of <em>Molecular Systems Biology</em></a>, teams led by the Donnelly Centre's <strong>Brenda Andrews</strong> and <strong>Charles Boone</strong>, both professors of molecular genetics, also describe DeepLoc’s ability to process images from other labs, illustrating its potential for wider use.</p> <p>“Right now, it only takes days to weeks to acquire images of cells and months to years to analyze them. Deep learning will ultimately bring the timescale of this analysis down to the same timescale as the experiments,” says <strong>Oren Kraus</strong>, a lead co-author on the paper and a graduate student co-supervised by Andrews and the Donnelly Centre's <strong>Brendan Frey</strong>, professor of electrical and computer engineering.&nbsp;</p> <h3><a href="/news/u-t-s-deep-genomics-applies-ai-develop-drugs-genetic-conditions">Learn more about Frey's startup&nbsp;which uses&nbsp;AI to develop drugs for genetic conditions</a></h3> <p>Kraus is now working with&nbsp;<strong>Jimmy Ba</strong>, a graduate student of AI pioneer <strong>Geoffrey Hinton</strong>, a ֱ&nbsp;<a href="http://www.provost.utoronto.ca/awards/uprofessors.htm">University Professor&nbsp;Emeritus</a>&nbsp;in computer science who is the chief scientific adviser of <a href="/news/vector-institute-points-toronto-global-hot-spot-ai-research">the newly established Vector Institute</a>, to&nbsp;commercialize the method through a new startup. The goal of the startup named, Phenomic AI, is to analyse cell image-based data for pharmaceutical companies.</p> <p>“In an image based drug screen, you can actually figure out how the drugs are affecting different cells based on how they look rather than some simplified parameters such as live/dead or cell size,” says Kraus. “This way you can extract a lot more information about cell state form these screens. We hope to make the early drug discovery process all the more accurate by finding more subtle effects of chemical compounds.”</p> <p>Similar to other types of AI, in which computers learn to recognize patterns in data, DeepLoc was trained to recognize diverse shapes made by glowing proteins – labelled a fluorescent tag that makes them visible – in cells. But unlike computer vision that requires detailed instructions, DeepLoc learns directly from image pixel data, making it more accurate and faster.</p> <p>Grys and Kraus trained DeepLoc on <a href="http://www.thedonnellycentre.utoronto.ca/news/new-map-uncovers-traffic-life-cell-0">previously published data</a> that shows an area in the cell occupied by more than 4,000 yeast proteins&nbsp;– three-quarters of all proteins in yeast. This dataset remains the most complete map showing exact position for a vast majority of proteins in any cell. When it was first released in 2015, the analysis was done with a complex computer vision and machine learning pipeline that took months to complete. DeepLoc crunched the data in a matter of hours.</p> <p>DeepLoc was able to spot subtle differences between similar images. The initial analysis identified 15 different classes of proteins, each representing distinct neighbourhoods in the cell. DeepLoc identified 22 classes. It was also able to sort cells whose shape changed due to a hormone treatment, a task that the previous pipeline couldn’t complete.</p> <p>Grys and Kraus were able to quickly retrain DeepLoc with images that differed from the original training set, showing that it can be used to process data from other labs. They hope that others in the field –&nbsp;where looking at images by eye is still the norm –&nbsp;will adopt their method.</p> <p>“Someone with some coding experience could implement our method,” says Grys. “All they would have to do is feed in the image-training set that we’ve provided and supplement this with their own data. It takes only an hour or less to retrain DeepLoc and then begin your analysis.”</p> <h3><a href="http://entrepreneurs.utoronto.ca/">Learn more about entrepreneurship and startups at ֱ</a></h3> </div> <div class="field field--name-field-news-home-page-banner field--type-boolean field--label-above"> <div class="field__label">News home page banner</div> <div class="field__item">Off</div> </div> Wed, 03 May 2017 15:56:25 +0000 ullahnor 107223 at This startup combines genomics with one of technology's hottest fields: deep learning /news/startup-combines-genomics-one-technologys-hottest-fields-deep-learning <span class="field field--name-title field--type-string field--label-hidden">This startup combines genomics with one of technology's hottest fields: deep learning</span> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span>sgupta</span></span> <span class="field field--name-created field--type-created field--label-hidden"><time datetime="2015-07-22T02:24:49-04:00" title="Wednesday, July 22, 2015 - 02:24" class="datetime">Wed, 07/22/2015 - 02:24</time> </span> <div class="field field--name-field-author-reporters field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/authors-reporters/marit-mitchell" hreflang="en">Marit Mitchell</a></div> </div> <div class="field field--name-field-author-legacy field--type-string field--label-above"> <div class="field__label">Author legacy</div> <div class="field__item">Marit Mitchell</div> </div> <div class="field field--name-field-topic field--type-entity-reference field--label-above"> <div class="field__label">Topic</div> <div class="field__item"><a href="/news/topics/breaking-research" hreflang="en">Breaking Research</a></div> </div> <div class="field field--name-field-story-tags field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/tags/brendan-frey" hreflang="en">Brendan Frey</a></div> <div class="field__item"><a href="/news/tags/collaboration" hreflang="en">Collaboration</a></div> <div class="field__item"><a href="/news/tags/commercialization" hreflang="en">Commercialization</a></div> <div class="field__item"><a href="/news/tags/engineering" hreflang="en">Engineering</a></div> <div class="field__item"><a href="/news/tags/entrepreneurship" hreflang="en">Entrepreneurship</a></div> <div class="field__item"><a href="/news/tags/genomics" hreflang="en">Genomics</a></div> <div class="field__item"><a href="/news/tags/health" hreflang="en">Health</a></div> <div class="field__item"><a href="/news/tags/international" hreflang="en">International</a></div> <div class="field__item"><a href="/news/tags/medicine" hreflang="en">Medicine</a></div> <div class="field__item"><a href="/news/tags/research" hreflang="en">Research</a></div> <div class="field__item"><a href="/news/tags/startup" hreflang="en">Startup</a></div> <div class="field__item"><a href="/news/tags/top-stories" hreflang="en">Top Stories</a></div> </div> <div class="field field--name-field-subheadline field--type-string-long field--label-above"> <div class="field__label">Subheadline</div> <div class="field__item">Meet Deep Genomics, a privately-held company that seeks to harness the power of deep learning to transform medicine</div> </div> <div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>It’s the first startup in the world to combine more than a decade of world-leading expertise in the fields of both deep learning and genome biology.</p> <p>Its goal: to transform the way genetic diseases are diagnosed and treated.</p> <p>Launched July 22, <a href="http://www.deepgenomics.com/">Deep Genomics </a>was spun out of research at the University of Toronto and its founders say it will transform genetic testing, pharmaceutical development and personalized medicine. The company is already grabbing headlines around the world. (<a href="http://www.washingtonpost.com/blogs/innovations/wp/2015/07/22/meet-deep-genomics-a-start-up-bringing-the-power-of-deep-learning-to-genomics/">Read the Washington Post story</a>. <a href="http://www.theglobeandmail.com/report-on-business/toronto-startup-aims-to-shake-up-genome-sequencing-market/article25612065/">Read the Globe and Mail story.</a>)</p> <p>“Our vision is to change the course of genomic medicine,” said&nbsp;<strong>Brendan Frey</strong>. The company’s president and CEO, Frey is also a professor in The Edward S. Rogers Sr. Department of Electrical &amp; Computer Engineering at ֱ and a senior fellow of the Canadian Institute for Advanced Research.&nbsp;</p> <p>“We’re inventing a new generation of deep learning technologies that can tell us what will happen within a cell when DNA is altered by natural mutations, therapies or even by deliberate gene editing.”</p> <p>Scientists have discovered how to read and write the DNA code in a living body, using hand-held genome sequencers and gene-editing systems. But knowing how to write is different from knowing what to write. To diagnose and treat genetic diseases, scientists must predict the biological consequences of both existing mutations and those they plan to introduce.</p> <p><img alt="photo of Brendan Frey with colleagues at white board" src="/sites/default/files/2015-07-22-Brendan-Frey-and-colleagues_credit-Roberta-Baker.jpg" style="width: 375px; height: 250px; margin: 10px; float: right;">“Companies like Google, Facebook and DeepMind have used deep learning to hugely improve image search, speech recognition and text processing. We’re doing something very different. The mission of Deep Genomics is to save lives and improve health,” said Frey (pictured at right, in yellow/<em>photo by Roberta Baker</em>).&nbsp;</p> <p>Deep Genomics is also releasing its first product, called SPIDEX, which provides information about how hundreds of millions of DNA mutations may alter splicing in the cell, a process that is crucial for normal development. Because errant splicing is behind many diseases and disorders, including cancers and autism spectrum disorder, SPIDEX has immediate and practical importance for genetic testing and pharmaceutical development. The science validating the SPIDEX tool was described in the January 9, 2015 issue of <a href="http://www.sciencemag.org/content/347/6218/1254806">the journal <em>Science</em></a>. (<a href="http://news.utoronto.ca/machine-learning-reveals-unexpected-genetic-roots-cancers-autism-and-other-disorders">Read more about that discovery</a>.)</p> <p>“The genome contains a catalogue of genetic variation that is our DNA blueprint for health and disease,” said Professor <strong>Stephen Scherer</strong>, director of The Centre for Applied Genomics at SickKids and the McLaughlin Centre atֱ, a CIFAR Senior Fellow, and an advisor to Deep Genomics.&nbsp;</p> <p>“Brendan has put together a fantastic team of experts in artificial intelligence and genome biology—if anybody can decode this blueprint and harness it to take us into a new era of genomic medicine, they can."&nbsp;</p> <h3><a href="http://entrepreneurs.utoronto.ca/">To learn more about entrepreneurship and startups at ֱ, visit its Banting &amp; Best Centre for Innovation &amp; Entrepreneurship</a></h3> <p>Until now, geneticists have spent decades experimentally identifying and examining mutations within specific genes that can be clearly connected to disease, such as the BRCA-1 and BRCA-2 genes for breast cancer. However, the number of mutations that could lead to disease is vast and most have not been observed before, let alone studied.&nbsp;</p> <p>These mystery mutations pose an enormous challenge for current genomic diagnosis. Labs send the mutations they’ve collected to Deep Genomics, and the company uses their proprietary deep learning system, which includes SPIDEX, to ‘read’ the genome and assess how likely the mutation is to cause a problem. It can also connect the dots between a variant of unknown significance and a variant that has been linked to disease.&nbsp;</p> <p>“Faced with a new mutation that’s never been seen before, our system can determine whether it impacts cellular biochemistry in the same way as some other highly dangerous mutation,” said Frey.</p> <p>Deep Genomics is committed to supporting publicly funded efforts to improve human health. “Soon after our Science paper was published, medical researchers, diagnosticians and genome biologists asked us to create a database to support academic research,” says Frey. “The first thing we’re doing with the company is releasing this database—that’s very important to us.”</p> <p>“Soon, you’ll be able to have your genome sequenced cheaply and easily with a device that plugs into your laptop. The technology already exists,” &nbsp;Frey said. “When genomic data is easily accessible to everyone, the big questions are going to be about interpreting the data and providing people with smart options. That’s where we come in.”</p> <p>Deep Genomics envisions a future where computers are trusted to predict the outcome of experiments and treatments, long before anyone picks up a test tube. To realize that vision, the company plans to grow its team of data scientists and computational biologists. Deep Genomics will continue to invent new deep learning technologies and work with diagnosticians and biologists to understand the many complex ways that cells interpret DNA, from transcription and splicing to polyadenylation and translation. Building a thorough understanding of these processes has massive implications for genetic testing, pharmaceutical research and development, personalized medicine and improving human longevity.</p> </div> <div class="field field--name-field-news-home-page-banner field--type-boolean field--label-above"> <div class="field__label">News home page banner</div> <div class="field__item">Off</div> </div> <div class="field field--name-field-picpath field--type-string field--label-above"> <div class="field__label">picpath</div> <div class="field__item">sites/default/files/2015-07-21-deep-genomics-one.jpg</div> </div> Wed, 22 Jul 2015 06:24:49 +0000 sgupta 7159 at Machine learning reveals unexpected genetic roots of cancers, autism and other disorders /news/machine-learning-reveals-unexpected-genetic-roots-cancers-autism-and-other-disorders <span class="field field--name-title field--type-string field--label-hidden">Machine learning reveals unexpected genetic roots of cancers, autism and other disorders</span> <span class="field field--name-uid field--type-entity-reference field--label-hidden"><span>sgupta</span></span> <span class="field field--name-created field--type-created field--label-hidden"><time datetime="2014-12-18T04:41:54-05:00" title="Thursday, December 18, 2014 - 04:41" class="datetime">Thu, 12/18/2014 - 04:41</time> </span> <div class="clearfix text-formatted field field--name-field-cutline-long field--type-text-long field--label-above"> <div class="field__label">Cutline</div> <div class="field__item"> Brendan Frey (centre) with first co-authors Leo Lee and Hui Xiong (Photo by Jennifer Wilson)</div> </div> <div class="field field--name-field-author-reporters field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/authors-reporters/marit-mitchell" hreflang="en">Marit Mitchell</a></div> </div> <div class="field field--name-field-author-legacy field--type-string field--label-above"> <div class="field__label">Author legacy</div> <div class="field__item">Marit Mitchell</div> </div> <div class="field field--name-field-topic field--type-entity-reference field--label-above"> <div class="field__label">Topic</div> <div class="field__item"><a href="/news/topics/breaking-research" hreflang="en">Breaking Research</a></div> </div> <div class="field field--name-field-story-tags field--type-entity-reference field--label-hidden field__items"> <div class="field__item"><a href="/news/tags/faculty-arts-science" hreflang="en">Faculty of Arts &amp; Science</a></div> <div class="field__item"><a href="/news/tags/autism" hreflang="en">Autism</a></div> <div class="field__item"><a href="/news/tags/brendan-frey" hreflang="en">Brendan Frey</a></div> <div class="field__item"><a href="/news/tags/cancer" hreflang="en">Cancer</a></div> <div class="field__item"><a href="/news/tags/engineering" hreflang="en">Engineering</a></div> <div class="field__item"><a href="/news/tags/global" hreflang="en">Global</a></div> <div class="field__item"><a href="/news/tags/health" hreflang="en">Health</a></div> <div class="field__item"><a href="/news/tags/international" hreflang="en">International</a></div> <div class="field__item"><a href="/news/tags/medicine" hreflang="en">Medicine</a></div> <div class="field__item"><a href="/news/tags/research" hreflang="en">Research</a></div> <div class="field__item"><a href="/news/tags/top-stories" hreflang="en">Top Stories</a></div> </div> <div class="field field--name-field-subheadline field--type-string-long field--label-above"> <div class="field__label">Subheadline</div> <div class="field__item">Researchers from engineering, biology and medicine teach computers to ‘read the human genome’ and rate likelihood of mutations causing disease, opening vast new possibilities for medicine</div> </div> <div class="clearfix text-formatted field field--name-body field--type-text-with-summary field--label-hidden field__item"><p>In the decade since the genome was sequenced in 2003, scientists, engineers and doctors have struggled to answer an all-consuming question: Which DNA mutations cause disease?&nbsp;</p> <p>A new computational technique developed at the University of Toronto may now be able to tell us.</p> <p>A Canadian research team led by Professor<strong> Brendan Frey</strong> has developed the first method for ‘ranking’ genetic mutations based on how living cells ‘read’ DNA, revealing how likely any given alteration is to cause disease. They used their method to discover unexpected genetic determinants of autism, hereditary cancers and spinal muscular atrophy, a leading genetic cause of infant mortality.</p> <p>Their findings appear in the December 18&nbsp;issue of the leading journal <em><a href="http://www.sciencemag.org/">Science</a>&nbsp;</em>and are already&nbsp;grabbing headlines. (Read the article in <a href="https://www.quantamagazine.org/20141218-machine-intelligence-cracks-genetic-controls/">Quanta Magazine</a>; <a href="http://www.theglobeandmail.com/news/national/canadian-team-makes-breakthrough-in-quest-to-unlock-dnas-secrets/article22147537/">read the Globe and Mail story</a>; <a href="http://www.autismspeaks.org/science/science-news/machine-learning-lets-researchers-search-genome-new-autism-clues">see&nbsp;coverage at autismspeaks.org</a>.)</p> <p>Think of the human genome as a mysterious text, made up of three billion letters.</p> <p>“Over the past decade, a huge amount of effort has been invested into searching for mutations in the genome that cause disease, without a rational approach to understanding why they cause disease,” said&nbsp;Frey. “This is because scientists didn’t have the means to understand the text of the genome and how mutations in it can change the meaning of that text.”</p> <p>It's a puzzle that Frey points out was captured by biologist Eric Lander of the Massachusetts Institute of Technology in a famous quote: “Genome. Bought the book. Hard to read.”&nbsp;</p> <p><img alt src="/sites/default/files/2014-12-18-dominoes-machine-learning.jpg" style="width: 250px; height: 313px; margin: 10px; float: right;">What was Frey’s approach? Scientists&nbsp;know that certain sections of the text, called exons, describe the proteins that are the building blocks of all living cells. What wasn’t appreciated until recently is that other sections, called introns, contain instructions for how to cut and paste exons together, determining which proteins will be produced. This ‘splicing’ process is a crucial step in the cell’s process of converting DNA into proteins, and its disruption is known to contribute to many diseases.&nbsp;</p> <p>(<em>Image at right: artist’s conception of how disease-causing genetic mutations reside with long&nbsp;chains of DNA; photo by Jessica Wilson.</em>)</p> <p>Most research into the genetic roots of disease has focused on mutations within exons, but increasingly scientists are finding that diseases can’t be explained by these mutations. Frey’s team took a completely different approach, examining changes to text that provides instructions for splicing, most of which is in introns.</p> <p>Frey’s team used a new technology called ‘deep learning’ to teach a computer system to scan a piece of DNA, read the genetic instructions that specify how to splice together sections that code for proteins, and determine which proteins will be produced.&nbsp;</p> <p>Unlike other machine learning methods, deep learning can make sense of incredibly complex relationships, such as those found in living systems in biology and medicine.</p> <p>“The success of our project relied crucially on using the latest deep learning methods to analyze the most advanced experimental biology data,” said&nbsp;Frey, whose team included members from ֱ’s Faculty of Applied Science &amp; Engineering, Faculty of Medicine and the Terrence Donnelly Centre for Cellular and Biomolecular Research, as well as Microsoft Research and the Cold Spring Harbor Laboratory.</p> <p>“My collaborators and our graduate students and postdoctoral fellows are world-leading experts in these areas.”&nbsp;</p> <p><iframe allowfullscreen frameborder="0" height="315" src="//www.youtube.com/embed/Zp-HcC5wQ1k?rel=0" width="560"></iframe></p> <p>Once they had taught their system how to read the text of the genome, Frey’s team used it to search for mutations that cause splicing to go wrong. They found that their method correctly predicted 94 per cent of the genetic culprits behind well-studied diseases such as spinal muscular atrophy and colorectal cancer, but more importantly, made accurate predictions for mutations that had never been seen before.</p> <p>They then launched a huge effort to tackle a condition with complex genetic underpinnings: autism spectrum disorder.</p> <p>“With autism there are only a few dozen genes definitely known to be involved and these account for a small proportion of individuals with this condition,” said&nbsp;Frey.&nbsp;</p> <p>In collaboration with Dr. <strong>Stephen Scherer</strong>, senior scientist and director of the University of Toronto McLaughlin Centre and&nbsp;The Centre for Applied Genomics at SickKids, Frey’s team compared mutations discovered in the whole genome sequences of children with autism, but not in controls. Following the traditional approach of studying protein-coding regions, they found no differences. However, when they used their deep learning system to rank mutations according to how much they change splicing, surprising patterns appeared.</p> <p>“When we ranked mutations using our method, striking patterns emerged, revealing 39 novel genes having a potential role in autism susceptibility,” Frey said.</p> <p>And autism is just the beginning&nbsp;–&nbsp;this mutation indexing method is ready to be applied to any number of diseases, and even non-disease traits that differ between individuals.</p> <p>Dr. Juan Valcárcel Juárez, a researcher with the Center for Genomic Regulation in Barcelona, Spain, who was not involved in this research, said: “In a way it is like having a language translator: it allows you to understand another language, even if full command of that language will require that you also study the underlying grammar. The work provides important information for personalized medicine, clearly a key component of future therapies.”</p> </div> <div class="field field--name-field-news-home-page-banner field--type-boolean field--label-above"> <div class="field__label">News home page banner</div> <div class="field__item">Off</div> </div> <div class="field field--name-field-picpath field--type-string field--label-above"> <div class="field__label">picpath</div> <div class="field__item">sites/default/files/2014-12-18-machine-learning.jpg</div> </div> Thu, 18 Dec 2014 09:41:54 +0000 sgupta 6707 at