AI Entrepreneur finds new drug candidates in the fight against Coronavirus
NEW YORK, NY, USA, March 12, 2020 /EINPresswire.com/ -- The race to find drugs that fight coronavirus is on, and a data scientist from New York, Matt O’Connor, recently won a competition to identify novel drugs to treat COVID-19. While multiple parties are at work on a vaccine, the competition, sponsored by Sage Health, tasked participants with using machine learning to accelerate the drug discovery process, to find a more immediate treatment to the virus. "It is very important to me that we respond with a comprehensive strategy, not just with an eye on long-term prevention, but also on finding drugs for treating situations like those we see today," O'Connor, partner at Reboot.ai, said in a statement.
The process included algorithms that combed through a database of over 5 million known molecules and compounds to discover those with the most potential in treating COVID-19, as well as generate entirely new, undiscovered molecules. Combining a generative Recurrent Neural Network model with principles of transfer learning and genetic algorithms, O'Connor uncovered multiple novel drug candidates potentially even more effective for treatment than the experimental drug Remdesivir, which was originally created to combat Ebola and is currently in trials for treating COVID-19. The competition’s organizer, Sage Health, indicated samples of the top candidate compounded would be sent to the Wuhan Institute of Virology for further analysis.
“I have long held the belief that artificial intelligence and data-based approaches can and should be used by everyday developers and by companies of any industry. AI is no longer only in the realm of PHDs, statisticians, or academia”, O'Connor continued. "It's not a matter of technology, but of human intelligence plus artificial intelligence. With the right strategy and approach, any company, individual, or government can unlock tremendous benefits through applied AI."
You can read through O’Connor’s winning submission here, drop him a message at matt@reboot.ai or tweet him @mattroconnor.
The process included algorithms that combed through a database of over 5 million known molecules and compounds to discover those with the most potential in treating COVID-19, as well as generate entirely new, undiscovered molecules. Combining a generative Recurrent Neural Network model with principles of transfer learning and genetic algorithms, O'Connor uncovered multiple novel drug candidates potentially even more effective for treatment than the experimental drug Remdesivir, which was originally created to combat Ebola and is currently in trials for treating COVID-19. The competition’s organizer, Sage Health, indicated samples of the top candidate compounded would be sent to the Wuhan Institute of Virology for further analysis.
“I have long held the belief that artificial intelligence and data-based approaches can and should be used by everyday developers and by companies of any industry. AI is no longer only in the realm of PHDs, statisticians, or academia”, O'Connor continued. "It's not a matter of technology, but of human intelligence plus artificial intelligence. With the right strategy and approach, any company, individual, or government can unlock tremendous benefits through applied AI."
You can read through O’Connor’s winning submission here, drop him a message at matt@reboot.ai or tweet him @mattroconnor.
Matt O'Connor
Reboot.ai
+1 646-236-4274
email us here
Visit us on social media:
Twitter
LinkedIn
Legal Disclaimer:
EIN Presswire provides this news content "as is" without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.