Glaxo turning to AI to mine other’s failures
Each year, drugmakers shelve, write off or discard thousands of possible treatments that don’t appear to have value in the clinic. To GlaxoSmithKline Plc, (with facilities in Upper Providence and Upper Merion, Montgomery County), those failures are a rich vein ready for prospecting, writes James Paton for bloomberg.com.
Glaxo is turning to supercomputers to trace the paths of millions of unsuccessful drugs, hoping to find a quicker course to the patient bedside and cut costs. The U.K.’s biggest pharmaceutical company is joining with U.S. government and university researchers in a bid to shrink the journey from identifying a biological target to finding a drug that hits it. It’s often a six-year process. Glaxo wants to do it in one.
[uam_ad id=”54865″]
“The industry has to begin sharing its failure,” John Baldoni, a senior vice president overseeing Glaxo’s artificial intelligence efforts, said in an interview at the company’s research site near Philadelphia. “We want to go where there has been a lot of failure because that’s where the most data are.”
Baldoni is betting that artificial intelligence, used in innovations like self-driving cars and facial recognition, will help debug the search for winners. Glaxo CEO Emma Walmsley is turning to such high-tech efforts to boost efficiency as she narrows the pharma unit’s focus on fewer, but bigger, opportunities and seeks to boost returns from research and development, the story continues.
Since so many drugs never take off, most information about how molecules behave in cells can be found in experiments that didn’t ultimately bear fruit. Harnessing that largely unexplored mountain of data can allow researchers to reach faster decisions on whether to test drugs in patients — or not, he said.
To read the complete story click here.
[uam_ad id=”54875″]
Stay Connected, Stay Informed
Subscribe for great stories in your community!
"*" indicates required fields
