THANK YOU FOR SUBSCRIBING
Pharma Tech Outlook | Tuesday, June 23, 2020
AI can reduce drug discovery time even for rare diseases, which traditionally requires years, as it involves checking thousands of molecule combinations.
Stay ahead of the industry with exclusive feature stories on the top companies, expert insights and the latest news delivered straight to your inbox. Subscribe today.
Fremont, CA: A small prescription drug can stop a devastating illness. But there is a long process to approve such medication. In the last 10 years, that process has become more complicated and expensive. The cost to develop a new medicine has gone up by 145 percent while taking 10 years on average to develop one.
Recently, scientists developed an AI model that predicts coronavirus in people without the need for testing using all the symptoms of COVID-19.
More than 400 million people suffer from rare diseases, and 95 percent of them have no approved treatment from the FDA. But, Artificial Intelligence (AI) with its subfields, machine learning (ML), and deep learning (DL) is set to change these statistics for the better, especially during the COVID-19 pandemic.
AI reduces the time taken to receive an accurate diagnosis while streamlining the path to develop new treatments for devastating illnesses. Not only that, in fields like oncology, gastroenterology where a huge chunk of data is already available, AI, ML, and DL can help in discovering compounds potential in becoming new drugs while repurposing existing drugs in short time spans.
How AI aids drug discovery?
[vendor_logo_first]Today, every major pharmaceutical company has partnerships with at least one AI-based drug development company, as their focus has shifted to address rare diseases that involve hundreds of proteins. Simple drugs target one or a few such proteins, but that’s not enough.
Discovering a drug has a process that begins identifying a target for the drug. Complex diseases need an understanding of how molecules interact with one another. Here, AI companies, such as Cyclica, have developed software that can process biophysical and biochemical properties of millions of molecules and matches them to the structures and properties of almost 150,000 proteins, thereby finding molecules that bind to a target.
Merck and Bayer, among the many big pharma companies, announced its partnership with Cyclica to aid them in drug discovery efforts. Not all pharma companies release information on what AI-generated drug candidates come out of such collaborations.
In some instances, AI aids when a researcher has discovered a target protein but is not able to determine a molecule that binds with the target. Major firms like Celgene, GSK, Sanofi, and Sunovion has partnered with Exscientia in attempting to solve such problems.
Exscientia compares the target protein against a database of a billion interactions. This process finds all possible compounds that might work on a target are narrowed to a few favorable drug candidates.
Centralizing Data for ML
For effective results, it should be ensured that all research data is readily available in one centralized location. Thermo Fisher Scientific stands at the forefront in enabling centralized data management for labs. They provide effective communication between laboratory instruments and software systems. Harnessing this, a central system can facilitate machine learning.
More in News