Using Artificial Intelligence to Discover New Drug Treatments: The Next Science


In the fields of medicine and pharmacology, drugs have historically been discovered either by identifying active ingredients from traditional remedies or by chance, in the same way that penicillin was discovered in 1928. But modern scientists have found an alternative method that makes it possible to research new pharmaceutical products. faster, cheaper and more efficient.

AI, or artificial intelligence, can be found in everything from computers playing chess and self-driving cars to applying maps on your phone when it calculates directions. And AI has found a new home: discovering new drugs.

It currently takes an average of 10 years and over $ 2 billion to create a new drug and get it approved. LSU’s DeepDrug team expects its AI tool, eSynth, to reduce the time it takes for drug discovery and preclinical testing from an average of three years to six to eight months.

DeepDrug is an interdisciplinary team at LSU led by Supratik Mukhopadhyay, Associate Professor in the Department of Computer Science, and Michal Brylinski, Associate Professor in the Department of Biological Sciences. If the team is successful, their AI could suggest antiviral drugs to reduce the impact of COVID-19 in just days. The team is careful to stress, however, that their end result will not be a vaccine or a complete cure.

“We cannot remove the virus from the body or prevent it from infecting more people,” Brylinski said. “What we can do is reduce the threat and the death rate, especially for people with serious illnesses who might have mild conditions instead. They will still be infected, but they will survive. More people could survive the pandemic. “

DeepDrug’s work recently reached the semi-finals of the IBM Watson AI XPRIZE, competing against AI research teams from around the world for a total prize of $ 5 million.

So how does AI help discover new drugs?
Several decades of research have led to the development of mathematical tools that have helped scientists improve their understanding of the nature of pathogens. They also identify potential drug targets and predict epidemics.

AI-based model systems hold promise because they can reason on huge amounts of data. Then, they can identify approaches to treat the disease by proposing a drug target, designing a molecule, and even defining patients to test it with. While still seen as the early days of AI’s full potential in the wider pharmaceutical industry, the technology is already advancing in this new role.

A few drugs, such as chloroquine, hydroxychloroquine, azithromycin, and remdesivir, have been approved by the FDA for the treatment of SARS-CoV-2 (betacoronavirus) infections. Most of these treatments were discovered through trial and error in different parts of the world.

What DeepDrug offers is a principled approach to drug discovery and drug reorientation (meaning that established drugs are used to treat new or different conditions) based on data sets that would be too large – or at least slow enough – to process without the use of AI.

Additionally, by helping to predict a drug’s pass / fail rate, AI could prevent researchers from wasting a lot of time exploring what is likely to be dead ends. DeepDrug’s AI would save time by exploring antivirals currently approved by the FDA to see if they could also treat COVID-19, as well as many possible combinations with other antivirals and drugs.

The team is also working on additional AI modules. eToxPred will ensure that the compounds identified are safe for humans. Another module, eDrugRes, will examine the protein-protein interaction network of pathogens to predict susceptibility and / or resistance to known drugs.

AI has already helped us discover new ways to mitigate the effects of the coronavirus. Recently, researchers in Singapore used AI to identify the best therapies to fight the virus that causes COVID-19. Their results identified a combination of the drugs remdesivir, ritonavir and lopinavir at specific doses.

Special thanks to Associate Professor Michal Brylinski of the Department of Biological Sciences at LSU for his contribution to this report.

LSU researchers enter semi-finals for the $ 5 million IBM Watson AI XPRIZE:
DeepDrug team website:


About Author

Comments are closed.