(KABC) Los Angeles Millions, if not billions, of dollars are required to transport a medicine from the laboratory to the pharmacy.
Only 37 medications were authorized by the FDA last year, the lowest number to pass regulatory review since 2016. There were thousands of drugs undergoing clinical testing. Artificial intelligence may soon make it possible to determine which medications will be effective and which won’t.
Dr. Sudipta Seal, a professor in the University of Central Florida’s department of Materials Science and Engineering, claimed that “we can reduce the experiments as well as the time.”
A new medicine takes, on average, 10 years to develop. Now, computer scientists are working to accelerate that process using a drug screening method based on AI that matches drug and protein interactions, a very tough task.
It verbalizes the intricate interactions that take place at each drug’s protein-binding site. After that, the AI model decodes that language to determine where region of the virus protein a given medication will attach to. It can forecast how effectively a medicine will function with 97% accuracy.
According to Mehdi Yazdani-Jahromi, a PhD candidate at UCF, “you can just, give it, for example, COVID protein and test it against all the FDA-approved drugs and see whether or not they bind or not.” The artwork’s beauty lies in that.
The model is called AttentionSiteDTI, and it can be utilized right away and without charge by anyone creating a novel medicine.
“I think it’s going to revolutionize medical fields in so many different ways,” said Dr. Ozlem Garibay of UCF’s computer science department.
It’s straightforward: the less time spent in the lab, the cheaper it will be to produce. Researchers think that consumers could benefit from savings.
They are now prepared to develop a website similar to ChatGPT that will make it easier for other scientists to input their data and determine whether or not their medicine will be effective.
Must read:Big Tech bounces back and is ready for game-changing AI investments