Have you ever wondered how long it usually takes to create a new medicine? Experts say it usually takes 10~15 years. It takes much longer than expected. Why would it take so long? It’s because humans each have so many different reactions to the medicine. But since AI is taking over almost all of the field, it is no exception for health care. Experts say that AI can reduce the amount of money and time that is used on making medicine.
Medicine is usually made in 6 steps. First, scientists try to find the drug candidates that have a potential to cure it. The second step is called Preclinical Testing. When they have found the right chemical compounds to use as a medicine, they test it on animals before testing it into humans and it is more studied in the lab. Third, if Preclinical Testing is done successfully, it is now tested on humans. When the test is done, they send their drug to the regulatory agencies like FDA to get confirmed. Once approved, the drug is manufactured in company and then sent to hospitals or pharmacies. Lastly, even after it is confirmed, the monitoring continues to prevent unnecessary incidents and side effects.
Then how can AI be helpful to this step? AI can be used to find the proper chemical compounds more efficiently, this will speed up the whole process much more because it usually takes 2~4 years but AI can reduce it to 6 months ~2 years. Then, it evaluates tons of chemical compounds to find out which one will best suit the protein that is related to the disease and can cure the disease. Next, AI suggests molecules structure that will likely achieve the needed properties. It also predicts the reactions of animals without having any animal testing but only according to the given information which will be more ethical and reduce the time it takes for the animal testing process. It can even create a more efficient chemical structure that can cure the disease and can predict side effects and poor absorption which will help reduce the trial and error process.
The reason drug development takes too long is that it takes too much time finding the right chemical compound and even though we say we found one, there are rarely ones that can be used for real drugs and medication. Also it is really hard and complicated in checking side effects or safety to humans. They have to check it through tons of people and animals to clearly make sure if it is safe since people’s lives are dependent on it.
AI has the potential to be the most efficient tool for drug development. Recently, NVIDIA released a new AI ‘Evo’ that predicts protein structure and will be used for drug development. Also AI got a nobel prize in chemistry for predicting the structure for almost all known proteins. It was awarded to Demis Hassabis and John M Jumper for creating AlphaFold2 (AI that was used to predict protein structure.) It helped predict almost all proteins which will contribute significantly in making new medicines, understanding new diseases at the molecular level and making it possible for a rapid response to new viruses like Covid-19.
Drug development is a long, complex process but with the help of AI, scientists expect that new medicines will become available to us faster, safer.
By. Madison Yi
Works Cited
“the nobel prize.” They cracked the code for protein’s amazing structure, the nobel prize, https://www.nobelprize.org/prizes/chemistry/2024/press-release/
Amisha et al. “overview of AI in medicine.” J family med prime care, PubMed, https://pmc.ncbi.nlm.nih.gov/articles/PMC6691444/
“E health.” Public Health, https://health.ec.europa.eu/ehealth-digital-health-and-care/artificial-intelligence-healthcare_en