Building AI for Drug Discovery
Gregg W. Casey
CTO/Founder TruthSayer AI | Leading AI/LLM Engineer on Transformative AI Solutions | Artificial Intelligence | Machine Learning | Generative AI
Drug discovery is perhaps the most exciting potential use case for Artificial Intelligence (AI).??AI-assisted Discovery of new drugs is already something over 43 companies have been working on for a few years now.
Unassisted (non-AI) drug discovery is a time-consuming process that can take over 10 years and as much as $2.5B to get a new drug to market. It can be a tedious process of testing thousands of compounds knowing that the likelihood of any of them becoming a profitable drug is small. But what if we could zero in on a small list of compounds that are much more likely to be successful?
So how can we use AI to lower risk, increase time-to-market, and save lives? It turns out that there are a lot of ways to tackle this problem and new and incredibly creative ideas are coming to light all the time. It takes a team with a don't-give-up attitude and an ability to think out of the box.
In the industry we initially started with what was not much more than but brute force techniques using AI to test millions of compounds in AI simulations. Then we tried things like "Virtual Screening" where we would predict whether a (known) compound was a likely candidate based on things like pK changes and toxicity. This process had lots of limitations and, although it was much faster than previous methods, it was still too slow to be practical.
There is still progress to be made but new processes are very encouraging. In an article here I go into detail on one method for using AI in Drug Discovery that has seen some success and how it builds on Generative AI techniques that have become so popular in the general public.
领英推荐
There is so much going on in this field that there is something new all the time. Soon I will also write on other methods like using Graph Neural Networks (GNNs) which is also exciting and efficient.
Please share your thoughts in a comment. If you are working in Drug Discovery, what techniques are you using?
More detailed article: https://medium.com/@gregg.casey_44706/breakthroughs-in-ai-for-drug-discovery-ae6e902cfff
CTO/Founder TruthSayer AI | Leading AI/LLM Engineer on Transformative AI Solutions | Artificial Intelligence | Machine Learning | Generative AI
10 个月I'll definitely add this to my reading list Ernest Bonat, Ph.D. We are about to launch our product so I'll have to wait a bit on reading it though.
Data Engineer | Data Scientist | Bioinformatics Scientist | Computer Science Faculty | Data Science Faculty
10 个月If anyone is interested in learning how to 'Apply Machine Learning Algorithms for Classification of Drug Discovery Data,' please read and understand this paper: https://ernest-bonat.medium.com/apply-machine-learning-algorithms-for-classification-drug-discovery-data-a37c71416414#9482. Feel free to contact me with any questions.
Great article Gregg. Not always watching this space so I appreciate your insights. I would expect AI in conjunction with digital twins would add significant efficiency and value.