Web platform to help identify new drug candidate molecules quickly and accurately by predicting drug target protein structures based on deep learning.
On average, it takes 15 years and costs 1 billion dollars just to develop one drug.
In the case of traditional drug discovery, the early-stage process takes 5 years and clinical trials take 10 years. With the introduction of AI to the process, the early stage can be reduced from 5 years to 2 years and the overall process can be cut down from 15 years to 10 years.
Here is the proposed solution, Deep Pharming.
Deep Pharming is a web-based platform that helps to find a new drug candidate by predicting the 3D structure of the drug target protein.
When pharmaceutical scientists input the sequence of the drug target protein into our platform, the platform will convert it into a 3D structure and find the optimal drug candidate.
Lock and Key Model
Drugs and Drug Target proteins can be compared to a lock and key model. When the right key is in the lock, it will open.
Similarly, when we know the exact shape of the drug target protein, we can make the most accurate drug fit in the shape of the Drug Target.
Only when we know the appropriate shape of the target protein can we produce the optimal drug to treat the disease effectively.
The conventional way to test the structure of potential drug candidates is through x-ray crystallography and NMR.
However, our platform will allow our AI to learn from an entire database of proteins at once. This will streamline the R&D process of drug creation by processing more images at once for a lower cost and greater accuracy.