Developing new drugs is a long and expensive process. We propose a new method that uses quantum computing and machine learning to speed up this process from 8-15 years to 1.5-2 years. Our method has three steps:
First, we use machine learning molecule generation to create many potential drug candidates based on the structure of the virus we want to target. Then, we will also use quantum computing simulation to filter top drug candidates' reaction and binding efficiency with the target virus from a large input initial pool of molecules.
Second, we will compare the candidates that were filtered and generated and optimize them by making more variations of them using molecular variation machine learning. The drug candidates that appear as a result of both processes will be designated as likely drug candidates and will undergo more optimization than the rest.
Third, we use quantum computing again to test all the variations under different conditions and standards. We select the best ones that are safe and effective for pre-clinical trials.