Artificial intelligence has created a new drug in just 21 days
Artificial intelligence has succeeded in many areas of our lives. And basically AI finds its application in the field of complex calculations, constructing mathematical models and so on. However, in the medical field of artificial intelligence may be less useful (if you follow the success of AI? We regularly talk about them in our Telegram channel). For example, recently the Hong Kong company Insilico Medicine published a study showing that its AI-based system and deep learning can create new drugs for certain pathologies in just 3 weeks. This is several tens of times faster than conventional methods.
AI use in drug development - a very promising technology
As artificial intelligence to create a drug
AI from the company Insilico Medicine is called GENTRL. Moreover, it is noteworthy that at the helm of the company is worth our compatriot Alex Lark. Mr. Lark is still in the mid-2000s, he received a master's degree from Johns Hopkins University, and then a doctorate at Moscow State University, where his research focused on the use of machine learning to study the physics of molecular interactions in biological systems. In 2006, he has worked on ATI company, until it was not acquired by the manufacturer of AMD chips. In 2014, Alex founded the already mentioned Insilico Medicine, they have had experience in the technology industry, and became interested in pharmacy.
This is interesting: How does the AI
If you go back to the AI, the developers themselves are called the core technology of artificial intelligence work "tensor generative learning". It allows the AI, if you do not go into details, more efficiently and quickly to learn the required skills.
We thought, if we can get the car to come up with zero new molecules with specific properties rather than making them sort through dozens of options available, - says Alex Lark.
Insilico GENTRL used to create multiple (or to be precise, 6) versions of drugs for the treatment of muscle fibrosis. By drugs inhibit DDR1 receptor, which is involved in the development of the disease. To do this, AI took 21 days, after which the researchers chose the most suitable options for drugs and tested them in the laboratory animals. It took another 25 days. Thus the choice of a potential drug took just 46 days. For comparison, the traditional process of development candidates for the drug takes about 8 years and costing the company several million dollars. While the creation of the AI took a total of 150 thousand dollars.
Left - normal muscle tissue. Right - a fabric with the development of fibrosis
Thus Insilico stress that they have not proved that a new drug is more effective than existing drugs. However, time and costs, which have gone from scientists to create drug candidates, much less than conventional methods of pharmacy. Their work was to demonstrate the enormous potential of the systems based on artificial intelligence in the development of new drugs. Now, scientists have focused on the improvement of technology and, of course, to test the effectiveness of new drugs developed AI.