Japanese pharmaceutical company begins testing drugs synthesized using a neural network

Pharmacists and programmers from Sumitomo Dainippon Pharma and Exscientia threw firewood into the bonfire of the dispute “where should the car’s independence and control begin manually”: on January 30, a press release was published on the Exscientia official website stating that they’ve developed AI Pharmaceutical Company Sumitomo Dainippon Pharma determined the formula and synthesized the active substance for the drug against OCD.



Of course, Exscientia marketers call their development “Artificial Intelligence”, but the development of a new drug was carried out using a learning neural network. In fact, the neural network determined the formula of a new drug through enumeration and analysis of combinations of known active substances. Development and synthesis are now complete, and SDP, the drug manufacturing company, is moving on to the first phase of animal clinical trials.

One of the advantages of involving neural networks in determining the formula of the active substance of drugs is the speed of the machine: the network, like scientists, sortes out possible combinations of substances and makes a forecast of their main effect, but unlike a team of biochemists, neural networks require the creation of rough formulas significantly less time. So, SDP specialists with technical support from Exscientia managed in 12 months instead of the usual ~ 4.5 years that such studies usually last.

The developed substance was called DSP-1181, and its scope is the fight against obsessive-compulsive disorder by increasing the response of the serotonin 5-HT1A receptor , that is, the active substance is a receptor agonist. Researchers claim that DSP-1181 is one of the most promising existing OCD treatments.

Representative of Sumitomo Dainippon Pharma in the person of the company's chief executive officer Toru Kimura speaks of the latest development: “We are very excited about the results of a joint study that allowed us to develop a candidate compound in such a short time. Our experience in the search for new drugs based on GPCR monoamines together with the capabilities of AI allowed us to work efficiently and ensure a successful result. "We will continue to work hard to ensure that our medicine passes clinical trials and reaches patients as soon as possible."

Unfortunately, representatives of Exscientia did not provide any technical details of the development in the press release, however, the following picture can be drawn up from fragments of individual phrases:

• neural network was used;
• Japanese company SDP provided an array of tagged data for training and network operation, as they have “experience in finding drugs based on GPCR monoamines ( monoamine neurotransmitters )”;
• the neural network was engaged in mechanical enumeration of candidate compounds for the active substance, after which the generated data array was checked by pharmaceutical company scientists;
• there is nothing breakthrough in technology itself.

It is worth noting that for all the banality of the process of searching for the active substance in the described case (and filtering a large data array using machine learning is now used in many fields, from mathematics and astrophysics, to oil production and geological exploration), the very fact that the formula of the active substance of the drug from such an unpleasant neurological disease as OCD, was originally bred by the machine - is impressive. The most important thing that Sumitomo Dainippon Pharma and Exscientia have done is to expand the horizon of the field of application of modern computers and pharmaceutical power. But the development of a drug is not a fast and extremely expensive process. On the topic, you can watch an interview (or read its transcript in the same place) with Alexander Zhavoronkov. Also itIt was also translated into Habr.

So, according to Zhavoronkov, neural networks are already actively used in pharmaceuticals, but usually they are used to analyze existing scientific works and the effectiveness of various studies: did the medicine go on sale or did its development stop on the basis of the initial publication and did not pass the clinical studies? Neural networks also processed the clinical data of patients; it is also known about the direct involvement of machine learning in the search for the active substance with subsequent synthesis. Basically, the role of “AI” is reduced to processing an array of indirect data and preparatory work - in what direction should researchers move.

With successful clinical trials, Sumitomo Dainippon Pharma and Exscientia are likely to be among the first companies to use the neural network and so quickly receive the active drug substance for a wide range of consumers.

Source: https://habr.com/ru/post/undefined/