Sberbank will start developing medicines using AI

“As part of the center’s strategy, it is planned to search for biological targets with the participation of associated pharmaceutical companies, generate new molecules, validate and carry out preclinical studies. The starting areas of the center will be the search for medicinal molecules for the treatment of oncology and Alzheimer’s disease, as well as for the fight against diabetes and obesity,” said the top executive.

The center itself will not produce drugs, but will focus on the search for new molecules and preclinical studies that demonstrate that the solutions obtained are effective and can move to the clinical trial phase, said Artur Kadurin, director of AI in Life. Scientific Group of Sciences of the AIRI Institute. He noted that the drugs will be synthesized by partners of Russian pharmaceutical companies. Kadurin did not name them.

Sberbank is already developing joint projects with drug manufacturers. For example, one lender is working with R-Pharm on a project that can “significantly reduce the timeline” for developing drugs, including for the treatment of melanoma, lung cancer, kidney cancer, lymphoma and other forms of of the disease, Andrey Burkov, director of strategic marketing and product portfolio development at R-Pharm.

According to him, in 2024 the bank and R-Pharm, within the framework of a pilot project, plan to develop a drug for the treatment of various oncological diseases. Plans include expanding cooperation and including in development not only biological drugs, but also small molecules. Burkov did not specify what role Sberbank plays in this project.

There are other companies in Russia that develop drugs using AI, for example Ligand Pro, which uses neural networks to create drugs in the early stages, and the Geropharm group of companies, which previously specialized in the production of insulin, noted Ekaterina Surkova, director . from the scientific communication department of the medical organization Genotek.

The use of AI in drug development can significantly reduce time and costs, but at the moment neural networks are used only in the initial stage, as they cannot replace clinical trials and the certification process, the researchers explained. experts interviewed by Vedomosti.