Center for Artificial Intelligence in Chemistry

Artificial intelligence Nanotechnology Drug Design

Head of Laboratory

Vinogradov, Vladimir V

DSc

CoLab Scopus

Publications

127

Citations

2400

h-index

Publications

127

Citations

2423

h-index

Authorization required.

In 2022, as part of the Data-Driven Nanomedicine project, ITMO University opened a separate department, the Center for Artificial Intelligence in Chemistry, and a new master's program, Chemistry and Artificial Intelligence. Now the Center is a community of undergraduates and graduate students who apply machine learning methods to scientific problems in the field of chemistry, the solution of which previously occurred either very poorly or was not solved at all. Our scientists are not afraid of big challenges and frontier goals. The principle of their work is think digital and without limits. We believe in innovative ideas and do not like to do routine work — that's why we are developing digital services and web platforms to solve global problems of chemists and materials scientists, articles about which are published by top journals included in the Nature Index reputation rating. Now we are implementing our own educational program, holding events, cooperating with the industry, winning competitions, giving lectures and striving for new scientific breakthroughs and collaborations.

Analytical methods, numerical modeling, programming
Artificial intelligence
Molecular dynamics and quantum chemical calculations
DFT calculations

Vladimir Vinogradov 🥼

Head of Laboratory

Andrey Dmitrenko

Principal researcher

Aleksandra Falchevskaya

Senior Researcher

Daniil Kladko 🤝 🥼

Senior Researcher

Nikita Serov

Senior Researcher

Julia Razlivina 🤝

Senior Researcher

Valeria Odegova

Researcher

Susan Jyakhwo 🤝

Researcher

Mariia Eremeyeva 🤝

Researcher

Nina Gubina

Researcher

Anastasiya Orlova

Researcher

Ivan Dubrovsky

Researcher

Research directions

AI in working with molecular machines

The main goal of the direction is to develop a set of tools using artificial intelligence, data science and computational methods. This will allow the creation of molecular machines, such as enzymes, to perform applied tasks. Such tools will allow us to better understand the structure of molecular machines, which will make it possible to form a system that we call "Lego molecular machines". This system implies a search/creation of separate and independent functional units or modules, as well as ways to assemble them into turnkey molecular machines with predetermined behavior and a set of activities.

The use of AI in the food industry

The Center for Artificial Intelligence in Chemistry focuses on two key areas in the field of food technology. This is the application of AI to optimize winemaking processes, including the generation of blends, the prediction of aromatic characteristics of wine and the creation of a recommendation system for choosing the best wine options. In addition, the Center is also engaged in the use of artificial intelligence to improve brewing processes. Both of these projects are aimed at creating digital services that help both consumers and manufacturers by optimizing production processes and improving the product selection experience.

Digitization and prediction of properties of nanomaterials

Our Center specializes in advanced methods using neural networks, machine learning and artificial intelligence to predict the properties of nanomaterials. Scientists are developing models and web platforms capable of predicting cellular toxicity, magnetic and catalytic properties of nanoparticles, which is important for the development of new materials with certain functional characteristics. This approach opens up new prospects in the field of chemical research, accelerating the processes of synthesis and optimization of the use of compounds obtained by scientists.

Data-driven drug discovery

The main idea of the data-driven drug discovery direction is the development of new safe and effective medicines based on data on existing medicines. With the help of generative artificial intelligence, fundamentally new compounds are created, and predictive machine learning algorithms, trained on a large number of experimental and calculated data, select the most optimal candidates among them for further development. This approach makes it possible to analyze a huge number of molecules in a short time and select those that will incur minimal risks at the stages of preclinical and clinical trials.