In four years, more than 3,500 students have successfully completed their studies at the NSU “Digital Department”

In 2026, 1,075 people completed training under the Novosibirsk State University’s “Digital Department” project. Students studied programming, machine learning, digital products, SMM, bioinformatics, digital law, and medicine — applying the acquired knowledge while working on their own projects. Since 2022, more than 3,500 students have graduated from NSU’s “Digital Department” programs.

The “Digital Department” allows students of fields unrelated to information technology to simultaneously gain knowledge of modern digital technologies and additional qualifications in the IT sector alongside their main educational program. Training for project participants is provided free of charge.

From programming to digital medicine

Since the project’s inception in 2022, the list and content of educational programs of the “NSU Digital Department” have been repeatedly updated. At the same time, two programs have been implemented since the very launch of the project — “Basic Programming in Python” and “Practical Bioinformatics and Molecular Biology.” The latter was previously called “Technologies for Managing the Properties of Biological Objects: Methods of Bioinformatics and Molecular Biology.”

Today, within the framework of the “NSU Digital Department,” seven additional professional education programs are being implemented. These programs allow students of fields and specialties not related to IT to simultaneously gain an understanding of modern digital technologies and their application in specific professional areas alongside their university studies, as well as additional qualifications in the field of information technology.

Industry specificity is especially pronounced in the programs “Digital Jurisprudence” and “Digital Medicine”. The first is aimed at training specialists who not only have a strong knowledge of law but also possess digital skills and can collaborate with technical experts. The second is intended for physicians in clinical specialties, medical informatics specialists, and developers of AI systems in healthcare.

A special place is occupied by the program “Practical Bioinformatics and Molecular Biology”. It is designed for specialists in bioinformatics, biotechnology, as well as molecular, synthetic, and structural biology. According to Marina Shashkova, head of the Digital Department at NSU, Today, this professional retraining program has no equivalents in Russian universities.

Universal digital skills can be acquired through the programs “Basic Programming in Python”, “Machine Learning and Neural Networks”, “SMM Tools for Promotion”, and “Design and Implementation of Digital Products.” They are available to students of all fields of study and specialties. These programs are most popular among students in social and humanitarian fields.

From training to practical solutions

One of the features of the “Digital Department of NSU” is the opportunity not only to study theoretical material but also to apply it when solving practical problems. Thus, participants in the program “Digital Medicine” developed clinical decision support systems, chatbots for patients and healthcare professionals, risk calculators, NLP tools for analyzing patient complaints, as well as analytical and predictive models for medical organizations.

One of the projects was “ZdravoMetrika” — an analytics system for the operation of the clinic’s medical information system. The project was created Evelina Fateeva, Darya Karasyova, Ilya Evgenov and Irina Karpets based at the State Budgetary Healthcare Institution of Novosibirsk Region “City Clinical Polyclinic No. 7”, which acted as the customer for the development.

The clinic is a multidisciplinary outpatient facility in Novosibirsk and includes several therapeutic and pediatric departments. Therefore, analyzing the workload of doctors and departments is of practical importance to the institution. The program participants developed a desktop application that automatically generates an analytical summary of the workload of doctors and departments. It is expected that using such a tool will reduce the workload on doctors and departments through better management of patient flows, optimize branch operations, and increase the speed of management decisions by department heads.

How to reduce queues in pharmacies

We solved a practical problem Olesya Pochikovskaya and Natalya Mishchenko. Their project was dedicated to developing a queue management system in pharmacies, taking into account changes in load throughout the day and week. The authors started from the hypothesis that long waiting times could affect not only customer satisfaction but also the pharmacy’s revenue and customers’ willingness to return.

To test the assumption, the participants analyzed the workload of pharmacists and the dependence of customer flow on temporal and external factors. Based on the obtained data, an algorithm was developed that allows forecasting the need to open additional registers and service windows. As a result, a predictive model was created, which helps reduce client waiting times and increase the efficiency of the pharmacy’s operations.

Digital technologies for biology and medicine

A separate direction of the “Digital Department” is the program “Practical Bioinformatics and Molecular Biology”, combining knowledge from biology, genetics, data analysis, and programming.

Students note that the skills acquired help solve real research problems.

— During the training, I learned Linux for bioinformatics. With its help, I now know how to write simple commands with arguments, perform basic file operations, as well as use advanced commands to work with and analyze them. The course also showed how to write, run, and manage Bash scripts and how to use conditional operators., — says Darya Pershina.

According to her, the part of the course dedicated to machine learning and neural network methods of image analysis became especially interesting.

— After completing it, I learned what image analysis methods exist and how they are specifically applied for data analysis. There was also a course on molecular biology methods and genetic engineering — the information complemented the knowledge I had already acquired and helped me recall important details., — adds Darya.

For Petra Rogacheva the training became an opportunity to delve deeper into sequencing data analysis.

— I was going to the digital department to study pipelines specifically for RNA-seq and other NGS data analysis. I really liked these videos: no fluff, but also not dry, they keep you focused. Therefore, I will highlight Alexander Bobrovsky, whose videos really helped me understand this topic, — Petr says.

The student also became acquainted with the Nextflow tool, which allows combining different scripts and programs into unified reproducible pipelines.

— I don’t have a technical background, so the possibility of scaling pipelines was new to me, — explains the student.

Victoria Milyukova notes that the program helped her systematize knowledge about statistics and master the tools used in bioinformatics:

— The course provided a closer look at statistics for processing experimental data and for work in bioinformatics. I mastered programs such as MEGA, UGENE, Conda, Linux. I particularly liked the practical playlist on working with sequences — concise and clear.

Machine learning: from DNA to chemical reactions

Another program of the NSU Digital Department — “Machine Learning and Neural Networks”. Here digital methods are applied to problems from a wide range of scientific fields.

Among the top projects of the participants are algorithm for automatic analysis of microfauna complexes using machine learning methods. Project author Ekaterina Bakanova Trained a neural network to recognize species of shelled microorganisms— foraminifera—from photographs taken from three angles. The best model achieved 96.2% accuracy across 12 classes.

Using the Grad-CAM method, it was also possible to determine which features the model uses during classification. It turned out that the neural network focuses on the same morphological characteristics that paleontologists use — the shape of the shell, the aperture, and the arrangement of chambers.

Yaroslav Demenchuk developed a project on the classification of plant metabarcoding data. To identify species based on short DNA fragments — the ITS region — classical ML models and convolutional neural networks were tested. The study showed that the specialized bioinformatics tool Kraken2 with a custom database proved to be more robust against noise in real sequencing data than neural network approaches.

Elizaveta Belko investigated the capabilities of neural networks for predicting cell-specific activity of K562 enhancers — DNA regions that regulate gene activity. The best hybrid model demonstrated the highest quality, and attention maps helped identify a specific GC-rich region resembling a transcription factor binding motif.

Machine learning has also been applied to problems in physics and chemistry.

Ivan Mayor used classical ML models to convert the color of thermosensitive liquid crystals into temperature values. This approach is used in aerodynamic experiments. The trained models showed accuracy higher than traditional polynomial calibration.

Yana Kiryushina worked on predicting the toxicity of molecules based on their structure. The task was to determine toxicity according to 12 biological tests at once. Two methods of representing molecules were used for this — a feature vector and a graph where atoms are vertices and chemical bonds are edges.

A Svetlana Boyarintseva developed an ML model for automatic selection of a solvent for a specific chemical reaction. A dataset of approximately 36,000 reactions with 82 molecular descriptors for each compound was collected for the project. After feature selection, the models demonstrated an accuracy of 95–96% in determining the suitability of a solvent.

“You can start from scratch”

Students note that the “Digital Department” programs are suitable not only for those who already have a technical background.

— The course, for me as someone from a completely different field, was very informative, clear, and interesting. Previously, I had no experience with machine learning, but that did not prevent me from completing the homework and successfully finishing, having tried myself in a new field., — says Ekaterina Buslavskaya.

Matvey Dyachkov also notes the accessibility of the program:

— The course is very comprehensive, interesting, and useful. It is suitable for mastering the topic “from scratch.” The instructors are responsive and explain everything very well. I gained a lot of useful knowledge and have already started applying it in practice.

For some participants, the training became directly linked to their research work.

— My thesis is related to processing tens of thousands of sequences, which is physically impossible to handle manually. Thanks to this course, I was able to automate this process and do my work effectively; currently, an article based on its results is being prepared for a high-ranking journal, — says Evgenia Lazareva.

Sofiya Setdikova particularly highlights the practical aspect of the training:

— The practice in the homework was prolonged, I had to Google, think, test hypotheses — I believe this is exactly how real skill is born. Huge thanks to the instructors for the relevant material and feedback. The course is definitely worth it!

A Ilya Rogozhin Among the main advantages is the ability to communicate directly with the teacher while completing assignments:

— In a dialogue, the teacher immediately sees if the student does not understand something, can highlight important points, explain quickly, and also asks questions that require reflection, which promotes development.

Who studies at the “Digital Department”

Among the graduates are students from all fields of study and specialties of NSU (except those related to the IT sector), as well as students from other universities.

Thus, the “Digital Department of NSU” becomes a platform where students from different specialties can acquire digital competencies and immediately try to apply them in professional and research tasks.

— Today, digital technologies are becoming a part of almost every professional field. Therefore, our task is not just to teach students to work with specific tools, but to show them how digital solutions can change familiar processes and help solve real problems. Our students’ projects clearly demonstrate how diverse these problems can be — from analyzing medical data and biological sequences to working with chemical reactions and creating digital products, — notes Marina Shashkova.

In the 2026/2027 academic year, the Digital Department programs became available to graduate students and residents.

Open until September 30 registration for the Digital Department programs at NSU for the 2026/2027 academic year