ITMO: Shiny Gatom Cellular Metabolism Analysis Service

Main article: Metabolism

2022: Developing a Service for Cell Metabolism Analysis

Bioinformatics ITMO developed a service for analyzing cellular metabolism - biochemical reactions responsible for the vital activity of cells - Shiny GATOM. Unlike analogues, the proposed tool considers cellular processes at the deepest level - not only at the level of substances and, genes but also at the level of atoms. This was reported by the press service of ITMO University on June 14, 2022. This facilitates the interpretation of the results. The service will be useful for solving problems biology in and - to medicine for example, it can help in the development drugs against autoimmune diseases and. cancer

Metabolism in cells plays an important role in the regulation of many biological processes, including the work of the immune system. In-depth study of metabolism helps to better understand how autoimmune diseases occur in living organisms or develop cancer, as well as to create other treatments for these diseases. However, existing solutions allow you to disassemble only a limited set of standard chemical reactions. For non-trivial tasks - for example, to analyze failures and "breakdowns" in cellular metabolism - more sophisticated software products are needed. In addition, tools are needed that can provide detailed information not only about the metabolic pathway as a whole, but also about its individual sites.

Work on this solution began back in 2016 - then scientists from ITMO, together with foreign colleagues, managed to develop a web service that detects links between changes in metabolism and genes. Now this is a full-fledged program that can be installed on a computer.

The principle of working with the program is quite simple. The researcher loads into it a table with data on metabolites (simple low molecular weight substances involved in metabolism) and gene activity, for example, a cancer cell. The program compares this data with the KEGG and Rhea databases, which describe most biological processes in the standard state. The result is given in the form of a certain map, a graph, where the path of transformation of substances is clearly presented and the connections between them are visible. The vertices of this graph are substances; lines between them ('edges') - reactions. Moreover, the algorithm itself highlights in color those metabolites and genes encoding them that need to be paid attention to. It all takes about a minute, told Alexey Sergushichev, Director of the Scientific and Educational Center for Genomic Diversity and Head of the Frontier Laboratory "Computational Methods for Systems Biology" at ITMO.

Over the past few years, the functionality of the service has expanded - it now knows how to analyze biochemical reactions not only at the level of substances, but also at the level of atoms. This allows you to better structure the graph and facilitates the interpretation of the results. However, in a graph with transformations of individual atoms, the search for subgraphs with the most pronounced changes becomes more complicated, because atoms can occur in several substances at once, which means that the number of reactions under consideration increases several times. Scientists managed to solve this problem - the service itself removes unnecessary repetitions and leaves only "edges" and vertices with positive weight, that is, a high indicator of significance.

Another update is the ability to work with lipidoma data (information about fats and their derivatives). By studying them, you can, for example, find out how different parts of the brain are arranged, where the variety of fat composition plays a large role. Lipidomics is an actively developing area, and there are practically no programs that can analyze metabolic lipid processes. However, reactions with lipids are already well described in the Rhea database - now it is also part of the GATOM algorithm, explained Maria Emelyanova, first author of the article, programmer of the ITMO Scientific and Educational Center for Genomic Diversity.

Thanks to the computational method developed, the authors were already able to link the development of Alzheimer's disease with a "breakdown" in the TREM2 gene and showed that it is possible to reduce the tumor growth rate by slowing down certain metabolic processes in cancer cells.