Research group

Bioinformatics Group

The bioinformatics group is dedicated to a development of efficient computational methods for important real-world biological and medical problems. The projects cover diverse directions starting from analysing metagenomic sequencing data, to gene expression analysis and metabolomics.

The group is based at Computer Technologies Department at ITMO University and supported by JetBrains. The group is actively collaborating with Maxim Artyomov’s laboratory at Washington University in St. Louis and Dmitry Alexeev’s laboratory at MIPT.

Main research projects

Algorithms for comparative metagenomics

Metagenomic projects usually deal with a lot of data: up to several hundred gigabytes of sequencing data per project. To be able to effectively analyse such amounts of data and to do a comparative analysis, a software called MetaFast was developed. It implements an algorithm of light-weight assembly which lies between traditional k-mer spectrum analysis and full metagenomic assembly. The algorithm allows MetaFast to combine advantages of the both methods: high computational performance with more informative and better interpretable extracted features, all without using reference genome sequences.

Analysis of open gene expression databases

Currently, there are several big databases, like TCGA or GEO Omnibus, that store data of a large number of experiments. Analysing these experiments and finding hidden interrelations is a very promising field. As a part of the project a web-service GeneQuery is being developed, which allows to find experiments with similar patterns of gene regulation. Using results of such search one can discover unobvious connections, which can facilitate hypothesis generation.

Computational methods for studying metabolic regulation

Recently a role metabolic regulation got recognized as one of the immune system and cancer hallmarks. In this project a number of methods are being developed to analyse gene expression and metabolomics data to study such regulation. Currently, some of the methods are available as a web-service GAM for integrative network analysis of gene expression and metabolomics profiling data.