• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Second semester 2018/2019

27/06/2019 Search for patterns of association between quadruplexes and epigenetic markers

The final laboratory seminar of this academic year was devoted to the report of Arina Nostaeva (master's program "Data Analysis in Biology and Medicine").

G-quadruplexes (G4) are dynamic structures that form in G-rich single-stranded regions of DNA. Sequencing of many genomes has shown that they are rich in motifs that can form G4 and that their location correlates with functionally important regions of the genome. The aim was to study the relationship between quadruplexes and histone labels and to determine their possible biological role in epigenetic genome formation.

June 27 (Thu), 18:10-19:30
3 Kochnovsky Proezd, Aud. 435

23/05/2019 Application of convolutional neural networks for recognition of stem-loop structures, quadruplexes, Z-DNA, epigenetic code and patterns of association between secondary structures and epigenetic code

At the laboratory seminar, students of bachelor's programs of the faculty Nazar Benkazarov, Pavel Latyshev, Alexander Knyshov and Evgeny Meshcheryakov presented the results of their course projects.

May 23 (Thu), 18:10-19:30
3 Kochnovsky Proezd, Aud. 435

16/05/2019 Features of the DNA shape arising from binding to transcription factors

Artyom Stolyarenko, a laboratory Research Assistant, at the seminar spoke about his research project related to the peculiarities of the DNA shape.

A study of the parameters of the shape of the double helix of DNA in the interaction with transcription factors is carried out. For research, packages x3dna and Curves + are used. The report is devoted to the formulation of the problem and problems, a description of the tools used and the current research results.

May 5 (Thu), 18:10-19:30
3 Kochnovsky Proezd, Aud. 435

18/04/2019 G-quadruplexes: State-of-the-art detection methods and links to DNA methylation & ChIP‐Atlas: a data‐mining suite powered by full integration of public ChIP‐seq data

Dmitry Konovalov, a master's student at the Physics Faculty of Moscow State University, spoke about guanine quadruplexes - non-canonical secondary DNA structures formed from guanine-rich DNA sequences. Quadruplexes are widespread in the human genome and are associated with a number of genetic diseases. The role of many of these structures in the genome is currently unknown. There is evidence that quadruplexes are associated with epigenetic regulation. It is known that quadruplexes can bind to DNMT1. At the same time, both an increase in the number of quadruplexes upon hypomethylation and an increase in the stability of individual quadruplexes upon hypomethylation were reported. To study the role and reveal general patterns of quadruplexes, high-performance methods of their detection are needed.
The first part of the report is devoted to modern high-performance experimental (G4-seq, G4-Chip-Seq) and computer methods for detecting G-quadruplexes. The second part of the report presents the results of testing a number of hypotheses about the association of quadruplexes with human DNA methylation (existence of two classes of quadruplexes; association with tissue-specific differentially methylated regions (DMR), cell-specific DMR, inactivation of the X chromosome).

Also Dmitry Svetlichny, Leading Researcher of the laboratory, gave a practical overview of ChIP-Atlas: a data-mining suite powered by full integration of public ChIP-seq data.

April 18 (Thu), 18:10-19:30
3 Kochnovsky Proezd, Aud. 205

11/04/2019 Recognition of 3’-end L1, Alu, processed pseudogenes, and mRNA stem-loops in the human genome using sequence-based and structure-based machine-learning models

Alexander Shein, laboratory Research Assistant, spoke about his research.

The role of 3’-end stem-loops in retrotransposition was experimentally demonstrated for transposons of various species, where LINE-SINE retrotransposons share the same 3’-end sequences, containing a stem-loop.          

We have discovered that 62-68% of processed pseduogenes and mRNAs also have 3’-end stem-loops. We investigated the properties of 3’-end stem-loops of human L1s, Alus, processed pseudogenes and mRNAs that do not share the same sequences, but all have 3’-end stem-loops. We have built sequence-based and structure-based machine-learning models that are able to recognize 3’-end L1, Alu, processed pseudogene and mRNA stem-loops with high performance. The sequence-based models use only sequence information and capture compositional bias in 3’-ends. The structure-based models consider physical, chemical and geometrical properties of dinucleotides composing a stem and position-specific nucleotide content of a loop and a bulge. The most important parameters include shift, tilt, rise, and hydrophilicity. The obtained results clearly point to the existence of structural constrains for 3’-end stem- loops of L1 and Alu, which are probably important for transposition, and reveal the potential of mRNAs to be recognized by the L1 machinery. The proposed approach is applicable to a broader task of recognizing RNA (DNA) secondary structures.

April 11 (Thu), 18:10-19:30
3 Kochnovsky Proezd, Aud. 509

04/04/2019 Investigation of species-specific regulation of gene expression in the nervous tissue of primates

Abusaid Shaimardanov, student of the master's program "Data Analysis in Biology and Medicine", presented his research on species-specific regulation of gene expression.

The differences in genetic sequence between humans and chimpanzees (the closest human relative) are extremely small. Comparison of homologous protein sequences also demonstrates a slight difference between species: about 30% of the polypeptides are identical, and for most other molecules, the differences are represented by one or two amino acids per protein. Thus, it can be assumed that one of the “human-forming” factors, possibly, is a change in the regulation of genes. In particular, in our work, we consider the differences between the cis-regulatory elements of humans and two apes. In the course of the work, human-specific, as well as common regulatory elements for primates were identified. Next, we tried to find the differences between the two groups of elements, and also searched for motives de novo and using open databases.

April 4 (Thu), 16:40-18:00
3 Kochnovsky Proezd, Aud. 511

18/03/2019 Methods for reconstructing gene networks from expression data

Leading researcher of the laboratory Dmitry Svetlichny made a report at the seminar.

The current level of development of experimental genomics makes it possible to conduct a large-scale study of gene activity in various types of cells and tissues. A key practical aspect of using the information obtained is to establish the relationship between genetic data and functional properties of cells (for example, the effect on the development of a disease), which requires data processing in order to interpret the results obtained. The solution to this problem relies heavily on modern methods of machine learning and data analysis. In order to study the influence of genes, computational methods are used to reconstruct the gene network, which is a graph that describes the connections between genes. It will talk about methods for reconstructing the topology of a gene network from experimental data using a combination of machine learning and statistics methods.

March 18 (Mon), 16:40-18:00
3 Kochnovsky Proezd, Aud. 322

25/02/2019 DEEP-learning for predicting DIFFerential gene expression from histone modifications

Svetlana Shishkova, student at the Faculty of Physics of Moscow State University, at the seminar presented two methods of DeepChrome (published in Bioinformatics 2016) and DeepDiff (published in Bioinformatics 2016) using deep learning methods with a graphical representation of the input data from Chip-Seq experiments. Using DeepChrome, the level of gene expression is predicted from histone modification data, and differential gene expression in different types of tissues is predicted using DeepDiff.

February 25 (Mon), 16:00-18:00
3 Kochnovsky Proezd, Aud. 322

18/02/2019 Mammal Intron Sliding

Irina Poverennaya, guest speaker from the Faculty of Bioengineering and Bioinformatics at Moscow State University, spoke about the sliding of introns in mammals.

Introns are intragenic sections of DNA that do not contain protein sequence information. Introns are a characteristic feature of eukaryotic genes for which maturation of mRNA after transcription is characteristic, during which introns are excised from the mRNA sequence. Intron sliding is a rare evolutionary event during which an intron moves a short distance (1-15 bp). Although the slide can lead to a change in the phase of the intron, reflecting its position relative to the open reading frame, it does not affect the sequence of mature mRNA. The unclear molecular mechanism of the slide and the frequent errors in gene annotation cast doubt on the existence of such an event. In our work, we look for cases of sliding between human genes and 13 species of mammals such as chimpanzees, mice, etc., by analyzing genomic alignments and transcriptome data.

February 18 (Mon), 16:00-18:00
3 Kochnovsky Proezd, Aud. 618

11/02/2019 Recognition of prokaryotic and eukaryoticpromoters using convolutional deep learningneural networks

At the seminar, postgraduate student of the faculty Anton Zaikin, together with the audience, sorted the article "Recognition of prokaryotic and eukaryoticpromoters using convolutional deep learningneural networks" (Ramzan Kh. Umarov, Victor V. Solovyev).

Abstact. Accurate computational identification of promoters remains a challenge as these key DNA regulatory regions have variable structures composed of functional motifs that provide gene-specific initiation of transcription. In this paper we utilize Convolutional Neural Networks(CNN) to analyze sequence characteristics of prokaryotic and eukaryotic promoters and build their predictive models. We trained a similar CNN architecture on promoters of five distant organisms: human, mouse, plant (Arabidopsis), and two bacteria (Escherichia coli and Bacillus subtilis). We found that CNN trained on sigma70 subclass of Escherichia coli promoter gives an excellent classification of promoters and non-promoter sequences (Sn =0.90, Sp = 0.96, CC = 0.84). The Bacillus subtilis promoters identification CNN model achieves Sn = 0.91, Sp = 0.95, and CC = 0.86. For human, mouse and Arabidopsis promoters we employed CNNs for identification of two well-known promoter classes (TATA and non-TATA promoters). CNN models nicely recognize these complex functional regions. For human promoters Sn/Sp/CC accuracy of prediction reached 0.95/0.98/0,90 on TATA and0.90/0.98/0.89 for non-TATA promoter sequences, respectively. For Arabidopsis we observed Sn/Sp/CC 0.95/0.97/0.91 (TATA) and 0.94/0.94/0.86 (non-TATA) promoters.Thus, the developed CNN models, implemented in CNNProm program, demonstrated the ability of deep learning approach to grasp complex promoter sequence characteristics and achieve significantly higher accuracy compared to the previously developed promoter prediction programs. 

PromotersProEuk 

February 11 (Mon), 16:00-18:00
3 Kochnovsky Proezd, Aud. 219

28/01/2019 From microarrays to SGS: an overview of sequencing methods and algorithms of sequencing data analysis

Irina Ponamareva, Bioinformatics lab, spoke about her research.

January 28 (Mon), 16:00-18:00
3 Kochnovsky Proezd, Aud. 219

21/01/2019 An image representation based convolutional network for DNA classification

The first workshop of 2019 was devoted to a discussion of the article "An image representation based convolutional network for DNA classification" (Bojian Yin, Marleen Balvert, etc.).  Iinitiator and moderator of the discussion - Arina Nostaeva, 1st year undergraduate student, Bioinformatics lab

Abstact. The folding structure of the DNA molecule combined with helper molecules, also referred to as the chromatin, is highly relevant for the functional properties of DNA. The chromatin structure is largely determined by the underlying primary DNA sequence, though the interaction is not yet fully understood. In this paper we develop a convolutional neural network that takes an image-representation of primary DNA sequence as its input, and predicts key determinants of chromatin structure. The method is developed such that it is capable of detecting interactions between distal elements in the DNA sequence, which are known to be highly relevant. Our experiments show that the method outperforms several existing methods both in terms of prediction accuracy and training time.

An image representation 

January 21 (Mon), 16:00-18:00
3 Kochnovsky Proezd, Aud. 322


 

Have you spotted a typo?
Highlight it, click Ctrl+Enter and send us a message. Thank you for your help!
To be used only for spelling or punctuation mistakes.