Seminars 2021/2022
22.09.2021 [online] RNA Editing Analysis: a short overview
Presenter: Alexander Fedorov, Junior Researcher
RNA editing is a method of post-transcriptional regulation of gene expression, an important mechanism of cellular immunity in eukaryotic cells. The report will describe the main “editors” of RNAs in mouse and human cells, proteins from the ADAR and APOBEC families, the current understanding of the role of A-I editing in regulating the immune response. A detailed description of the existing computational approaches to the analysis of editing at the level of individual loci, regions, and the entire genome will be given. Their main advantages and limitations, as well as typical applications, will be described.
13.10.2021 [online] Z-RNA induction as a response to viral infection
Presenter: Dmitry Konovalov, Research Assistant
Every person during his life is faced with influenza and herpes viruses. The replicating virus has been shown to generate Z-RNAs that activate ZBP1 in the nuclei of infected cells, eventually causing cell death. Research on the mechanisms of response to infection can help fight these diseases. This work presents an analysis of the expression of genes that form Z-RNA 8 and 12 hours after infection of mouse embryonic fibroblasts with influenza type A virus and herpes simplex virus type 1.
27.10.2021 [online] Clustering and description of IFNb-regulated genes
Speaker: Alexander Fedorov, Junior Researcher
Interferons (IFNs) are a group of signaling proteins with similar properties that are synthesized by cells in response to Damage Associated Molecular Patterns (DAMPs) and/or Pathogen Associated Molecular Patterns (PAMPs). According to the classical concepts, these proteins are required to stimulate cellular immunity, for example, the expression of proteins that prevent viral replication. This view, however, is not entirely correct for IFN beta (IFNb), as we were able to show using public Total RNA-seq experiments from IFNb-treated mouse fibroblasts. Specifically, after 48 hours of stimulation, almost half of all mRNA in the cell does not belong to known genes of the immune system but belongs to specific translation factors or ribosomal proteins. Only some of these factors were previously associated with the stress state of the cell or the immune response.
03.11.2021 [online] Search for conservative secondary structures in viruses of the Coronaviridae family
Presenter: Alexander Shein, Junior Researcher
The Coronaviridae virus family has drawn a lot of public attention in the last two years. Viruses from it infect large groups of vertebrates, including birds, reptiles and mammals. As humanity is experiencing the ongoing Covid pandemics, research of SARS COV 2 and it's siblings is a topic of significant importance. One of quite promising research subjects are RNA secondary structures. The discussed study includes search of conservative stem-loops in genomes of 39 different coronaviruses. One of key findings is that sequences of ORF1 a and b, ORF 7 and Surface Glycoprotein include a conservative stem-loop region. This fact suggests that these specific structures play a significant role in the virus' life cycle. Further analysis of conservative secondary structures and their functions may enrich our knowledge of host cell - virus interaction mechanisms.
17.11.2021 [online] Deep learning methods for predicting G-quadruplexes using omics data
Presenter: Latyshev Pavel, Research Assistant
Currently, there are no computational methods capable of predicting with high quality the formation of stable G-quadruplexes on chromatin. This talk proposes a method to predict the formation of G-quadruplexes using a neural LSTM network. The use of omics data from ChIP-Atlas to predict these structures will also be discussed, which gave an increase in recognition quality. In addition, a method for finding significant epigenetic features associated with the formation of G-quadruplexes will be proposed.
24.11.2021 [online] Methods of Interpretation of omics data in the tasks of Z-DNA predictions
Presenter: Nazar Beknazarov, Junior Researcher
The problem of Z-DNA prediction has been unresolved for a long time. There was an understanding that using only the DNA sequence was not enough to solve this problem. DeepZ solves the problem of Z-DNA prediction using machine learning methods and additional omix data. This approach has not only solved the problem of Z-DNA prediction, but also opens up broad prospects for further application of previously unsolved problems
08.12.2021 [online] Domain adaptation method for the annotation of mouse genome G-quadruplexes
Presenter: Latyshev Pavel, Research Assistant
The annotation of the genome by stable G-quadruplexes, which are located on the chromatin of a cell, obtained by the G4-ChIP-Seq method is only available for the human genome. In this report, a method for annotating the mouse genome with G-quadruplexes using the DANN (Domain-Adversarial Training of Neural Networks) method from the unsupervised domain adaptation class will be presented. A comparison was made between naive application of a model trained on human data on the mouse genome and a model obtained using DANN, resulting in a decrease in false positives from 61% to 49%.
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.