|
Cytokine networks and their impact on the immune profile of acute myeloid leukemia (AML) patients
Ptáček, Antonín ; Musil, Jan (advisor) ; Fišer, Karel (referee)
Acute myeloid leukemia (AML) is a malignant hematological disorder characterized by aberrant expansion of blasts in the bone marrow and peripheral blood. The immune system protects the body from leukemia by eliminating transformed cells. However, in AML, the abilities of immune cells are affected both by direct contact between leukemic cells and effector cells, as well as by cytokines, metabolites and other soluble proteins that, together with the cells, form the specific AML microenvironment. The effects of cytokines and other soluble molecules in the AML microenvironment are not sufficiently described yet. This thesis aimed to optimize and implement a multiparametric flow cytometry panel for the measurement of cell populations and to implement multiparametric assays for the analysis of cytokine levels, chemokines and other soluble proteins in plasma. The following goal was to use these methods to characterize the frequency and functional phenotype of cell populations and the levels of the soluble proteins and to describe their influence on disease severity and overall survival of the patients. We also tried to find novel biomarkers of the immune escape of leukemic cells. In patients, we observed a suppressive microenvironment with aberrant levels of soluble receptors and other proteins. This...
|
|
Gene expression analysis on a subgene level
Kloda, František ; Fišer, Karel (advisor) ; Novotný, Marian (referee)
RNA sequencing allows investigation of expression of singular genes in cells. It is possible to interpret the arisen data on multiple levels, each level providing a different type of information. Apart from measuring expression of whole genes, it is possible to quantify expression of singular exons, or transcripts (gene isoforms), which allows more detailed study of regulatory mechanisms. The main difference between approaches is in determining the origin of short reads. This step is significantly more complex in analysis of expression of transcripts, as transcripts derived from the same gene have typically larger rate of sequential similarity. In this thesis, we describe eleven tools for subgene level expression analysis a as comparison we have tested three of these tools on real patient data. The results provided by all three tools proved to be very similar with the greatest difference being the time needed for the analysis.
|
|
Disruption of transcription regulatory networks in childhood Acute lymphoblastic leukemia
Sadílek, Vojen ; Fišer, Karel (advisor) ; Frič, Jan (referee)
Transcription factor aberrations are the initiating lesion in a majority of pediatric Acute Lymphoblastic Leukemia (ALL) cases, often acquired in utero. These mutations play a crucial role in determining the disease phenotype and prognosis. This work advocates for a broad evaluation of the transcriptomic landscape from the viewpoint of transcription factors and their targets. We adapted SCENIC (Aibar et al., 2017), a tool for single-cell gene regulatory network inference, for an extensive dataset of 633 bulk RNA sequencing samples of pediatric ALL. This allows us to evaluate the gene expression profile as a result of transcription factor activity. Furthermore, we aimed to deepen our understanding of valuable preexisting data by incorporating information from several "-omics" databases. We utilized resources of single cell RNA sequencing data, protein-protein interactions and putative cis-regulatory elements in our analysis. Our approach covers the transcriptomic landscape without an overt bias for differentially expressed or highly variable genes. We demonstrate that most B-ALL subtypes have specific patterns of transcription factor activity and, for subtypes initiated by transcription factor mutations, we examine the influence of such aberrations on their activity. With MEF2D as an example, we...
|
|
A modern implementation of LALR(1) parser generator
Fišer, Karel ; Bednárek, David (advisor) ; Hnětynka, Petr (referee)
The goal of this thesis is to design and implement a modern parser generator. The result is a program that reads description of some context-free LALR(1) grammar and semantic actions from an input file. To output files the program generates source codes in several target modern object-oriented programming languages for implementation of the syntax analyzer which, when parsing the language corresponding to the given grammar, executes the given semantic actions. Powered by TCPDF (www.tcpdf.org)
|
|
Bombic 2
Fišer, Karel ; Lidický, Bernard (advisor) ; Jelínková, Eva (referee)
Title: Bombic 2 Author: Karel Fišer Department: Department of Applied Mathematics Supervisor: RNDr. Bernard Lidický Supervisor's E-mail address: bernard@kam.mff.cuni.cz Abstract: The aim was to analyze the shortcomings of the first version of the game Bombic and to desing game and map editor removes these deficiencies. This concept then implement and create a programming and user documentation. As a result of the work originated programming piece, which consists of two programs. In the chapters of this work I describe in details the shortcomings of the first version, design of data storage and architecture of both programs. In chapter 6. I bring the user documentation. Read here how to install the programs, how to set up and play the game. I'm describing how to work with the map editor and how to create a new map. Keywords: Bombic, side-scrolling, arcade game, deathmatch
|
|
Sources, identification and removal of ChIP-seq artifacts
Shumilova, Aleksandra ; Převorovský, Martin (advisor) ; Fišer, Karel (referee)
Chromatin immunoprecipitation is used to enrich DNA sequences that are associ- ated with a protein of interest, and is used to map those sequences to the genomic regions. Studying these DNA-protein binding regions provides an understanding of gene regulation and chromatin remodeling. However, some signals in fact rep- resent no binding event and are known as false positives. This thesis discusses the main sources of false-positive signals that commonly arise during ChIP-seq analysis, and offers possible solutions on how to minimize or filter them. Keywords: ChIP-seq, chromatin imunoprecipitation, quality control, data filtra- tion iii
|
| |
|
Leukaemia associated immunophenotype in childhood acute leukaemias and its development during the course of disease
Podolská, Tereza ; Fišer, Karel (advisor) ; Drbal, Karel (referee)
Acute lymphoblastic leukaemia (ALL) is the most frequent childhood malignancy. One of the recent improvements in ALL treatment was the introduction of minimal residual disease (MRD) monitoring that enables risk stratification based treatment adaptation. The same MRD monitoring helps to choose relapse treatment, to guide indication for stem cell transplantation (SCT) and allows for a more personalized management of patients undergoing SCT. One of the main routes of MRD levels detection is characterisation of leukemic blasts using flow cytometry. However, flow cytometry is limited by its mainly manual expertise-based analysis. Such analysis is subjective and clearly insufficient for current complex data. While new computational tools are available for multidimensional flow cytometry data, there is an urgent need to test and adapt them for the use in clinical environment. The goal of this thesis is to detect immunophenotypes associated with leukaemia and their development by leveraging machine-assisted analysis of a set of diagnostic files selected based on information about more than three hundred thousand of multiparameter flow cytometry datasets. Advanced bioinformatic tools will help to detect blast and healthy haematopoietic populations, to derive their immunophenotypes and to identify individual...
|
|
Ambiguous representation of genetic variants in the VCF format
Neitzert, Gesa-Maret ; Daněček, Petr (advisor) ; Fišer, Karel (referee)
The variant call format (VCF) is a file format used to represent and store information about DNA variation. Genetic variants in VCF can be represented in multiple ways because the VCF specification allows for ambiguity, which can arise because of different variant calling pipelines or differences in sequence alignment. Ambiguities interfere with the comparison of VCF files and the variants therein, leading to complications in further analysis of variants. This thesis explores the differences in the representation of genetic variants that can occur, as well as their causes and impacts on further analysis. Furthermore, the normalization of VCF files is addressed and an algorithm for the atomization and deatomization of VCF files is shown. Keywords: VCF, variant call format, ambiguous variant representation, variant comparison, variant atomization, variant deatomization
|
| |