Research about ribonucleic acid (RNA) is gaining popularity as we widen our knowledge about its function. But to properly examine new structures, we need robust computational tools to analyse different properties. One of such tools is structural superposition, which is a method to align two structures over each other and quantify their similarity. This tool can be used on tertiary RNA structures for visual comparison, clustering or the assessment of their function. The aim of this thesis is to present a novel approach for achieving RNA superposition using information about secondary RNA structure and its link to trees. Tree edit distance algorithms are used to compare the trees, and a multitude of methods for generating the structural superposition from the calculated tree similarities is presented. The new method is aligned in the context of existing works, and its accuracy is compared to the best current approaches for structural superposition. The implementation can be accessed at https://github.com/gyfis/rawted.

Volumes of data obtained from the next generation sequencing platforms is growing faster than the available capacity of storage media. Sequencers mainly produce short reads of DNA. However, output of the sequencing machines also contains other information, for example information about read reliability/quality. This data must be archived even after successful complete genome assembly. Standard file format used for this type of data is format SAM (Sequence Alignment/Mapping Format) and its binary compressed version BAM. In this thesis we describe the construction of a better lossless compression scheme for compression of files in the SAM/BAM format. This compression scheme provides better compression ratios than the BAM format. In addition, random access to data in the compressed file is retained. Implementation of this compression scheme is platform independent and allows simple configuration of the compression process. Implementation also offers easy extensibility. Thanks to this, we will be able to respond to changes in current sequencing platforms as well as to changes in the SAM format.

The issue of comparing ribosomal RNA secondary structures is currently an open research problem. Goal of this work is to design and implement program comparing so called mutual conservancy of a group of rRNA secondary structures. This program is included as a new module of an existing system called rPredictor consisting of the structure database and an algorithm that predicts these structures. This thesis covers an introduction into the problem of secondary structures conservancy identification and summarizes known approaches towards the problem. The result of this work is a stand alone program for the secondary structure comparison and also a new analytic module of the rPredictor system, where the secondary structure comparison program can be used for comparison of the structures from the rPredictor database. Powered by TCPDF (www.tcpdf.org)

This diploma thesis deals with a development of a strategy for data evaluation generated by next-generation sequencing. Using bioinformatics tools such as Galaxy, Velvet and Enterovirus genotyping tool new aproach of data processing was optimized. There were 22 samples analyzed which of 10 were grown on cell culture. Remaining 12 were obtained from real stool samples. All samples were taken from children at the highest genetic risk of type 1 diabetes. All of them were enterovirus positive. Enteroviruses and their following infections have been suspecting to be involved in ehiology of type 1 diabetes for a long time. That's a disease resulting to an absolut insulin deficiency due to autoimmune destruction of pancreatic beta cells. Genetic components seems to be relatively well defined (the HLA, INS, STLA4, PTPN22, CTLA4, IFIH1 and numerous other genes), the environmental part of the etiology remains obscured. We were able to assemble 22 genomes de novo. However, there were numerous gaps among the particular contigs. For the first nine samples these gaps were complemented by Sanger sequencing. Nine full-length genomes were assempled this way. The main contribution of this work was to create a universal process of analyzing data from next-generation sequencing. This has already been using for further...

The broader purpose of the work is to find and interpret polymorphism in the genes of natural immunity of cattle to be used to improve disease resistance. The NGS method on the PacBio platform was applied for the resequencing of the gene for the key receptor of innate immunity TLR2 and two genes coding for the components of the downstream transcriptional factor NF-κB. In the population of 149 bulls of the Czech Simmental breed, 22 polymorphisms were found in the gene NFKB1 (5 new), while in the NFKB2 gene 13 SNP were found (10 new). 21 SNP were found in the TLR2 gene (3 new). Of the 56 found polymorphisms, 6 SNPs were nonsynonymous. One SNP leads to a change R474G in the NFKB1 product and five to changes E63D, R152Q, I211V, R563H and H665Q in the protein TLR2. Knowledge of the haplotypes facilitated the development of individual genotyping reactions. In TLR2, a high number of haplotypes was detected, both from the PacBio reads and the statistical reconstruction. In addition, two clusters of haplotypes were ditinguished inTLR2, possibly due to diversifying selection or introgression. The knowledge of genetic diversity in the population allows for the planned association studies with health data. Localization in functional domains allow to define the change with the greatest effect, in particular...

Determination of protein structure in space is a crucial part of protein function analysis. But structure determination is an expensive and time consuming pro- cess, therefore structure prediction model raised on popularity. The most notable subproblem of protein structure prediction is prediction of local conformation of the adjacent amino acids, ie. secondary structure. This thesis studies usage of deep neural networks for protein secondary structure prediction. We implemented pre- diction model and different modifications are evaluated. Especially compassion of LSTM and GRU memory cells was done. Furthermore, two new preprocessing me- thods are evaluated. Fast PSSM calculation method was proposed and prediction of tertiary structure was used as input for prediction model. Last part of this thesis examine application of filtering methods for models predicting secondary structure with eight classes. 1

DNA is the fundamental molecule in all domains of life, its role in heredity is well established. Although the famous double helical complementary form is indispensable for replication mechanism DNA can occupy wide range of conformations. In the past studies performed in the laboratory, DNA oligomers related to single stranded bacterial Repetitive Extragenic Palindromic (REP) showed spectral behavior suggesting complex equilibria including double helical, hairpin, and tetraplex conformations. The studies presented in this thesis extended the scope of analyzed sequences and employed circular dichroism spectroscopy and X-ray crystallography. We report spectral data and X-ray structures of three successfully crystalized oligonucleotides. All three structures acquire double helical architecture with two consecutive T- T mismatches in the center. To improve the convergence of the refinement process of the crystal structures we used novel dinucleotide conformational classes, NtC classes. The NtC class classification was also used to analyze geometries of selected non-canonical base pairs in all DNA crystal structures in the Protein Data Bank. We measured the fit between geometries of the dinucleotides involved in the non-canonical base pairing and the NtC classes and correlated this fit to the electron...

This diploma thesis is dealing with methods of identification genetic variability in sequencing data. The resarch is targeted to bacterial strains of Treponema pallidum. The sequencing was performed by Illumina platform. There is a proposition of method to identificate variable spots in resequenced genomes and their analysis and comparation across all processed genomes.

Analýza a interpretace variací DNA je důležité pro zkoumání genetického pozadí dědičnosti, nemocí a jiných fenotypových rysů. Tato práce stručně úvadí oblasti molekulární biologie a základních principů genetiky, popisuje metody pro anotační analýzy genetických variací, genomové asociační studie a metody pro analýzy obohacení s jejich implementací. V rámci této práce jsme představili nový webový nástroj Varanto, který může být použit k anotaci, vizualizaci a analýze genetických variací. Může být použit k analýze obohacení anotací pomocí hypergeometrického testu pro danou množinu variací. Varanto obsahuje uživatelské webové rozhraní vyvinuté pomocí frameworku Shiny jazyka R. Výkon a funkcionalita nástroje jsou testovány a demonstrovány podle výkonových benchmarků a na základě analýzy a interpretace dat z dříve publikovaných genomových asociačních studií.

Protein engineering is a young dynamic discipline with great amount of potential practical applications. However, its success is primarily based on perfect knowledge and usage of all existing information about protein function and structure. To achieve that, protein engineering is supported by plenty of bioinformatic tools and analysis. The goal of this project is to create a new tool for protein engineering that would enable researchers to identificate related proteins with modified function in still growing biological databases. The tool is designed as an automated workflow of existing bioinformatic analyses that leads to identification of proteins with the same type of enzymatic function, but with slightly modified properties - primarily in terms of selectivity, reaction speed and stability.