Bioinformatics tools for the elucidation of genomic properties

Completion of the Human Genome Project stimulated experimental research in a multitude of areas related to human genetics. This has resulted in terabytes of information being generated everyday around the globe and modern computational and bioinformatics methods are required for analysis of this data. This thesis describes several methods that were applied and/or developed which focus on potential mechanisms of regulation of gene transcription.;Biological signals can be detected if a group of DNA fragments that share the same property can be analyzed together. The CLOVER, algorithm developed by M. Frith, was applied to several potential regulatory regions. CLOVER allows detection of overrepresented motifs inside a set of nucleotide sequences. One of the objects of interest was the elucidation of DNasel hypersensitive sites which would indicate open chromatin and thus, a potential regulatory role. This study elucidated potential binding places for transcription factors in different cell types and confirmed their cell-specific regulatory functions.;DNA-protein interactions can be detected by different techniques. One such technique is based on chromatin immunoprecipitation with subsequent hybridization to DNA arrays (ChIP-chip). A procedure was developed which allows detection of this signal in a time efficient manner and inter-lab comparison showed that this procedure was among the best for the main three available commercial ChIP-chip platforms.;Lastly, this research covered another high throughput method: ChIP-sequencing. Here, H3K4me3 histone modification was analyzed in a variety of human tissues including the prefrontal cortex of the human brain. A pipeline was developed for efficient analysis of gigabytes of data generated by large-scale experiments examining different sources of samples. Additionally, it was shown how this particular histone modification varies with type of tissue, age of patient, and disease state.;This work demonstrates how bioinformatics methods can be used in the analysis of different experimental data. Furthermore, results of this analysis can be used in the development of improved therapeutics for patients with mental diseases. All results and methods described in this thesis are published and available to other researchers.