The cell cycle genes homology region (CHR) has been identified as a DNA element with an important role in transcriptional regulation of late cell cycle genes. It has been shown that such genes are controlled by DREAM, MMB and FOXM1-MuvB and that these protein complexes can contact DNA via CHR sites. However, it has not been elucidated which sequence variations of the canonical CHR are functional and how frequent CHR-based regulation is utilized in mammalian genomes. Here, we define the spectrum of functional CHR elements. As the basis for a computational meta-analysis, we identify new CHR sequences and compile phylogenetic motif conservation as well as genome-wide protein-DNA binding and gene expression data. We identify CHR elements in most late cell cycle genes binding DREAM, MMB, or FOXM1-MuvB. In contrast, Myb- and forkhead-binding sites are underrepresented in both early and late cell cycle genes. Our findings support a general mechanism: sequential binding of DREAM, MMB and FOXM1-MuvB complexes to late cell cycle genes requires CHR elements. Taken together, we define the group of CHR-regulated genes in mammalian genomes and provide evidence that the CHR is the central promoter element in transcriptional regulation of late cell cycle genes by DREAM, MMB and FOXM1-MuvB.

Modern humans originated in Africa around 200,000 years ago and today have settled in nearly every corner of earth. During migrations humans became exposed to new pathogens, food sources and have encountered vastly different environments. Natural selection likely contributed to the survival under such diverse conditions by promoting the raise in frequency of advantageous alleles. Thereby natural selection leaves genetic footprints that we can identify. The thesis at hand is about understanding how natural selection has shaped different human populations by analyzing these genetic footprints. In the first study, I infer the evolutionary history of an insertion-substitution variant using present-day human genomic data. This variant is interesting because the ancestral allele encodes a previously unannotated open-reading frame for a gene with antiviral ac- tivity (IFNL4 ), while the derived allele truncates this open-reading frame and is strongly associated with improved clearance of Hepatitis C, a major health care problem. Using an approximate bayesian computation approach I infer a complex evolutionary history, where the derived, truncating allele evolved under weak positive selection in Africa, with selection strength increasing in non-African populations, especially in East Asian popu- lations where the truncating allele is nearly fixed today. Hence, the changes in selection and resulting population differences in allele frequency contribute to the variation in Hep- atitis C clearance observed across human populations today. In the second study, I use ancient human genomes to estimate genome-wide allele frequencies in the past to understand present-day population differentiation. I develop a new statistic and incorporate the genome of Ust’-Ishim, a modern human that lived 45,000 year ago in Siberia, to study to what extent natural selection and drift have contributed to human population differentiation. The results suggest that European populations carry high frequency alleles in protein-coding (genic) regions that evolved under strong, recent positive selection. Further, the genic alleles that rose in frequency recently in Europeans were already present in ancient hunter-gatherers more often than in ancient farmers. This suggests that during the colonization of Europe local, positive selection changed the frequency of advantageous alleles in hunter-gatherer populations prior to the influx of farming individuals and those alleles remained beneficial also in the later admixed populations.

Understanding the evolutionary basis for human-specific phenotypes such as complex speech and language, advanced cognition or the unique preparation of their food is a topic of broad interest. Approaches focusing on comparisons of the genomic DNA (deoxyribonucleic acid) or RNA (ribonucleic acid) sequence between species, individuals or tissues allow for the identification of evolutionary sequence changes, some of these changes may underlie differences in phenotypes. In addition, differences in when, where and how much of a particular gene is present may also contribute to functional changes and therefore also to phenotypic differences. The resources to make such comparisons using genetic data are now available. The genome sequences of a number of outgroups: all living great apes, as well two archaic humans, are now publically available. Studying gene expression on the RNA level - a precursor of the protein expression - is considerably easier and cheaper than the measurement of expression of the protein itself. It has been shown that the RNA and protein expression levels are well correlated and therefore measuring RNA levels provides a good proxy for the expression of the protein. Using high-throughput sequencing techniques, relatively unbiased expression comparison is now possible because the RNA from any species can be sequenced directly, rather than being captured on arrays which are designed based on a particular reference sequence. The aim of this research was to use gene expression as a molecular phenotype to identify changes relevant to human-specific biology and study the difference between humans and their closest living relatives to understand patterns and differences in the gene expression and in gene expression regulation in multiple tissues in primates using high-throughput sequencing techniques. In my thesis, I describe two analyses to address open questions in the field of gene expression and genes expression regulation in humans. In the first part I will analyze how the effect of different diets impact gene expression using a mouse model. Two key components of the human diet that differ substantially from the diet of other primates, the frequent use of meat of many humans and the cooking of their food which is common for almost all human populations, are modeled in the experiment. I tested for their impact on liver gene expression. I found that both the differences in food substrates - meat and tuber - as well as in their preparation affect gene expression in mice significantly. The effect is bigger between food substrates than between methods of preparation. Differentially expressed genes between food substrates and food preparation were predominantly related to metabolic functions. In addition, immune-genes showed differential expression between the comparisons of raw meat to both, raw tuber and cooked meat, respectively. The results indicate that different food substrates and food preparations activate different metabolic pathways and that the cooking of food and particularly of meat has an influence on the immune also changes immune-reactions of the body. I showed that expression differences in these mice are correlated with the differences observed between humans and other primates, and that there is evidence that adaptation to these diets dates to more than 300.000 years. Finally, I showed that transcription factors play in important role in regulation of gene expression with respect to different food preparation. In the second part I analyzed the expression of one key regulator of gene expression: microRNAs (miRNAs). Using miRNA expression data from multiple primate species and for multiple tissues I found that expression differences vary between tissues. While heart and brain show only few expression differences between primates, other tissues are more variable in expression. The most extreme expression differences in all three primate species were found in the brain, which may reflect the importance of miRNAs in the regulation of gene expression in the brain. Expression differences in testis were significantly larger between humans and macaques than between chimpanzees and macaques, indicating that miRNAs evolved differently in human compared to chimpanzees. MiRNA expression differences were correlated with expression differences of their target genes genome-wide which underlines the regulatory importance of miRNAs. I also showed that differentially expressed miRNAs between species/tissues preferentially targeted transcription factors, which are important gene expression regulators as well. This finding that suggests complex regulatory pathways involving both miRNAs and transcription factors in the control of gene expression. Finally, I used the miRNA sequencing data to annotate new miRNAs in primates and was able to increase the number of annotated miRNAs substantially, especially for the non-human primates which were previously not extensively annotated. The overlap of miRNAs annotated in multiple primate species thereby also increased which will support future studies to investigate the evolutionary changes of miRNAs between these primates.

Screening of a deeply sequenced transcriptome using Illumina sequencing as well as the genome of the tardigrade Hypsibius dujardini revealed a set of five opsin genes. To clarify the phylogenetic position of these genes and to elucidate the evolutionary history of opsins in Panarthropoda (Onychophora + Tardigrada + Arthropoda), we reconstructed the phylogeny of broadly sampled metazoan opsin genes using maximum likelihood and Bayesian inference methods in conjunction with carefully selected substitution models. According to our findings, the opsin repertoire of H. dujardini comprises representatives of all three major bilaterian opsin clades, including one r-opsin, three c-opsins, and a Group 4 opsin (neuropsin/opsin-5). The identification of the tardigrade ortholog of neuropsin/opsin-5 is the first record of this opsin type in a protostome, but our screening of available metazoan genomes revealed that it is also present in other protostomes. Our opsin phylogeny further suggests that two r-opsins, including an "arthropsin", were present in the last common ancestor of Panarthropoda. While both r-opsin lineages were retained in Onychophora and Arthropoda, the "arthropsin" was lost in Tardigrada. The single (most likely visual) r-opsin found in H. dujardini supports the hypothesis of monochromatic vision in the panarthropod ancestor, whereas two duplications of the ancestral panarthropod c-opsin have led to three c-opsins in tardigrades. Although the early-branching nodes are unstable within the metazoans, our findings suggest that the last common ancestor of Bilateria possessed six opsins: two r-opsins, one c-opsin, and three Group 4 opsins, one of which (Go opsin) was lost in the ecdysozoan lineage.

Background: Imputation of partially missing or unobserved genotypes is an indispensable tool for SNP data analyses. However, research and understanding of the impact of initial SNP-data quality control on imputation results is still limited. In this paper, we aim to evaluate the effect of different strategies of pre-imputation quality filtering on the performance of the widely used imputation algorithms MaCH and IMPUTE. Results: We considered three scenarios: imputation of partially missing genotypes with usage of an external reference panel, without usage of an external reference panel, as well as imputation of ompletely un-typed SNPs using an external reference panel. We first created various datasets applying different SNP quality filters and masking certain percentages of randomly selected high-quality SNPs. We imputed these SNPs and compared the results between the different filtering scenarios by using established and newly proposed measures of imputation quality. While the established measures assess certainty of imputation results, our newly proposed measures focus on the agreement with true genotypes. These measures showed that pre-imputation SNP-filtering might be detrimental regarding imputation quality. Moreover, the strongest drivers of imputation quality were in general the burden of missingness and the number of SNPs used for imputation. We also found that using a reference panel always improves imputation quality of partially missing genotypes. MaCH performed slightly better than IMPUTE2 in most of our scenarios. Again, these results were more pronounced when using our newly defined measures of imputation quality. Conclusion: Even a moderate filtering has a detrimental effect on the imputation quality. Therefore little or no SNP filtering prior to imputation appears to be the best strategy for imputing small to moderately sized datasets. Our results also showed that for these datasets, MaCH performs slightly better than IMPUTE2 in most scenarios at the cost of increased computing time.

Motivation: Although various high-throughput technologies provide a lot of valuable information, each of them is giving an insight into different aspects of cellular activity and each has its own limitations. Thus, a complete and systematic understanding of the cellular machinery can be achieved only by a combined analysis of results coming from different approaches. However, methods and tools for integration and analysis of heterogenous biological data still have to be developed. Results: This work presents systemic analysis of basic cellular processes, i.e. cell viability and cell cycle, as well as embryonic stem cell pluripotency and differentiation. These phenomena were studied using several high-throughput technologies, whose combined results were analysed with existing and novel clustering and hit selection algorithms. This thesis also introduces two novel data management and data analysis tools. The first, called DSViewer, is a database application designed for integrating and querying results coming from various genome-wide experiments. The second, named PhenoFam, is an application performing gene set enrichment analysis by employing structural and functional information on families of protein domains as annotation terms. Both programs are accessible through a web interface. Conclusions: Eventually, investigations presented in this work provide the research community with novel and markedly improved repertoire of computational tools and methods that facilitate the systematic analysis of accumulated information obtained from high-throughput studies into novel biological insights.

Background: In the life sciences, the amount of literature and experimental data grows at a tremendous rate. In order to effectively access and integrate these data, biomedical ontologies – controlled, hierarchical vocabularies – are being developed. Creating and maintaining such ontologies is a difficult, labour-intensive, manual process. Many computational methods which can support ontology construction have been proposed in the past. However, good, validated systems are largely missing. Motivation: The biocuration community plays a central role in the development of ontologies. Any method that can support their efforts has the potential to have a huge impact in the life sciences. Recently, a number of semantic search engines were created that make use of biomedical ontologies for document retrieval. To transfer the technology to other knowledge domains, suitable ontologies need to be created. One area where ontologies may prove particularly useful is the search for alternative methods to animal testing, an area where comprehensive search is of special interest to determine the availability or unavailability of alternative methods. Results: The Dresden Ontology Generator for Directed Acyclic Graphs (DOG4DAG) developed in this thesis is a system which supports the creation and extension of ontologies by semi-automatically generating terms, definitions, and parent-child relations from text in PubMed, the web, and PDF repositories. The system is seamlessly integrated into OBO-Edit and Protégé, two widely used ontology editors in the life sciences. DOG4DAG generates terms by identifying statistically significant noun-phrases in text. For definitions and parent-child relations it employs pattern-based web searches. Each generation step has been systematically evaluated using manually validated benchmarks. The term generation leads to high quality terms also found in manually created ontologies. Definitions can be retrieved for up to 78% of terms, child ancestor relations for up to 54%. No other validated system exists that achieves comparable results. To improve the search for information on alternative methods to animal testing an ontology has been developed that contains 17,151 terms of which 10% were newly created and 90% were re-used from existing resources. This ontology is the core of Go3R, the first semantic search engine in this field. When a user performs a search query with Go3R, the search engine expands this request using the structure and terminology of the ontology. The machine classification employed in Go3R is capable of distinguishing documents related to alternative methods from those which are not with an F-measure of 90% on a manual benchmark. Approximately 200,000 of the 19 million documents listed in PubMed were identified as relevant, either because a specific term was contained or due to the automatic classification. The Go3R search engine is available on-line under www.Go3R.org.