Bioinformatics Analyses of Alternative Splicing: Non-EST Based Prediction, Influence of Secondary Structures and Tandem Splice Sites

This thesis investigates three aspects of alternative splicing by means of computational large-scale analyses. In the first part, we introduce a new approach for the ab initio prediction of alternative splice events. We introduce an efficient algorithm to reduce the computational complexity during the search for new splice events. Applying this algorithm to the human genome, we predict and verify novel splice events. In the second part, we investigate the influence of mRNA secondary structures on the regulation of the splicing process. We show that experimentally verified binding sites of splicing regulatory proteins have a higher single-strandedness. Then, we develop a new motif finding method that benefits from taking the single-strandedness of motif occurrences into account. In the third part, we analyze a group of splice events that occur at tandem splice sites and result in minor changes of the mRNA and the protein. Genome-wide analyses provide evidence for a non-random distribution of these splice events, for tissue-specific regulation, and for evolutionary conservation. We conclude that these splice events represent one major mechanism to increase the proteome diversity.