Work

Statistische Analyse von Sequenzpopulationen in der Virologie und Immunologie

Year: 2017
Type: dissertation
Abstract: In this thesis I have examined various topics regarding the relationship between viruses and the human immune system. I expanded and refined a tool (which can now be found as R-package SeqFeatR on C-RAN) for the analysis of sequence data and features of this sequences like HLA type or tropism (see chapter 4) and checked with this tool if there are differences between some multiple correction approaches for sequence data, and how Bayesian inference could be used in this context (see chapter 5). It could be shown that Bayesian inference is superior to the frequentistic methods for this kind of problem, because multiple correction approaches ignore the fact that different positions in a sequence alignment may be connected in the protein product of this sequence and are therefor not independent. Furthermore, I have examined sequences from HCV with a form of bootstrap algorithm to find sequence areas which can be used in unknown transmission cases in court. Two areas were found, one in the hypervariable region and the other at the end of the non-structural protein NS5B (see chapter 9). Proteasomal cleavage of alien amino acid sequences inside human cells leads to a presentation of fragments of these sequences on the surface of the cell as epitopes. To present such a fragment, not only must it bind to the MHC, but also needs to be in the correct length to be presented. Therefore viral evolution should favor those viruses, which cannot be cut into presentable epitopes. With epitope data from IEDB and predicted viral sequences which bind the MHC, I searched for amino acids inside the flanking regions around the epitope that may indicate a possible escape mutation against the proteasomal cleavage processes. Fourteen such amino acids and positions were found (see chapter 7). I created a model of HBV reverse transcriptase to check if mutations in certain positions could influence binding with the nucleotide analogue reverse transcriptase inhibitor Tenofovir. Mutations which were inside the binding pocket for Tenofovir showed, in an experimental design by the group of Mengji Lu, a decreased affinity towards the drug (see chapter 10). Together with Ralf Küppers group I examined NGS from different types of B cells to search for almost identical sequences between those. We found similar to identical sequences from two, three and even four kinds of cells in the blood samples of both donors (see chapter 6).
Language: English
Open Access status: green
OpenAlex ID: W7034122985
Cites: 449
Cited by:
Related to: 9