Datenbestand vom 13. Juni 2019

Warenkorb Datenschutzhinweis Dissertationsdruck Dissertationsverlag Institutsreihen     Preisrechner

WICHTIGER HINWEIS
DER VERLAG IST IN DER ZEIT VOM 12.06.2019 BIS 23.06.2019 AUSCHLIESSLICH PER EMAIL ERREICHBAR.

aktualisiert am 13. Juni 2019

ISBN 9783868538823

Euro 72,00 inkl. 7% MwSt


978-3-86853-882-3, Reihe Informatik

Sebastian Fröhler
Algorithms for the analysis of the primary and tertiary structure of genomes.

164 Seiten, Dissertation Eberhard-Karls-Universität Tübingen (2011), Softcover, A5

Zusammenfassung / Abstract

The development of an organism is a complex and complicated process, being tightly regulated by several mechanisms. A lot of regulatory events take place at the level of genome structure.

In this thesis, we present approaches to analyze the primary- as well as the tertiary structure of genomes.

Here, we present a method to screen for polymorphisms in the genomic sequence (the primary structure) of an organism. Our approach exploits quality information from both: the reference genomic sequence as well as from a set of sequences deter- mined using the latest sequencing technology. By assessing this quality information using a Bayesian statistical formalism, our method is able to accurately identify real polymorphisms while reliably rejecting low-quality polymorphisms.

We also present a method integrating polymorphism data with gene expression data in order to predict the phenotypic behavior of an organism in the presence of small molecules (drugs) supplied. This is achieved by using probabilistic graphical models extended by Gaussian Processes, thereby enabling our method to detect non- linear statistical dependencies as can be found in many biological gene regulation networks.

We finally present a method to support experimentalists in studying the tertiary structure of genomes by detecting and quantifying long-ranging DNA-DNA inter- actions. This kind of interactions, in many cases, represent transient regulatory interactions between dispersed genomic loci but may also be simple random col- lisions. Our method aims to generate an optimized experiment setup to reliably discern these two kinds of contacts.

We give detailed descriptions of each method and perform extensive evaluations to assess the accuracy of each method.