Mathematical models of signal transduction pathways are characterized by a large number of proteins and uncertain parameters, yet only a limited amount of quantitative data is available. The dissertation addresses this problem using two different approaches: the first approach deals with a model simplification procedure for signaling pathways that reduces the model size but retains the physical interpretation of the remaining states, while the second approach deals with creating rich data sets by computing transcription factor profiles from fluorescent images of green-fluorescent-protein (GFP) reporter cells. For the first approach a model simplification procedure for signaling pathway models is presented. The technique makes use of sensitivity and observability analysis to select the retained proteins for the simplified model. The presented technique is applied to an IL-6 signaling pathway model. It is found that the model size can be significantly reduced and the simplified model is able to adequately predict the dynamics of key proteins of the signaling pathway. An approach for quantitatively determining transcription factor profiles from GFP reporter data is developed as the second major contribution of this work. The procedure analyzes fluorescent images to determine fluorescence intensity profiles using principal component analysis and K-means clustering, and then computes the transcription factor concentration from the fluorescence intensity profiles by solving an inverse problem involving a model describing transcription, translation, and activation of green fluorescent proteins. Activation profiles of the transcription factors NF-κB, nuclear STAT3, and C/EBPβ are obtained using the presented approach. The data for NF-κB is used to develop a model for TNF-α signal transduction while the data for nuclear STAT3 and C/EBPβ is used to verify the simplified IL-6 model. Finally, an approach is developed to compute the distribution of transcription factor profiles among a population of cells. This approach consists of an algorithm for identifying individual fluorescent cells from fluorescent images, and an algorithm to compute the distribution of transcription factor profiles from the fluorescence intensity distribution by solving an inverse problem. The technique is applied to experimental data to derive the distribution of NF-κB concentrations from fluorescent images of a NF-κB GFP reporter system.

Die Aufklärung der Mechanismen zur Kontrolle der Genexpression ist eines der wichtigsten Probleme der modernen Molekularbiologie. Detaillierte experimentelle Untersuchungen sind enorm aufwändig aufgrund der komplexen und kombinatorischen Wechselbeziehungen der beteiligten Moleküle. Infolgedessen sind bioinformatische Methoden unverzichtbar. Diese Dissertation stellt drei Methoden vor, die die Vorhersage der regulatorischen Elementen der Gentranskription verbessern. Der erste Ansatz findet Bindungsstellen, die von den Transkriptionsfaktoren erkannt werden. Dieser sucht statistisch überrepräsentierte kurze Motive in einer Menge von Promotersequenzen und wird erfolgreich auf das Genom der Bäckerhefe angewandt. Die Analyse der Genregulation in höheren Eukaryoten benötigt jedoch fortgeschrittenere Techniken. In verschiedenen Datenbanken liegen Hunderte von Profilen vor, die von den Transkriptionsfaktoren erkannt werden. Die Ähnlichkeit zwischen ihnen resultiert in mehrfachen Vorhersagen einer einzigen Bindestelle, was im nachhinein korrigiert werden muss. Es wird eine Methode vorgestellt, die eine Möglichkeit zur Reduktion der Anzahl von Profilen bietet, indem sie die Ähnlichkeiten zwischen ihnen identifiziert. Die komplexe Natur der Wechselbeziehung zwischen den Transkriptionsfaktoren macht jedoch die Vorhersage von Bindestellen schwierig. Auch mit einer Verringerung der zu suchenden Profile sind die Resultate der Vorhersagen noch immer stark fehlerbehafted. Die Zuhilfenahme der unabhängigen Informationsressourcen reduziert die Häufigkeit der Falschprognosen. Die dritte beschriebene Methode schlägt einen neuen Ansatz vor, die die Gen-Anotation mit der Regulierung von multiplen Transkriptionsfaktoren und den von ihnen erkannten Bindestellen assoziiert. Der Nutzen dieser Methode wird anhand von verschiedenen wohlbekannten Sätzen von Transkriptionsfaktoren demonstriert. / Understanding the mechanisms which control gene expression is one of the fundamental problems of molecular biology. Detailed experimental studies of regulation are laborious due to the complex and combinatorial nature of interactions among involved molecules. Therefore, computational techniques are used to suggest candidate mechanisms for further investigation. This thesis presents three methods improving the predictions of regulation of gene transcription. The first approach finds binding sites recognized by a transcription factor based on statistical over-representation of short motifs in a set of promoter sequences. A succesful application of this method to several gene families of yeast is shown. More advanced techniques are needed for the analysis of gene regulation in higher eukaryotes. Hundreds of profiles recognized by transcription factors are provided by libraries. Dependencies between them result in multiple predictions of the same binding sites which need later to be filtered out. The second method presented here offers a way to reduce the number of profiles by identifying similarities between them. Still, the complex nature of interaction between transcription factors makes reliable predictions of binding sites difficult. Exploiting independent sources of information reduces the false predictions rate. The third method proposes a novel approach associating gene annotations with regulation of multiple transcription factors and binding sites recognized by them. The utility of the method is demonstrated on several well-known sets of transcription factors. RNA interference provides a way of efficient down-regulation of gene expression. Difficulties in predicting efficient siRNA sequences motivated the development of a library containing siRNA sequences and related experimental details described in the literature. This library, presented in the last chapter, is publicly available at http://www.human-sirna-database.net