Mathematics and Computer Science

We introduce a novel genotype-phenotype mapping based on
the relation between RNA sequence and its secondary structure for the use in evolutionary studies. Various extensive studies concerning RNA folding in the context of neutral theory yielded insights about properties of the structure space and the mapping itself. We intend to get a better understanding of some of these properties and especially of the evolution of RNA-molecules as well as their effect on the evolution of the entire molecular system. We investigate the constitution of the neutral
network and compare our mapping with other artificial approaches using cellular automatons, random boolean networks and others also based on RNA folding. We yield the highest extent, connectivity and evolvability of the underlying neutral network. Further, we successfully apply the mapping in an existing model for the evolution of a ribozyme-catalyzed
metabolism

One important aspect of computational systems biology
includes the identification and analysis of functional response
networks within large biochemical networks. These functional
response networks represent the response of a biological system
under a particular experimental condition which can be used to
pinpoint critical biological processes.
For this purpose, we have developed a novel algorithm to calculate
response networks as scored/weighted sub-graphs spanned by
k-shortest simple (loop free) paths. The k-shortest simple path
algorithm is based on a forward/backward chaining approach
synchronized between pairs of processors. The algorithm scales
linear with the number of processors used. The algorithm
implementation is using a Linux cluster platform, MPI lam
and mpiJava messaging as well as the Java language for the
application.
The algorithm is performed on a hybrid human network consisting
of 45,041 nodes and 438,567 interactions together with
gene expression information obtained from human cell-lines
infected by influenza virus. Its response networks show the early
innate immune response and virus triggered processes within
human epithelial cells. Especially under the imminent threat of
a pandemic caused by novel influenza strains, such as the current
H1N1 strain, these analyses are crucial for a comprehensive
understanding of molecular processes during early phases of
infection. Such a systems level understanding may aid in the
identification of therapeutic markers and in drug development
for diagnosis and finally prevention of a potentially dangerous
disease.

Background: The metabolic architectures of extant organisms share many key pathways such as the citric acid
cycle, glycolysis, or the biosynthesis of most amino acids. Several competing hypotheses for the evolutionary
mechanisms that shape metabolic networks have been discussed in the literature, each of which finds support
from comparative analysis of extant genomes. Alternatively, the principles of metabolic evolution can be studied
by direct computer simulation. This requires, however, an explicit implementation of all pertinent components: a
universe of chemical reaction upon which the metabolism is built, an explicit representation of the enzymes that
implement the metabolism, of a genetic system that encodes these enzymes, and of a fitness function that can
be selected for.
Results: We describe here a simulation environment that implements all these components in a simplified ways so
that large-scale evolutionary studies are feasible. We employ an artificial chemistry that views chemical reactions as
graph rewriting operations and utilizes a toy-version of quantum chemistry to derive thermodynamic parameters.
Minimalist organisms with simple string-encoded genomes produce model ribozymes whose catalytic activity is
determined by an ad hoc mapping between their secondary structure and the transition state graphs that they
stabilize. Fitness is computed utilizing the ideas of metabolic flux analysis. We present an implementation of the
complete system and first simulation results.
Conclusions: The simulation system presented here allows coherent investigations into the evolutionary mecha-
nisms of the first steps of metabolic evolution using a self-consistent toy universe.

We developed a simulation tool for investigating the evolution
of early metabolism, allowing us to speculate on the formation
ofmetabolic pathways fromcatalyzed chemical reactions
and development of characteristic properties. Our model
consists of a protocellular entity with a simple RNA-based
genetic system and an evolving metabolism of ribozymecatalyzed
enzymes that manipulate a rich underlying chemistry.
Ensuring an almost open-ended and fairly realistic simulation
is crucial for understanding the first steps in metabolic
evolution. We show here, how our simulation tool can be
helpful in arguing for or against hypotheses on the evolution
of metabolic pathways. We demonstrate that seemingly mutually
exclusive hypotheses may well be compatible when we
take into account that different processes dominate different
phases in the evolution of a metabolic system. Our results
suggest that forward evolution shapes metabolic network in
the very early steps of evolution. In later and more complex
stages, enzyme recruitment supersedes forward evolution,
keeping a core set of pathways from the early phase.

We extend our previous work on the exploration of static metabolic networks to evolving, and therefore dynamic, pathways. We apply our visualization software to data from a simulation of early metabolism. Thereby, we show that our technique allows us to test and argue for or against different scenarios for the evolution of metabolic pathways. This supports a profound and efficient analysis of the structure and properties of the generated metabolic networks and its underlying components, while giving the user a vivid impression of the dynamics of the system. The analysis process is inspired by Ben Shneiderman’s mantra of information visualization. For the overview, user-defined diagrams give insight into topological changes of the graph as well as changes in the attribute set associated with the participating enzymes, substances and reactions. This way, “interesting features” in time as well as in space can be recognized. A linked view implementation enables the navigation into more detailed layers of perspective for in-depth analysis of individual network configurations.

Life emerged, I suggest, not simple, but complex and whole, and has remained complex and whole ever since–not because of a mysterious elan vital, but thanks to the simple, profound transformation of dead molecules into an organization by which each molecule's formation is catalyzed by some other molecule in the organization. The secret of life, the wellspring of reproduction, is not to be found in the beauty of Watson-Crick pairing, but in the achievement of collective catalytic closure.

Abstract The formation of complex systems and their properties is an intriguing field of research in biology with many unresolved questions. The knowledge that can be gained from it is not only beneficial for the understanding and optimization of existing systems but also for constructing many kinds of novel artificial systems.

Abstract In the course of evolution, biological organisms have developed certain desirable properties such as the robustness against metabolite fluctuations or mutational errors, as well as the ability to switch flexibly between functional modules. Knowledge about the emergence of these properties is beneficial both for understanding the underlying evolutionary mechanisms as well as for developing principles for the construction of artificial systems.

Clathrin mediated endocytosis (CME) is an ubiquitous cellular pathway that regulates central aspects of cell physiology such as nutrient uptake, modulation of signal transduction, synaptic transmission and membrane turn-over. Endocytic vesicle formation depends on the timed production of specific phosphoinositides and their interactions with various endocytic proteins. Recently, it has been found that phosphatidylinositol-3,4-bisphosphate (PI(3,4)P2) produced by the class II phosphatidylinositol 3-kinase C2alpha plays a key role in the recruitment of the PX-BAR domain protein SNX9, which is proposed to play a role in the constriction of the endocytic vesicle neck [Posor et al, Nature 499, p233 (2013)]. Interestingly, SNX9 and its close paralog SNX18 are not fully specific to PI(3,4)P2 but can also bind other phospholipids, in particular to PI(4,5)P2, an abundant plasma membrane lipid required for the recruitment of many endocytic proteins. In order to understand the dynamical interplay between phospholipids and endocytic proteins, we developed a computational model of the temporal changes in the population of the phosphoinositide-associated endocytic proteins and their spatial distribution at a clathrin-coated pit (CCP). The model resolves single molecules in time and space, and incorporates the complex interplay of proteins and lipids, as well as their movement within the CCP. We find that the comparably small differences in lipid binding affinities of endocytic proteins are amplified by competition among them, allowing for the selective enrichment of SNX9 at late stage CCPs as a result of timed PI(3,4)P2 production.

Die selbstständige und eigenhändige Ausfertigung versichert an Eides statt Berlin, den Unterschrift This thesis provides a description of an application environment realisation for model transforma-tion / integration based on the triple graph approach. The implementation is performed in Mathem-atica. Due to the use of functional programming in Mathematica, the resulting application shows a low abstraction from the theoretical concepts. The implementation is tested in two case studies for model transformation between class and entity-relationship diagrams. Additionally, the basis for a third case study is introduced. This case study will apply the model transformation and integration to business service and IT service models by the use of ABT-Reo diagrams.

Gene expression studies revealed a large degree of variability in gene expression patterns particularly in tissues even in genetically identical individuals. It helps to reveal the components majorly fluctuating during the disease condition. With the advent of gene expression studies many microarray studies have been conducted in prostate cancer, but the results have varied across different studies. To better understand the genetic and biological regulatory mechanisms of prostate cancer, we conducted a meta-analysis of three major pathways i.e. androgen receptor (AR), mechanistic target of rapamycin (mTOR) and Mitogen-Activated Protein Kinase (MAPK) on prostate cancer. Meta-analysis has been performed for the gene expression data for the human species that are exposed to prostate cancer. Twelve datasets comprising AR, mTOR, and MAPK pathways were taken for analysis, out of which thirteen potential biomarkers were identified through meta-analysis. These findings were compiled based u...

Testing sensor network applications is an essential and a difficult task. Due to their distributed and faulty nature, severe resource constraints, unobservable interactions, and limited human interaction, sensor networks, make monitoring and debugging of applications strenuous and more challenging.

Several ways to estimate the position of a Wireless-Sensor-Network (WSN) node have been discussed in the past years. Unlike in outdoor solutions where the Global Positioning System (GPS) could be used in most applications, a general solution for indoor usage has not been found. The few existing indoor solutions on the market are highly specialized and rely on infrastructure or on very expensive special designed hardware. Both -infrastructure and expensive hardware -does not fit well into most scenarios where WSNs are commonly used because of the adhoc characteristics and the large amount of nodes of such installations. In this paper we present a solution to get a precise estimation of the distance between two nodes without the needs for special purpose chips or a redesign of already existent nodes. We use radio runtime measurement to calculate the distance between nodes and present algorithms to refine the measurements. A comparison with a professional solution which is available on the market is also presented.

The MANIAC Challenge is a competition for cooperation strategies in wireless ad-hoc networks with the focus on experimental evaluation. We present the results of the MANIAC Challenge and discuss characteristics of real networks such as link instability and mobility, which are often simplified in common network simulators. We introduce our strategy Friendly Clustering that won the Performance Award. Friendly Clustering is based on monitoring the neighbors' forwarding behavior and assessing their willingness to relay future data packets. Further on, we describe the challenges in implementing the strategy for real networks, such as the DES-Testbed at the Freie Universität Berlin and the ad-hoc network of the MANIAC Challenge.

The Friendly Clustering approach designed for the MANIAC challenge [1] suggests a strategy for fair forwarding in mobile ad-hoc networks (MANETs). The strategy relies on the classification of neighboring nodes into cooperative and uncooperative ones by examining their forwarding behavior. Based on the importance of forwarded traffic, we assign scores to neighboring nodes and deduce sophisticated forwarding policies by calculating forwarding probabilities. The adaptive feature of the strategy detects any change of a node's forwarding behavior quickly and therefore prevents a blind trust situation. By actively sharing all gathered statistics between nodes with the same strategy, in case of topology changes due to mobility, we are able to estimate the behavior of a previously unknown node. We evaluate our approach on the DES-Testbed, a hybrid WLAN Mesh and wireless sensor network hosted at the Freie Universität Berlin, which currently consists of 30 mesh routers.

Wireless sensor networks (WSN) are in a transition from research to real world applications. Robust and efficient hardware platforms are needed. These have to offer sufficient processing power and memory while retaining energy efficiency. In this technical report we present the newest ScatterWeb hardware platform that fits the needs of research and prototyping applications of the near future. Index Terms -wireless sensor networks (WSN), embedded systems, ScatterWeb platform 1

Localization in wireless sensor networks has been a big challenge for researchers in the past years. Besides physical problems like estimating the distance between two nodes, applicable algorithms are still on the list of open research issues. While single-hop localization with direct connection to fixed anchors is well researched, the localization with fixed anchors over a multihop route is still at its beginning. Especially the combination of multi-hop networks and mobile nodes needs further research. In this paper, we present and discuss a refined algorithm and a simulation-based approach regarding the mentioned scenario. Using a five phase structure that pursues a greedy approach, including a refining anchor selection, we investigate and discuss the precision of localization in a mobile environment. To approximate the distance between node and anchor over a multi-hop route, we make use of the mentioned greedy algorithm which fits best for processing multiple distance measurements between nodes. Furthermore, we evaluate different simulation-based experiments with mobile nodes and multi-hop routes.

The rapid spread of different social media applications provides a new way for people to interact and to share information on-line all over the world. This massive and enormous volume of on-line information needs to be structured and organized in a useful way for users to get oriented opinion summary from text related to their search. In this paper we consider the problem of sentiment classification of on-line product features to obtain their overall semantic orientation. We present a novel method for extracting product features and to obtain their associated sematic orientations. Our proposed method is based on multi-agent technology with emerging machine learning classification techniques that together promise a better level of efficiency.

Multi-valued network models can be described by their topology and a set of parameters capturing the effects of the regulators for each component. Dynamics can then be derived and represented as state transition systems. Different network models may lead to the same transition system, meaning dynamics analysis of a representative model covers a larger class of models. While rather clear in the Boolean case, the properties contributing to this effect become more involved for multi-valued models. We analyze these properties and present a mathematical description of the resulting model equivalence classes.