No abstract

Living cells are constantly subjected to a plethora of environmental stimuli that require integration into an appropriate cellular response. This integration takes place through signal transduction events that form tightly interconnected networks. The understanding of these networks requires capturing their dynamics through computational support and models. ANIMO (analysis of Networks with Interactive Modeling) is a tool that enables the construction and exploration of executable models of biological networks, helping to derive hypotheses and to plan wet-lab experiments. The tool is based on the formalism of Timed Automata, which can be analyzed via the UPPAAL model checker. Thanks to Timed Automata, we can provide a formal semantics for the domain-specific language used to represent signaling networks. This enforces precision and uniformity in the definition of signaling pathways, contributing to the integration of isolated signaling events into complex network models. We propose an approach to discretization of reaction kinetics that allows us to efficiently use UPPAAL as the computational engine to explore the dynamic behavior of the network of interest. A user-friendly interface hides the use of Timed Automata from the user, while keeping the expressive power intact. Abstraction to single-parameter kinetics speeds up construction of models that remain faithful enough to provide meaningful insight. The resulting dynamic behavior of the network components is displayed graphically, allowing for an intuitive and interactive modeling experience.

One of the core problems in soft computing is dealing with uncertainty in data. In this paper, we revisit the formal foundation of a class of probabilistic databases with the purpose to (1) obtain data model independence, (2) separate metadata on uncertainty and probabilities from the raw data, (3) better understand aggregation, and (4) create more opportunities for optimization. The paper presents the formal framework and validates data model independence by showing how to a obtain probabilistic Datalog as well as a probabilistic relational algebra by applying the framework to their non-probabilistic counterparts. We conclude with a discussion on the latter three goals.

A bioinformatician has a large number of homology data sources to choose from. These data sources need to be combined before a query can be posed over the combined data. We propose a generic probabilistic approach to combining grouping data from multiple sources. Our approach incorporates an iteratively evolving view on trust, allowing the bioinformatician to express his fine-grained view on how much the data in the sources can be trusted. We evaluate our approach by combining 3 real-world biological databases and show that it scales well for realistic amounts of data and uncertainty.

No abstract

No abstract

No abstract