Refine
Year of publication
Document Type
- Article (29) (remove)
Keywords
- Access control (1)
- Arabidopsis (1)
- August 2002 flood (1)
- Brassica napus (1)
- Central Europe (1)
- Data federation (1)
- Floods Directive (1)
- Genotypic differences (1)
- Information federation (1)
- Information integration (1)
Institute
- Institut für Biochemie und Biologie (7)
- Institut für Informatik und Computational Science (6)
- Institut für Chemie (4)
- Institut für Geowissenschaften (4)
- Institut für Umweltwissenschaften und Geographie (3)
- Fachgruppe Volkswirtschaftslehre (1)
- Institut für Mathematik (1)
- Institut für Physik und Astronomie (1)
- MenschenRechtsZentrum (1)
- Sozialwissenschaften (1)
Jede Software ist "Social"
(2010)
In this article we report about a successful application of modern machine learning technology, namely Support Vector Machines, to the problem of assessing the 'drug-likeness' of a chemical from a given set of descriptors of the Substance. We were able to drastically improve the recent result by Byvatov et al. (2003) on this task and achieved an error rate of about 7% on unseen compounds using Support Vector Machines. We see a very high potential of such machine learning techniques for a variety of computational chemistry problems that occur in the drug discovery and drug design process
The transmembrane tight junction protein occludin is sensitive to oxidative stress. Occludin oligomerizes; however, its function in the tight junction is unknown. The cytosolic C-terminal tail contains a coiled coil-domain and forms dimers contributing to the oligomerization. The regulation of the oligomerization remains unclear. As the domain area contains sulfhydryl residues, we tested the hypothesis that the dimerization of the coiled coil-domain depends on these residues. We showed that the dimerization is modulated by the thiol concentration in the low-millimolar range, which is relevant both for physiological and pathophysiological conditions. Masking the sulfhydryl residues in the fragment by covalent binding of 4-vinyl pyridine prevented the dimerization but did not affect its helical structure and cylindric shape. The data demonstrate, for the first time, that disulfide bridge formation of murine cystein 408 is involved in the dimerization. This process is redox-sensitive but the secondary structure of the domain is not. It is concluded that the dimerization of occludin may play a regulatory role in the tight junction assembly under physiological and pathological conditions.
Information integration across company borders becomes increasingly important for the success of product lifecycle management in industry and complex supply chains. Semantic technologies are about to play a crucial role in this integrative process. However, cross-company data exchange requires mechanisms to enable fine-grained access control definition and enforcement, preventing unauthorized leakage of confidential data across company borders. Currently available semantic repositories are not sufficiently equipped to satisfy this important requirement. This paper presents an infrastructure for controlled sharing of semantic data between cooperating business partners. First, we motivate the need for access control in semantic data federations by a case study in the industrial service sector. Furthermore, we present an architecture for controlling access to semantic repositories that is based on our newly developed SemForce security service. Finally, we show the practical feasibility of this architecture by an implementation and several performance experiments.
Robust ensemble learning
(2000)
fundamental challenge for product-lifecycle management in collaborative value networks is to utilize the vast amount of product information available from heterogeneous sources in order to improve business analytics, decision support, and processes. This becomes even more challenging if those sources are distributed across multiple organizations. Federations of semantic information services, combining service-orientation and semantic technologies, provide a promising solution for this problem. However, without proper measures to establish information security, companies will be reluctant to join an information federation, which could lead to serious adoption barriers.
Following the design science paradigm, this paper presents general objectives and a process for designing a secure federation of semantic information services. Furthermore, new as well as established security measures are discussed. Here, our contributions include an access-control enforcement system for semantic information services and a process for modeling access-control policies across organizations. In addition, a comprehensive security architecture is presented. An implementation of the architecture in the context of an application scenario and several performance experiments demonstrate the practical viability of our approach.