Elke Eisenschmidt

SynProt: a database for proteins of detergent-resistant synaptic protein preparations

Chemical synapses are highly specialized cell–cell contacts for communication between neurons in the CNS characterized by complex and dynamic protein networks at both synaptic membranes. The cytomatrix at the active zone (CAZ) organizes the apparatus for the regulated release of transmitters from the presynapse. At the postsynaptic side, the postsynaptic density constitutes the machinery for detection, integration, and transduction of the transmitter signal. Both pre- and postsynaptic protein networks represent the molecular substrates for synaptic plasticity. Their function can be altered both by regulating their composition and by post-translational modification of their components. For a comprehensive understanding of synaptic networks the entire ensemble of synaptic proteins has to be considered. To support this, we established a comprehensive database for synaptic junction proteins (SynProt database) primarily based on proteomics data obtained from biochemical preparations of detergent-resistant synaptic junctions. The database currently contains 2,788 non-redundant entries of rat, mouse, and some human proteins, which mainly have been manually extracted from 12 proteomic studies and annotated for synaptic subcellular localization. Each dataset is completed with manually added information including protein classifiers as well as automatically retrieved and updated information from public databases (UniProt and PubMed). We intend that the database will be used to support modeling of synaptic protein networks and rational experimental design.

Synaptic proteome changes in mouse brain regions upon auditory discrimination learning

Changes in synaptic efficacy underlying learning and memory processes are assumed to be associated with alterations of the protein composition of synapses. Here, we performed a quantitative proteomic screen to monitor changes in the synaptic proteome of four brain areas (auditory cortex, frontal cortex, hippocampus striatum) during auditory learning. Mice were trained in a shuttle box GO/NO‐GO paradigm to discriminate between rising and falling frequency modulated tones to avoid mild electric foot shock. Control‐treated mice received corresponding numbers of either the tones or the foot shocks. Six hours and 24 h later, the composition of a fraction enriched in synaptic cytomatrix‐associated proteins was compared to that obtained from naïve mice by quantitative mass spectrometry. In the synaptic protein fraction obtained from trained mice, the average percentage (±SEM) of downregulated proteins (59.9 ± 0.5%) exceeded that of upregulated proteins (23.5 ± 0.8%) in the brain regions studied. This effect was significantly smaller in foot shock (42.7 ± 0.6% down, 40.7 ± 1.0% up) and tone controls (43.9 ± 1.0% down, 39.7 ± 0.9% up). These data suggest that learning processes initially induce removal and/or degradation of proteins from presynaptic and postsynaptic cytoskeletal matrices before these structures can acquire a new, postlearning organisation. In silico analysis points to a general role of insulin‐like signalling in this process.

Integrally indecomposable polytopes and the survivable network design problem

We consider the survivable network design problem for fractional flows and integral capacities and demands. While this problem was modelled using so-called metric inequalities in the past, we will present an integer program which is based on the automatic linearization of non-linear constraints. It turns out that the linear relaxation of the latter formulation is actually stronger than the linear relaxations of the previously known models. Our model making use of integrally indecomposable polytopes, we introduce a new way of computing these polytopes via the chamber decomposition of the parameter space.

A polynomial time approximation algorithm for the two-commodity splittable flow problem

We consider a generalization of the unsplittable maximum two-commodity flow problem on undirected graphs where each commodity

Integer Minkowski Programs and the Design of Survivable Networks

{i \in \{1, 2\}}

Discovering all associations in discrete data using frequent minimally infrequent attribute sets

can be split into a bounded number ki of equally-sized chunks that can be routed on different paths. We show that in contrast to the single-commodity case this problem is NP-hard, and hard to approximate to within a factor of α > 1/2. We present a polynomial time 1/2-approximation algorithm for the case of uniform chunk size over both commodities and show that for even ki and a mild cut condition it can be modified to yield an exact method. The uniform case can be used to derive a 1/4-approximation for the maximum concurrent (k1, k2)-splittable flow without chunk size restrictions for fixed demand ratios.