Assessing the gain of biological data integration in gene networks inference

Background: A current challenge in gene annotation is to define the gene function in the context of the network of relationships instead of using single genes. The inference of gene networks (GNs) has emerged as an approach to better understand the biology of the system and to study how several components of this network interact with each other and keep their functions stable. However, in general there is no sufficient data to accurately recover the GNs from their expression levels leading to the curse of dimensionality, in which the number of variables is higher than samples. One way to mitigate this problem is to integrate biological data instead of using only the expression profiles in the inference process. Nowadays, the use of several biological information in inference methods had a significant increase in order to better recover the connections between genes and reduce the false positives. What makes this strategy so interesting is the possibility of confirming the known connections through the included biological data, and the possibility of discovering new relationships between genes when observed the expression data. Although several works in data integration have increased the performance of the network inference methods, the real contribution of adding each type of biological information in the obtained improvement is not clear. Methods: We propose a methodology to include biological information into an inference algorithm in order to assess its prediction gain by using biological information and expression profile together. We also evaluated and compared the gain of adding four types of biological information: (a) protein-protein interaction, (b) Rosetta stone fusion proteins, (c) KEGG and (d) KEGG+GO. Results and conclusions: This work presents a first comparison of the gain in the use of prior biological information in the inference of GNs by considering the eukaryote (P. falciparum) organism. Our results indicates that information based on direct interaction can produce a higher improvement in the gain than data about a less specific relationship as GO or KEGG. Also, as expected, the results show that the use of biological information is a very important approach for the improvement of the inference. We also compared the gain in the inference of the global network and only the hubs. The results indicates that the use of biological information can improve the identification of the most connected proteins.

Class-specific metrics for multidimensional data projection applied to CBIR

Content-based image retrieval is still a challenging issue due to the inherent complexity of images and choice of the most discriminant descriptors. Recent developments in the field have introduced multidimensional projections to burst accuracy in the retrieval process, but many issues such as introduction of pattern recognition tasks and deeper user intervention to assist the process of choosing the most discriminant features still remain unaddressed. In this paper, we present a novel framework to CBIR that combines pattern recognition tasks, class-specific metrics, and multidimensional projection to devise an effective and interactive image retrieval system. User interaction plays an essential role in the computation of the final multidimensional projection from which image retrieval will be attained. Results have shown that the proposed approach outperforms existing methods, turning out to be a very attractive alternative for managing image data sets.

Interaction among nitric oxide (NO)-related genes in migraine susceptibility

The pathogenic mechanisms involved in migraine are complex and not completely clarified. Because there is evidence for the involvement of nitric oxide (NO) in migraine pathophysiology, candidate gene approaches focusing on genes affecting the endothelial function have been studied including the genes encoding endothelial NO synthase (eNOS), inducible NO synthase (iNOS), and vascular endothelial growth factor (VEGF). However, investigations on gene-gene interactions are warranted to better elucidate the genetic basis of migraine. This study aimed at characterizing interactions among nine clinically relevant polymorphisms in eNOS (T-786C/rs2070744, the 27 bp VNTR in intron 4, the Glu298Asp/rs1799983, and two additional tagSNPs rs3918226 and rs743506), iNOS (C(-1026)A/rs2779249 and G2087A/rs2297518), and VEGF (C(-2578)A/rs699947 and G(-634)C/rs2010963) in migraine patients and control group. Genotypes were determined by real-time polymerase chain reaction using the Taqman(A (R)) allele discrimination assays or PCR and fragment separation by electrophoresis in 99 healthy women without migraine (control group) and in 150 women with migraine divided into two groups: 107 with migraine without aura and 43 with aura. The multifactor dimensionality reduction method was used to detect and characterize gene-gene interactions. We found a significant interaction between eNOS rs743506 and iNOS 2087G/A polymorphisms in migraine patients compared to control group (P < 0.05), suggesting that this combination affect the susceptibility to migraine. Further studies are needed to determine the molecular mechanisms explaining this interaction.

Epistasis among eNOS, MMP-9 and VEGF maternal genotypes in hypertensive disorders of pregnancy

Polymorphisms of the endothelial nitric oxide synthase (eNOS), matrix metalloproteinase-9 (MMP-9) and vascular endothelial growth factor (VEGF) genes were shown to be associated with hypertensive disorders of pregnancy. However, epistasis is suggested to be an important component of the genetic susceptibility to preeclampsia (PE). The aim of this study was to characterize the interactions among these genes in PE and gestational hypertension (GH). Seven clinically relevant polymorphisms of eNOS (T-786C, rs2070744, a variable number of tandem repeats in intron 4 and Glu298Asp, rs1799983), MMP-9 (C-1562T, rs3918242 and -90(CA)(13-25), rs2234681) and VEGF (C-2578A, rs699947 and G-634C, rs2010963) were genotyped by TaqMan allelic discrimination assays or PCR and fragment separation by electrophoresis in 122 patients with PE, 107 patients with GH and a control group of 102 normotensive pregnant (NP) women. A robust multifactor dimensionality reduction analysis was used to characterize gene-gene interactions. Although no significant genotype combinations were observed for the comparison between the GH and NP groups (P>0.05), the combination of MMP-9-1562CC with VEGF-634GG was more frequent in NP women than in women with PE (P<0.05). Moreover, the combination of MMP-9-1562CC with VEGF-634CC or MMP-9-1562CT with VEGF-634CC or-634GG was more frequent in women with PE than in NP women (P<0.05). These results are obscured when single polymorphisms in these genes are considered and suggest that specific genotype combinations of MMP-9 and VEGF contribute to PE susceptibility. Hypertension Research (2012) 35, 917-921; doi:10.1038/hr.2012.60; published online 10 May 2012

Effect of external fields in Axelrod's model of social dynamics

The study of the effects of spatially uniform fields on the steady-state properties of Axelrod's model has yielded plenty of counterintuitive results. Here, we reexamine the impact of this type of field for a selection of parameters such that the field-free steady state of the model is heterogeneous or multicultural. Analyses of both one- and two-dimensional versions of Axelrod's model indicate that the steady state remains heterogeneous regardless of the value of the field strength. Turning on the field leads to a discontinuous decrease on the number of cultural domains, which we argue is due to the instability of zero-field heterogeneous absorbing configurations. We find, however, that spatially nonuniform fields that implement a consensus rule among the neighborhood of the agents enforce homogenization. Although the overall effects of the fields are essentially the same irrespective of the dimensionality of the model, we argue that the dimensionality has a significant impact on the stability of the field-free homogeneous steady state.

Cross-cultural adaptation and psychometric properties of the Brazilian-Portuguese version of the Duke Anticoagulation Satisfaction Scale

Aim. The aim of this study was to evaluate the internal reliability and validity of the BrazilianPortuguese version of Duke Anticoagulation Satisfaction Scale (DASS) among cardiovascular patients. Background. Oral anticoagulation is widely used to prevent and treat thromboembolic events in several conditions, especially in cardiovascular diseases; however, this therapy can induce dissatisfaction and reduce the quality of life. Design. Methodological and cross-sectional research design. Methods. The cultural adaptation of the DASS included the translation and back-translation, discussions with healthcare professionals and patients to ensure conceptual equivalence, semantic evaluation and instrument pretest. The BrazilianPortuguese version of the DASS was tested among subjects followed in a university hospital anticoagulation outpatient clinic. The psychometric properties were assessed by construct validity (convergent, known groups and dimensionality) and internal consistency/reliability (Cronbachs alpha). Results. A total of 180 subjects under oral anticoagulation formed the baseline validation population. DASS total score and SF-36 domain correlations were moderate for General health (r = -0.47, p < 0.01), Vitality (r = -0.44, p < 0.01) and Mental health (r = -0.42, p < 0.01) (convergent). Age and length on oral anticoagulation therapy (in years) were weakly correlated with total DASS score and most of the subscales, except Limitation (r = -0.375, p < 0.01) (Known groups). The Cronbachs alpha coefficient was 0.79 for the total scale, and it ranged from 0.76 (hassles and burdens)0.46 (psychological impact) among the domains, confirming the internal consistency reliability. Conclusions. The BrazilianPortuguese version of the DASS has shown levels of reliability and validity comparable with the original English version. Relevance to clinical practice. Healthcare practitioners and researchers need internationally validated measurement tools to compare outcomes of interventions in clinical management and research tools in oral anticoagulation therapy.

A comparative study on multiscale fractal dimension descriptors

Fractal theory presents a large number of applications to image and signal analysis. Although the fractal dimension can be used as an image object descriptor, a multiscale approach, such as multiscale fractal dimension (MFD), increases the amount of information extracted from an object. MFD provides a curve which describes object complexity along the scale. However, this curve presents much redundant information, which could be discarded without loss in performance. Thus, it is necessary the use of a descriptor technique to analyze this curve and also to reduce the dimensionality of these data by selecting its meaningful descriptors. This paper shows a comparative study among different techniques for MFD descriptors generation. It compares the use of well-known and state-of-the-art descriptors, such as Fourier, Wavelet, Polynomial Approximation (PA), Functional Data Analysis (FDA), Principal Component Analysis (PCA), Symbolic Aggregate Approximation (SAX), kernel PCA, Independent Component Analysis (ICA), geometrical and statistical features. The descriptors are evaluated in a classification experiment using Linear Discriminant Analysis over the descriptors computed from MFD curves from two data sets: generic shapes and rotated fish contours. Results indicate that PCA, FDA, PA and Wavelet Approximation provide the best MFD descriptors for recognition and classification tasks. (C) 2012 Elsevier B.V. All rights reserved.

Psychometric properties of the Spanish version of the mindful attention awareness scale (MAAS) in patients with fibromyalgia

Background Mindful-based interventions improve functioning and quality of life in fibromyalgia (FM) patients. The aim of the study is to perform a psychometric analysis of the Spanish version of the Mindful Attention Awareness Scale (MAAS) in a sample of patients diagnosed with FM. Methods The following measures were administered to 251 Spanish patients with FM: the Spanish version of MAAS, the Chronic Pain Acceptance Questionnaire, the Pain Catastrophising Scale, the Injustice Experience Questionnaire, the Psychological Inflexibility in Pain Scale, the Fibromyalgia Impact Questionnaire and the Euroqol. Factorial structure was analysed using Confirmatory Factor Analyses (CFA). Cronbach's α coefficient was calculated to examine internal consistency, and the intraclass correlation coefficient (ICC) was calculated to assess the test-retest reliability of the measures. Pearson’s correlation tests were run to evaluate univariate relationships between scores on the MAAS and criterion variables. Results The MAAS scores in our sample were low (M = 56.7; SD = 17.5). CFA confirmed a two-factor structure, with the following fit indices [sbX2 = 172.34 (p < 0.001), CFI = 0.95, GFI = 0.90, SRMR = 0.05, RMSEA = 0.06. MAAS was found to have high internal consistency (Cronbach’s α = 0.90) and adequate test-retest reliability at a 1–2 week interval (ICC = 0.90). It showed significant and expected correlations with the criterion measures with the exception of the Euroqol (Pearson = 0.15). Conclusion Psychometric properties of the Spanish version of the MAAS in patients with FM are adequate. The dimensionality of the MAAS found in this sample and directions for future research are discussed.

TCAD approaches to multidimensional simulation of advanced semiconductor devices

Technology scaling increasingly emphasizes complexity and non-ideality of the electrical behavior of semiconductor devices and boosts interest on alternatives to the conventional planar MOSFET architecture. TCAD simulation tools are fundamental to the analysis and development of new technology generations. However, the increasing device complexity is reflected in an augmented dimensionality of the problems to be solved. The trade-off between accuracy and computational cost of the simulation is especially influenced by domain discretization: mesh generation is therefore one of the most critical steps and automatic approaches are sought. Moreover, the problem size is further increased by process variations, calling for a statistical representation of the single device through an ensemble of microscopically different instances. The aim of this thesis is to present multi-disciplinary approaches to handle this increasing problem dimensionality in a numerical simulation perspective. The topic of mesh generation is tackled by presenting a new Wavelet-based Adaptive Method (WAM) for the automatic refinement of 2D and 3D domain discretizations. Multiresolution techniques and efficient signal processing algorithms are exploited to increase grid resolution in the domain regions where relevant physical phenomena take place. Moreover, the grid is dynamically adapted to follow solution changes produced by bias variations and quality criteria are imposed on the produced meshes. The further dimensionality increase due to variability in extremely scaled devices is considered with reference to two increasingly critical phenomena, namely line-edge roughness (LER) and random dopant fluctuations (RD). The impact of such phenomena on FinFET devices, which represent a promising alternative to planar CMOS technology, is estimated through 2D and 3D TCAD simulations and statistical tools, taking into account matching performance of single devices as well as basic circuit blocks such as SRAMs. Several process options are compared, including resist- and spacer-defined fin patterning as well as different doping profile definitions. Combining statistical simulations with experimental data, potentialities and shortcomings of the FinFET architecture are analyzed and useful design guidelines are provided, which boost feasibility of this technology for mainstream applications in sub-45 nm generation integrated circuits.

Vortex dynamics in the high-temperature superconductors YBa 2 Cu 3 O 7 and Bi 2 Sr 2 CaCu 2 O 8 + delta in low- and high-dissipative transport measurements

This thesis investigates phenomena of vortex dynamics in type II superconductors depending on the dimensionality of the flux-line system and the strength of the driving force. In the low dissipative regime of Bi_2Sr_2CaCu_2O_{8+delta} (BSCCO) the influence of oxygen stoichiometry on flux-line tension was examined. An entanglement crossover of the vortex system at low magnetic fields was identified and a comprehensive B-T phase diagram of solid and fluid phases derived.In YBa_2Cu_3O_7 (YBCO) extremely long (>100 mm) high-quality measurement bridges allowed to extend the electric-field window in transport measurements by up to three orders of magnitude. Complementing analyses of the data conclusively produced dynamic exponents of the glass transition z~9 considerably higher than theoretically predicted and previously reported. In high-dissipative measurements a voltage instability appearing in the current-voltage characteristics of type II superconductors was observed for the first time in BSCCO and shown to result from a Larkin-Ovchinnikov flux-flow vortex instability under the influence of quasi-particle heating. However, in an analogous investigation of YBCO the instability was found to appear only in the temperature and magnetic-field regime of the vortex-glass state. Rapid-pulse measurements fully confirmed this correlation of vortex glass and instability in YBCO and revealed a constant rise time (~µs).

An evolutionary approach to the design of experiments for combinatorial optimization with an application to enzyme engineering

In a large number of problems the high dimensionality of the search space, the vast number of variables and the economical constrains limit the ability of classical techniques to reach the optimum of a function, known or unknown. In this thesis we investigate the possibility to combine approaches from advanced statistics and optimization algorithms in such a way to better explore the combinatorial search space and to increase the performance of the approaches. To this purpose we propose two methods: (i) Model Based Ant Colony Design and (ii) Naïve Bayes Ant Colony Optimization. We test the performance of the two proposed solutions on a simulation study and we apply the novel techniques on an appplication in the field of Enzyme Engineering and Design.

Advanced Numerical Simulation of Silicon-Based Solar Cells

Photovoltaic (PV) conversion is the direct production of electrical energy from sun without involving the emission of polluting substances. In order to be competitive with other energy sources, cost of the PV technology must be reduced ensuring adequate conversion efficiencies. These goals have motivated the interest of researchers in investigating advanced designs of crystalline silicon solar (c-Si) cells. Since lowering the cost of PV devices involves the reduction of the volume of semiconductor, an effective light trapping strategy aimed at increasing the photon absorption is required. Modeling of solar cells by electro-optical numerical simulation is helpful to predict the performance of future generations devices exhibiting advanced light-trapping schemes and to provide new and more specific guidelines to industry. The approaches to optical simulation commonly adopted for c-Si solar cells may lead to inaccurate results in case of thin film and nano-stuctured solar cells. On the other hand, rigorous solvers of Maxwell equations are really cpu- and memory-intensive. Recently, in optical simulation of solar cells, the RCWA method has gained relevance, providing a good trade-off between accuracy and computational resources requirement. This thesis is a contribution to the numerical simulation of advanced silicon solar cells by means of a state-of-the-art numerical 2-D/3-D device simulator, that has been successfully applied to the simulation of selective emitter and the rear point contact solar cells, for which the multi-dimensionality of the transport model is required in order to properly account for all physical competing mechanisms. In the second part of the thesis, the optical problems is discussed. Two novel and computationally efficient RCWA implementations for 2-D simulation domains as well as a third RCWA for 3-D structures based on an eigenvalues calculation approach have been presented. The proposed simulators have been validated in terms of accuracy, numerical convergence, computation time and correctness of results.

Reinforcement Lernen mit Regularisierungsnetzwerken

Die Arbeit behandelt das Problem der Skalierbarkeit von Reinforcement Lernen auf hochdimensionale und komplexe Aufgabenstellungen. Unter Reinforcement Lernen versteht man dabei eine auf approximativem Dynamischen Programmieren basierende Klasse von Lernverfahren, die speziell Anwendung in der Künstlichen Intelligenz findet und zur autonomen Steuerung simulierter Agenten oder realer Hardwareroboter in dynamischen und unwägbaren Umwelten genutzt werden kann. Dazu wird mittels Regression aus Stichproben eine Funktion bestimmt, die die Lösung einer "Optimalitätsgleichung" (Bellman) ist und aus der sich näherungsweise optimale Entscheidungen ableiten lassen. Eine große Hürde stellt dabei die Dimensionalität des Zustandsraums dar, die häufig hoch und daher traditionellen gitterbasierten Approximationsverfahren wenig zugänglich ist. Das Ziel dieser Arbeit ist es, Reinforcement Lernen durch nichtparametrisierte Funktionsapproximation (genauer, Regularisierungsnetze) auf -- im Prinzip beliebig -- hochdimensionale Probleme anwendbar zu machen. Regularisierungsnetze sind eine Verallgemeinerung von gewöhnlichen Basisfunktionsnetzen, die die gesuchte Lösung durch die Daten parametrisieren, wodurch die explizite Wahl von Knoten/Basisfunktionen entfällt und so bei hochdimensionalen Eingaben der "Fluch der Dimension" umgangen werden kann. Gleichzeitig sind Regularisierungsnetze aber auch lineare Approximatoren, die technisch einfach handhabbar sind und für die die bestehenden Konvergenzaussagen von Reinforcement Lernen Gültigkeit behalten (anders als etwa bei Feed-Forward Neuronalen Netzen). Allen diesen theoretischen Vorteilen gegenüber steht allerdings ein sehr praktisches Problem: der Rechenaufwand bei der Verwendung von Regularisierungsnetzen skaliert von Natur aus wie O(n**3), wobei n die Anzahl der Daten ist. Das ist besonders deswegen problematisch, weil bei Reinforcement Lernen der Lernprozeß online erfolgt -- die Stichproben werden von einem Agenten/Roboter erzeugt, während er mit der Umwelt interagiert. Anpassungen an der Lösung müssen daher sofort und mit wenig Rechenaufwand vorgenommen werden. Der Beitrag dieser Arbeit gliedert sich daher in zwei Teile: Im ersten Teil der Arbeit formulieren wir für Regularisierungsnetze einen effizienten Lernalgorithmus zum Lösen allgemeiner Regressionsaufgaben, der speziell auf die Anforderungen von Online-Lernen zugeschnitten ist. Unser Ansatz basiert auf der Vorgehensweise von Recursive Least-Squares, kann aber mit konstantem Zeitaufwand nicht nur neue Daten sondern auch neue Basisfunktionen in das bestehende Modell einfügen. Ermöglicht wird das durch die "Subset of Regressors" Approximation, wodurch der Kern durch eine stark reduzierte Auswahl von Trainingsdaten approximiert wird, und einer gierigen Auswahlwahlprozedur, die diese Basiselemente direkt aus dem Datenstrom zur Laufzeit selektiert. Im zweiten Teil übertragen wir diesen Algorithmus auf approximative Politik-Evaluation mittels Least-Squares basiertem Temporal-Difference Lernen, und integrieren diesen Baustein in ein Gesamtsystem zum autonomen Lernen von optimalem Verhalten. Insgesamt entwickeln wir ein in hohem Maße dateneffizientes Verfahren, das insbesondere für Lernprobleme aus der Robotik mit kontinuierlichen und hochdimensionalen Zustandsräumen sowie stochastischen Zustandsübergängen geeignet ist. Dabei sind wir nicht auf ein Modell der Umwelt angewiesen, arbeiten weitestgehend unabhängig von der Dimension des Zustandsraums, erzielen Konvergenz bereits mit relativ wenigen Agent-Umwelt Interaktionen, und können dank des effizienten Online-Algorithmus auch im Kontext zeitkritischer Echtzeitanwendungen operieren. Wir demonstrieren die Leistungsfähigkeit unseres Ansatzes anhand von zwei realistischen und komplexen Anwendungsbeispielen: dem Problem RoboCup-Keepaway, sowie der Steuerung eines (simulierten) Oktopus-Tentakels.

Simulations of charge transport in organic compounds

To aid the design of organic semiconductors, we study the charge transport properties of organic liquid crystals, i.e. hexabenzocoronene and carbazole macrocycle, and single crystals, i.e. rubrene, indolocarbazole and benzothiophene derivatives (BTBT, BBBT). The aim is to find structure-property relationships linking the chemical structure as well as the morphology with the bulk charge carrier mobility of the compounds. To this end, molecular dynamics (MD) simulations are performed yielding realistic equilibrated morphologies. Partial charges and molecular orbitals are calculated based on single molecules in vacuum using quantum chemical methods. The molecular orbitals are then mapped onto the molecular positions and orientations, which allows calculation of the transfer integrals between nearest neighbors using the molecular orbital overlap method. Thus we obtain realistic transfer integral distributions and their autocorrelations. In case of organic crystals the differences between two descriptions of charge transport, namely semi-classical dynamics (SCD) in the small polaron limit and kinetic Monte Carlo (KMC) based on Marcus rates, are studied. The liquid crystals are investigated solely in the hopping limit. To simulate the charge dynamics using KMC, the centers of mass of the molecules are mapped onto lattice sites and the transfer integrals are used to compute the hopping rates. In the small polaron limit, where the electronic wave function is spread over a limited number of neighboring molecules, the Schroedinger equation is solved numerically using a semi-classical approach. The results are compared for the different compounds and methods and, where available, with experimental data. The carbazole macrocycles form columnar structures arranged on a hexagonal lattice with side chains facing inwards, so columns can closely approach each other allowing inter-columnar and thus three-dimensional transport. When taking only intra-columnar transport into account, the mobility is orders of magnitude lower than in the three-dimensional case. BTBT is a promising material for solution-processed organic field-effect transistors. We are able to show that, on the time-scales of charge transport, static disorder due to slow side chain motions is the main factor determining the mobility. The resulting broad transfer integral distributions modify the connectivity of the system but sufficiently many fast percolation paths remain for the charges. Rubrene, indolocarbazole and BBBT are examples of crystals without significant static disorder. The high mobility of rubrene is explained by two main features: first, the shifted cofacial alignment of its molecules, and second, the high center of mass vibrational frequency. In comparsion to SCD, only KMC based on Marcus rates is capable of describing neighbors with low coupling and of taking static disorder into account three-dimensionally. Thus it is the method of choice for crystalline systems dominated by static disorder. However, it is inappropriate for the case of strong coupling and underestimates the mobility of well-ordered crystals. SCD, despite its one-dimensionality, is valuable for crystals with strong coupling and little disorder. It also allows correct treatment of dynamical effects, such as intermolecular vibrations of the molecules. Rate equations are incapable of this, because simulations are performed on static snapshots. We have thus shown strengths and weaknesses of two state of the art models used to study charge transport in organic compounds, partially developed a program to compute and visualize transfer integral distributions and other charge transport properties, and found structure-mobility relations for several promising organic semiconductors.

Thermoelektrische Eigenschaften dünner Schichten und Übergitterstrukturen von IV-VI- und V-VI-Verbundhalbleitern

Als ein vielversprechendes Konzept zur Erhöhung der thermoelektrischen Effizienz wird seit Anfang der 90er Jahre die Nutzung niederdimensionaler Systeme angesehen. Aus theoretischen Arbeiten von Hicks und Dresselhaus folgt, dass in ein- und zweidimensionalen Systemen eine Erhöhung der thermoelektrischen Effizienz möglich ist, die einen Durchbruch für die Anwendung thermoelektrischer Wandler zur Folge haben könnte. Die Realisierung solcher niederdimensionaler Systeme ist in geeigneten Mehrlagenstrukturen und durch Verwendung von Halbleiterverbindungen mit unterschiedlicher Energiebandlücke möglich. Ziel des Verbundprojektes Nitherma war es Mehrfachschichtsysteme mit 2-dimensionalem Transportverhalten aus thermoelektrischen Materialien (Pb1-xSrxTe, Bi2(SexTe1-x)3) herzustellen und auf die erwartete hohe thermoelektrische Effizienz zu untersuchen. Diese wurde messtechnischrndurch die Bestimmung der elektrischen Leitfähigkeit, des Seebeck-Koeffizienten und der Wärmeleitfähigkeit parallel zu den Schichtebenen (in-plane-Transporteigenschaft) ermittelt. Ziel dieser Arbeit war einerseits die Verbesserung der Präparations- und Messtechnik bei der Untersuchung der Wärmeleitfähigkeit von Schichten und Schichtsystemen sowie die Demonstration der Reproduzierbarkeit, andererseits die Interpretation der an niederdimensionalen Strukturen ermittelten Transportmessungen. Um den Einfluß der Niederdimensionalität auf die Wärmeleitfähigkeit zu ermitteln, wurden umfangreiche Messungen an unterschiedlich dimensionierten Übergitter- und Multi-Quantum-Well-Strukturen (MQW-Strukturen) durchgeführt. Die Verifizierung der von den Projektpartnern durchgeführten Transportmessungen wurde durch die Messung des Seebeck-Koeffizienten unterstützt.Neben der Charakterisierung durch Transportmessungen erfolgte die Bestimmung der thermoelektrischen Effizienz.