Qucosa - Publikationsserver der Universität Leipzig
Not a member yet
16020 research outputs found
Sort by
Molecular Characterization of Serpin-Recepter Interactions
No full textThe serine protease inhibitor vaspin (SERPINA12) plays a key role in protecting adipose tissue from inflammation and preserving insulin sensitivity in obesity. Unlike many serpins, vaspin interacts with several ligands in addition to its target proteases. The aim of this thesis was to identify new binding molecules and interaction partners of vaspin, focusing on cell surface receptors, and including studies on receptor binding of antitrypsin (AAT, SERPINA1) and neuroserpin (SERPINI1). First, we found that endocytosed vaspin partially localizes to the nucleus and binds strongly to DNA, which also accelerates the inhibition of KLK7. This identifies vaspin as only the second human DNA-binding serpin. We also crystallized polyphosphate-bound vaspin and confirmed a shared polyphosphate- and heparin-binding site atop of central β-sheet A. Secondly, by investigating internalization of vaspin in detail, we identified the LRP1 as the major endocytosis receptor. LRP1 binding depends on prior contact with extracellular matrix heparan sulfates and is further facilitated by insulin-stimulated LRP1 translocation in mature adipocytes. We found that native and polymerized, but not RCL-cleaved vaspin, are endocytosed and partially targeted for lysosomal degradation. Thirdly, we investigated AAT binding to the glucocorticoid receptor (GR) and other reported receptors, comparing native and oxidized, non-inhibitory AAT variants. We found that both native and oxidized AAT bound the GR, suggesting that oxidized AAT may modulate GR-dependent inflammatory responses. Extending these receptor studies to other AAT variants, neuroserpin, and vaspin, we examined serpin binding to various receptors, focusing on the LDL receptor family and other reported receptors. AAT and oxidized AAT bound similar receptors, confirming our previous results, whereas ZZ-AAT, cleaved AAT and neuroserpin did not bind any of the receptors tested. Interestingly, vaspin bound several LDLR family receptors with high affinity. Together, this thesis provides new insights by expanding vaspin’s molecular repertoire, identifying novel ligands and receptors, highlighted by the detailed investigation of LRP1-mediated endocytosis. It also established that oxidized AAT likely exhibits similar signaling capabilities via GR binding
The Double Translation Surfaces of Sophus Lie
No full textDouble translation surfaces were established by Sophus Lie in 1882. These surfaces
in three-dimensional space possess the characteristic property of being
associated with a quartic curve Q in the projective plane. This association allows
for the classification of the surfaces based on the type of their associated
curve. Furthermore, by determining Abelian integrals over Q one obtains parameterizations
of the corresponding surface.
Focusing on developments up to the early twentieth century, the genesis of the
theory of such surfaces is discussed from both mathematical and historical perspectives.
To this end, we follow the outline of two historical dissertations written
under the supervision of Sophus Lie at Leipzig University and present the most
relevant theorems, the classification of the surfaces and typical examples. Since
some of the surfaces have been realized as plaster models also the history of the
collection of mathematical models at Leipzig University is addressed. Finally,
the thesis provides a chapter on computational mathematics for the cases where
Q decomposes into four straight lines
Early Precursors of Theory of Mind in the Developing Brain
No full textAs social beings, humans rely heavily on the ability to reason about others’ thoughts, known as
Theory of Mind (ToM). Traditionally, ToM was thought to emerge around age four, but recent
research using non-verbal tasks suggests that even infants may predict others’ actions based on their
mental states. A recent theory proposes that infants pass non-verbal tasks by being socially biased
to attend to what others deem relevant, leading to better encoding and memory of shared
experiences. This thesis explores how infants pass non-verbal ToM tasks, focusing on social biases,
precursors, and supporting neural networks. The research investigates two main questions: (A)
What social cognitive processes and neural developments drive non-verbal ToM success? (Studies
1 and 3) (B) What precursors might help infants pass non-verbal tasks through mechanisms other
than mature ToM reasoning? (Studies 2 and 3)
The first study examined preschoolers' structural brain networks involved in non-verbal and
verbal ToM tasks. Results revealed two distinct networks: a ventral network for non-verbal ToM
tasks and a dorsal network for verbal ToM reasoning, overlapping with the Default network. The
non-verbal tasks were linked to connections within the Salience network, which filters salient
environmental stimuli, among which social cues are particularly salient. This suggests that young
children use these cues, such as others’ attention, to predict other’s actions and succeed in non-
verbal ToM tasks.
The second study focused on joint attention, a behavioral precursor to ToM. Joint attention
involves coordinating one’s attention on objects with others and thus may bias infants toward social
cues. Resting-state fMRI data of infants showed that connectivity patterns predictive of joint
attention also predicted verbal ToM abilities in preschoolers. These findings emphasized the role
of the Salience and Default networks in helping infants transition from perceptual processing in the
Salience network to more abstract social reasoning in the Default network.
The third study tested whether joint attention enhances infants' memory of object locations,
a key element in non-verbal ToM tasks. Using an adapted A-not-B task, where infants repeatedly
search for an object in one location before it is moved to another, we found that joint attention did
not improve infants' memory of the object's new location. This suggests that social cues' influence
on memory may be fragile, and infants may use other strategies in non-verbal ToM tasks.
Overall, the findings suggest that young children use strategies other than mature ToM
reasoning to understand others, questioning theories of mature ToM before 4 years. Non-verbal
ToM tasks in infancy likely rely on simpler mechanisms, such as socially influenced attention
The role of L1 translation form in L2 compound processing: the case of native Czech speakers processing German noun-noun compounds
No full textThis study explores how Czech-German late bilinguals process German (L2) noun-noun compounds. Using a lexical decision task combined with translation constituent priming, we investigated two factors potentially influencing the L2 compound processing: (a) the compound translation corresponds to one derived noun (e.g., Abendstern—večernice, ‘evening star') or to an adjective + noun phrase (e.g., Weizenmehl—pšeničná mouka, ‘wheat flour'); and (b) the compound translation entails translations of compound constituents (L1 translation of Abendstern, večernice, includes only first constituent, i.e. modifier, Abend = večer, ‘evening'; L1 translation of Weizenmehl, pšeničná mouka, includes both constituents, Weizen = pšenice, ‘wheat', Mehl = mouka, ‘flour'). Two experiments were conducted; one focussing on head priming, the other on modifier priming. The results are in line with non-selective bilingual access and decomposition of L2 compounds. They reveal no influence of factor (a), while (b) affects processing
Gut-Brain Interactions after Bariatric Surgery: on the Relationship of Gut Hormones and Working Memory
No full textAdipositas ist eine chronische Erkrankung mit hohem Gesundheitsrisiko und steigender Prävalenz. Die bariatrische Chirurgie gilt als wirksamste Langzeittherapie für anhaltende Gewichtsreduktion, weist jedoch eine hohe interindividuelle Variabilität der Ergebnisse auf. Mechanistisch werden der Gewichtsreduktion u. a. tiefgreifende postoperative Veränderungen von Stoffwechsel, Appetit und Sättigung zugeschrieben – insbesondere eine veränderte Ausschüttung hormoneller Signale der Darm-Hirn-Achse (GLP-1, PYY, Ghrelin). Parallel sind bei Adipositas Beeinträchtigungen exekutiver Funktionen, vor allem des Arbeitsgedächtnisses, beschrieben, für die nach bariatrischer Chirurgie teilweise Verbesserungen berichtet werden. Unklar ist, inwieweit Darm-Hirn-Signale diese kognitiven Veränderungen und einen nachhaltig erfolgreichen Gewichtsverlauf mitbestimmen.
In der vorliegenden Arbeit wurde dieser Zusammenhang konzeptionell untersucht, indem langfristig erfolgreiche und weniger erfolgreiche Operierte gegenübergestellt und die postprandiale Sekretion von Magen-Darm-Hormonen sowie Leistungen des Arbeitsgedächtnisses erfasst wurden. Insgesamt zeigte sich, dass günstigere Langzeitergebnisse mit einer appetitzügelnden hormonellen Antwort und besseren Leistungen in komplexeren Arbeitsgedächtnisaufgaben einhergingen. Explorativ deutete sich an, dass insbesondere die Dynamik des Hormons Ghrelin den Zusammenhang zwischen Gewichtsreduktion und Kognition vermitteln könnte.
Die Ergebnisse ordnen bariatrische Behandlungserfolge in ein neurobiologisches Rahmenmodell ein, in dem dopaminerge Prozesse eine Brückenrolle zwischen Energiehaushalt, Essverhalten und kognitiver Kontrolle spielen könnten. Daraus ergeben sich Perspektiven für personalisierte Nachsorge und Interventionen, die Darm-Hirn-Signale gezielt modulieren. Angesichts des post-hoc-Studiendesigns und der begrenzten Stichprobengröße werden größere und prospektive Studien empfohlen – idealerweise mit zerebraler Bildgebung und unter Einbezug weiterer potenzieller Mediatoren wie Entzündungsstatus, Vagusnerv, Mikrobiom und Gallensäuren.:1 Introduction 3
1.1 Obesity – a Global Challenge 3
1.2 Therapy Success and Variability of Bariatric Surgery 3
1.3 Mechanisms behind Weight Loss after Bariatric Surgery 4
1.4 Gut-Brain Communication after Bariatric Surgery 5
1.5 The Role of Cognitive Functions in Obesity and after Bariatric Surgery 9
1.6 Research Aims 10
1.7 A Central Method for Assessing Working Memory 11
2 Original Publication 13
3 Summary 29
3.1 Introduction and Research Question 29
3.2 Methods 30
3.3 Results and Discussion 31
3.4 Implications 33
4 Bibliography 35
5 Appendix 51
5.1 List of Abbreviations 51
5.2 Academic Contribution 53
5.3 Declaration of Authenticity 54
5.4 Curriculum Vitae 55
5.5 Publications 56
5.6 Acknowledgments 5
Methodological Investigation of the Band Gap Determination of Solid Semiconductors via UV/Vis Spectroscopy
No full textSeveral different evaluation methods (EM) to obtain the band gap energy (Eg) of semiconductors via UV/Vis diffuse reflectance spectroscopy (DRS) are used in literature and no clear consensus which EM to use is established. By using P25 as a model semiconductor, evaluating 32 literature sources and our own experiments, we show that the determined Eg energy is heavily influenced by the EM (differences in Eg of up to 0.87 eV). For P25 loaded with CuO even contradictive effects of metal loading on the Eg are observed using different EM. Different phase compositions of TiO2 are also shown to yield EM-dependent results. This paper thus comprehensively investigates the most common EM on different relevant photocatalyst material classes, illustrates their mathematical and graphical determination, gives recommendations which method to use based on a quantitative indicator and illustrates common pitfalls in determining Eg via UV/Vis DRS
Expression of adhesion GPCRs on innate immune cells after traumatic injury
No full textInjuries claim 4.4 million lives annually worldwide, accounting for nearly 8% of all deaths.
Following trauma, damaged cells and necrotic cells release damage-associated molecular patterns (DAMPs) into circulation, recognized by receptors on innate immune cells, including G-protein coupled receptors (GPCRs).
aGPCRs of the “E-family”, EMR1, EMR2, EMR3 and CD97, are mainly expressed in circulating neutrophils, monocytes/macrophages and/or DCs. It is likely that they are involved in the post-traumatic immune response. The aim of our studies is to quantify the expression of EMR2, EMR3, and CD97 on innate immune cells up to 240 h following trauma, compared with uninjured volunteers, and to investigate their potential value as biomarkers by correlating their expression with clinical data of the patients.
Patients were divided into two groups based on the injury severity score (ISS): ISS <25 (moderately/ severely injured), and ISS ≥25 (very severely injured). Blood samples were collected from patients at 1, 8, 24, 48, 120, and 240 h upon admission to the hospital. Five uninjured volunteers underwent the same time course in blood taking. Leukocyte count was determined in an automatic blood cell analyzer. Leukocytes were prepared using red blood cell lysis, stained with a newly established multi-color fluorophore-labeled antibody panel for circulating myeloid cells in a single sample, and analyzed by flow cytometry. The cells were separated into mature and immature neutrophils, monocyte subsets, dendritic cells (DCs), and natural killer (NK) cells. All subsets were analyzed for the percentage of aGPCR-positive (aGPCR+) cells and the median fluorescence intensity (MFI) of aGPCRs expression.
Neutrophil counts were elevated in very severely injured patients compared to moderately/ severely injured patients at most post-traumatic time points. The percentage of EMR2+ mature neutrophils and their EMR2 expression levels followed a uniform posttraumatic pattern: both remained low from 1 to 8 h, peaked at 48 h, and then declined by 120 to 240 h but stayed above initial levels after trauma. This time course was seen in all patients, regardless of injury severity or type. EMR2+ neutrophils were more activated or primed for activation. At 240 h post-trauma, EMR2 levels on neutrophils remained elevated in many patients and correlated with CRP, IL-6, and PCT levels as well as duration of hospitalization. CD97 was upregulated in mature neutrophils after injury, with levels lowest at 8 h, peaking at 48 h, then normalizing, without significant correlation with clinical data. The percentage of EMR3+ mature neutrophils and their EMR3 levels, which were ISS-dependent, decreased significantly, reaching a minimum at 48 h in very severely injured patients. At 48 h post-trauma, EMR3 levels showed negative correlation with ISS, ICU days, and IL-6/ IL-8 levels.
Monocyte counts were higher in very severely injured patients compared to moderately/severely injured patients at 120 h after trauma; this also tended to be seen at other time points. Very severe trauma induced a phenotypic shift in monocytes: classical monocytes were elevated at 8 and 120 h, with a general upward trend, while non-classical monocytes decreased from 8 to 240 h. Intermediate monocytes, initially low (1-8 h), peaked at 24 h and then decreased until 120h.
In each monocyte subset, EMR2, CD97 and EMR3 showed the lowest levels in classical and the highest in nonclassical monocytes respectively, in both uninjured volunteers and patients.
In traumatized patients, classical and intermediate monocytes showed an early ISS-dependent decrease in the percentage of EMR2+ cells. EMR2 levels were significantly lower immediately and in tendency up to 48 h after trauma in classical and intermediate monocytes. In intermediate monocytes, EMR2+ levels increased to peak at 48 h, slightly decreased afterwards, without obvious difference between the two groups of patients.
The expression of EMR2 and EMR3 in neutrophils and of EMR2 in monocytes is regulated during the innate immune response after trauma, which suggests that these receptors likely play critical roles in (sterile) inflammatory immune response.
EMR2 may serve as a prognostic biomarker, potentially providing insights into patient outcomes in the later stages of recovery following trauma. The loss of EMR3 in neutrophils during the early stages of trauma can serve as an indicator of trauma severity, potentially acting as a prognostic biomarker and a marker of inflammation severity.
In summary, this study systematically investigated the expression of EMR2, EMR3, and CD97 on circulating innate immune cells following trauma, providing novel insights into the immune response. Through a multi-level temporal analysis, the study correlated receptor expression with serum inflammatory markers, cytokines, and clinical parameters, revealing dynamic immune changes over time.:Abbreviation 1
1.Introduction 2
1.1 Polytrauma 2
1.1.1 Background 2
1.1.2 Injury related scales and scores 2
1.1.3 Damage-associated molecular patterns (DAMPs) 2
1.2 Innate immune cells 3
1.2.1 Bone marrow hematopoiesis 3
1.2.2 Neutrophilic granulocytes 3
1.2.3 Monocytes 4
1.2.4 Dendritic cells (DCs) 4
1.2.5 Natural killer cells (NKs) 5
1.3 Flow cytometric analysis of circulating leukocytes 5
1.3.1 Cell surface receptors on circulating leukocytes 5
1.3.2 Isolation and staining of circulating leukocytes 6
1.3.3 Flow cytometric analysis 6
1.4 Adhesion G-protein coupled receptor (aGPCRs) 6
1.4.1 Overview 6
1.4.2 aGPCRs on immune cells 7
1.4.3 EMR2 on neutrophils and monocytes 8
1.4.4 CD97 and EMR3 on neutrophils and monocytes 8
2. Aim of the study 9
3. Materials and methods 9
3.1 Ethics statement and clinical study 9
3.2 Blood sample preparation and determination of parameters 10
3.3 Flow cytometric analysis of patients’ leukocytes 10
3.4 Assess differences in activation between EMR2- and EMR2+ neutrophils 14
3.5 qRT-PCR analysis 14
3.6 Statistics 14
4. Results 15
4.1 Higher blood neutrophil counts in very severely injured patients 15
4.2 Expression of EMR2 on circulating neutrophils 15
4.3 Expression of CD62L and CD11c on EMR2- and EMR2+ neutrophils 17
4.4 Correlation between neutrophilic EMR2 and CRP 18
4.5 Expression of CD97 on circulating neutrophils 19
4.6 Expression of EMR3 on circulating neutrophils 20
4.7 Very severe trauma caused a phenotypic shift among circulating monocytes 22
4.8 Expression of EMR2, CD97 and EMR3 on circulating monocyte subsets 24
4.9 Decreased circ
ulating DCs correlates with adverse clinical parameters late after injury 27
4.10 Expression of EMR2 and CD97 on circulating DCs 28
4.11 Expression of CD97 on NK cells 30
5. Discussion 31
5.1 Absolute number of circulating neutrophils and monocytes after trauma 31
5.2 Expression of EMR2 on circulating neutrophils and monocytes after trauma 32
5.2.1 Upregulation of EMR2 on neutrophils after trauma 32
5.2.2 Expression of EMR2 on monocytes after trauma 33
5.2.3 Expression of EMR2 on DCs after trauma 34
5.3 Expression of CD97 on myeloid and NK cells after trauma 34
5.3.1 Upregulation of CD97 on neutrophils and monocytes after trauma 34
5.3.2 Regulation of CD97 on NKs and DCs after trauma 36
5.4 EMR3 on neutrophils and monocytes after trauma 37
5.5 Study strengths and limitations 38
6. Zusammenfassung der Arbeit 39
7. References 41
Erklärung über die eigenständige Abfassung der Arbeit 47
Curriculum Vitae 48
Publications 50
Acknowledgments 6
Establishing a new method for automatized UPD detection in exome sequencing data applied on a large cohort
No full textUniparental disomy (UPD) describes the inheritance of two homologous of a chromosome from the mother (maternal) or the father (paternal) only. Depending on formation mechanisms, UPDs can either consist of both copies of an identical chromosome (isodisomy) or of two different ones (heterodisomy). UPDs are not mandatorily pathogenic but can lead to disease if chromosomes with associations to imprinting disorder like 6,7,11,14,15, or 20 are affected, if homozygous recessive pathogenic variants appear, or if chromosomal disturbances occur. Currently, there is no special treatment for UPDs but effectively diagnosing them can increase the diagnostic yield and can affect genetic counseling. Especially in cases in which the risk of reoccurrence of a certain disorder in case of a new pregnancy is significantly lower for a homozygous pathogenic variant caused by UPD (<1%) in comparison to a homozygous pathogenic variant caused by an autosomal recessive inheritance mechanism (~25%).
With few exceptions, the majority of UPD cases have gone unnoticed in our institute, with UPD detection not being a part of our routine diagnostics pipeline for exome sequencing data. Therefore, the goal of this project and this dissertation was to close this diagnostic gap and to contribute to establishing a method for UPD detection within our institute using exome sequencing data and to reanalyze available data for before missed UPD cases in our cohort.
As first step and as foundation of my work, a method for UPD detection in our institute based on NGS data had to be established. While performing literature research, Regions of Homozygosity (ROHs) and an Inheritance ratio (IR) could be identified as possible parameters useful for UPD detection. Using UPD positive controls collected from other institutes and in close cooperation with our bioinformatics team, cut-off´s for UPD detection could be set up. Using those parameters a UPD detection pipeline and our diagnostic webapp for manual curation called altAFplotter could be developed and improved. This webapp consists of an interactive user interface illustrating the alternative allele frequency of every chromosome, an overview table and flagging of remarkable variances of the above mentioned cut-off´s for ROH and IR. Since it is not possible to distinguish ROHs caused by UPD or ROHs caused by consanguinity in consanguine families, a consanguinity flagging was included into our tool to avoid false-positive UPD results. Furthermore, all cases with known or suspected consanguinity are excluded from analysis with our UPD detection method. A closer description of the detection tool is included in this dissertation as manuscript accepted for publication.
After the establishment of the UPD detection method UPD, reanalysis was performed with all exome sequencing data available in our institute. Out of 9212 cases in a single, duo or trio constellation 656 samples had to be excluded due to suspected consanguinity. Of the remaining 8556 cases, 14 UPD positive cases could be identified including 13 isodisomy’s and one heterodisomy. This results in an UPD frequency in our cohort of 0,16% which is consistent with the reported incidence ranges between 0,05% and 0,3%, depending on the composition of the investigated cohort. Confirmation of the UPD positive cases took place with multiplex ligation-dependent probe amplification, microsatellite analysis and sanger sequencing. Of the 14 detected UPD positive cases nine are possibly pathogenic. Six cases because of an imprinting disorder including four cases of Angelman syndrome (chr 15), one case of Silver-Russell syndrome (chr 7) and one case of Temple syndrome (chr 14). Three cases were rated as likely pathogenic because of a homozygous recessive pathogenic variant in the patient which was found heterozygous in an unaffected parent (including chr 2 twice and chr 18). Of the remaining five cases, the clinical significance of the found UPD is unremarkable or not yet known.
To gain more information, methylation site analysis of UPD positive cases affecting chromosomes on which no imprinting disorders are known so far was initiated. The results will be reported in context of a new project and are not part of this dissertation. Performing UPD reanalysis of our cohort was the main aim of this project, and the results were published as an article in Human Genetics.
During our cohort analysis, a special case of an individual with a paternal isodisomy affecting the whole chromosome 18 presenting with muscle weakness, scoliosis and global developmental delay was found. The case was rated pathogenic because of a homozygous pathogenic loss of function variant in the PIEZO 2 gene associated with an arthrogryposis syndrome. Performing segregation and microsatellite analysis could prove that the variant is also present in the father but heterozygous. Therefore, the father is not affected. Moreover, by performing methylation site analysis, demethylation of the promotor region of PARD6G-AS1 could be identified. This region was previously described as maternally imprinted and we suggested that this could have an impact on the individual’s phenotype. What is also special about this case is that to our knowledge this is the first known total isodisomy on chromosome 18 covering the whole chromosome described so far and was published as a case report in Frontiers in Genetics.
To conclude, our UPD detection pipeline and the webapp altAFplotter are now part of our routine diagnostic pipeline in our institute, and there is still work to be done in refining cut off´s based on data and experiences generated through everyday use. Using this tool, the diagnostic gap for UPD detection could be closed and the diagnostic yield in our institute elevated. Hopefully some more UPD cases can be solved in the future and the assessment of the risk of reoccurrence for other family members in cases with homozygous pathogenic variants will be more accurate if we have knowledge of a present UPD. 14 UPD cases within our cohort, nine of them likely pathogenic, could be detected and reported back to the affected families. For the final statements about the clinical significance of these cases, the results of the methylation site analysis data need to be further evaluated. This dissertation is based on three publications. The manuscript about our UPD detection method includes me as a co-author and the other two publications about the cohort analysis and our case report include me as a first author.:Table of Contents:
1.Abbreviations .......................................................................................................................... Ⅱ
2.List of illustrations ................................................................................................................ Ⅲ
3.Introduction ............................................................................................................................. 1
3.1 Genetics of UPD .......................................................................................................... 1
3.2 Mechanism of UPD formation ...................................................................................... 2
3.3 Clinical Significance and Imprinting ............................................................................ 4
3.4 Methods of UPD detection ............................................................................................ 8
3.5 Rationale ....................................................................................................................... 8
3.6 Results ........................................................................................................................... 9
4. Publications .......................................................................................................................... 10
4.1 altAFplotter: a web app for reliable UPD detection in NGS diagnostics ................... 10
4.2 Automatized detection of uniparental disomies in a large cohort .............................. 18
4.3 Case report: Complete paternal isodisomy on chromosome 18 induces methylation
changes in PARD6G-AS1 promotor in a case with arthrogryposis ................................... 29
5. Summary .............................................................................................................................. 36
6. References ........................................................................................................................... 39
7. Presentation of personal scientific contribution ................................................................... 42
8. Declaration of Authorship .................................................................................................... 48
9. Curriculum vitae ................................................................................................................... 49
10. Acknowledgements ............................................................................................................ 5
Complementary Extensions of Concept Learning in Description Logics
No full textInstead of a sheer amount of data, having information with explicit semantics is a key enabler for turning data into knowledge. Data provided in semantically explicit formats allows machines to calculate insights from it. A technique that leverages the explicit semantics and automated reasoning capabilities is Concept Learning (CL). The broad idea of CL is to learn concept descriptions from data. It has proven successful in many real-world use cases and is advantageous over black-box systems. In many application fields, a learned classifier that can provide further insights, besides mere classification capabilities, is favorable.
However, despite its solid formal foundation, we see unsolved challenges in Concept Learning. A first challenge (Ch1) concerns the lack of solutions for a systematic evaluation of CL algorithms. Another challenge (Ch2) is the size of the search space, especially in application domains that are modeled in great detail, and, hence, have a very large number of defined concepts. In many existing CL algorithms, this causes a ‘combinatorial explosion’ regarding the generated candidate concepts. This circumstance requires new exploration strategies to cover the search space more efficiently. Finally, we see a rise in the amount of spatial data. However, the knowledge representation formalisms underlying Concept Learning usually do not cover the extraction of spatial relations. We see the integration of these implicit relations into CL as a further challenge (Ch3).
To address Ch1, we propose a framework for the automatic evaluation of different CL algorithms. We perform a systematic literature review to find a diverse and established set of CL scenarios. We implement a free software system called SML-Bench to run a suite of freely configurable benchmark runs and get one unified evaluation result.
Regarding Ch2, we found that it is worth considering mechanisms for balancing the explorative and exploitative behavior of CL algorithms, especially in cases involving more extensive background knowledge bases. While more explorative algorithms typically cover a wider area of the search space and thus generate a more diverse set of hypotheses, algorithms dominated by exploitative behavior focus on areas where the greatest improvement is achieved. However, in their extremes, a fully explorative, i.e., random, CL algorithm would be inefficient, and greedy, purely improvement-driven algorithms would be prone to getting stuck in local optima. To better fine-tune the search space exploration behaviors of CL algorithms, we reviewed the literature on metaheuristics. As a first in-depth study, we focus on a CL approach based on the Simulated Annealing meta-heuristic and demonstrate performance improvements that surpass the existing state-of-the-art.
To fully utilize spatial information in Concept Learning and address Ch3, we focus on a set of 20 spatial relations, define them formally, demonstrate how to automatically infer these relations from polygon data, and explain how to integrate them into the CL formalism. Furthermore, we discuss their computational complexities. We apply this extension in different experiments to show its effectiveness.:1 Introduction
1.1 Motivation
1.2 Research Questions
1.3 Overview of the Thesis
2 Preliminaries
2.1 Knowledge Representation and Reasoning
2.1.1 Foundations
2.1.2 Ontology and Ontologies
2.1.3 Automated Reasoning
2.2 First-Order Logic
2.3 Horn Clause Logic and Logic Programming
2.3.1 Logic Programming
2.3.2 Prolog
2.4 Description Logics
2.4.1 Relation Between Description Logics and Logic Programming
2.5 Semantic Web Technologies
2.6 Machine Learning
2.7 Inductive Logic Programming
2.8 Concept Learning
3 Related Work
3.1 Concept Learning
3.1.1 Inductive Learning
3.1.2 Inductive Learning in Logic Programming (ILP)
3.1.3 Inductive Learning in Description Logics and OWL (Concept Learning)
3.2 Benchmarking Concept Learning
3.3 Metaheuristics for Concept Learning
3.4 Spatial Concept Learning
4 The DL-Learner Framework
4.1 Learning Problems
4.2 Overview of the Framework
4.3 Learning Algorithms
4.3.1 Refinement Operator-based Algorithms
4.3.2 OWL Schema Learning Algorithms
4.3.3 Other Algorithms
4.4 Implementation
4.5 Contributions
5 Benchmarking Concept Learning
5.1 Challenges of Structured Machine Learning
5.2 Benchmarking Structured Machine Learning Algorithms
5.3 Datasets
5.4 SML-Bench Framework
5.4.1 Architecture
5.4.2 Benchmark Scenarios
5.4.3 Benchmarked Tools
5.4.4 Benchmark Settings
5.4.5 Available Learning Systems
5.5 Evaluation
5.6 Discussion
6 Extended Metaheuristics for Concept Learning
6.1 Approach
6.1.1 Refinement-based Concept Learning in Description Logics as a Hill Climbing Approach
6.1.2 Simulated Annealing
6.1.3 Simulated Annealing in Concept Learning
6.2 Empirical Evaluation
6.2.1 Evaluation Setup
6.2.2 Results
6.3 Discussion
7 Spatial Concept Learning
7.1 A Calculus for Polygon Data
7.2 Spatial Inference on Geospatial Polygon Data in Description Logics
7.3 Spatial Concept Learning
7.4 Implementation
7.5 Evaluation
7.5.1 Spatial Inference
7.5.2 Spatial Concept Learning
8 Conclusions
8.1 The DL-Learner Framework
8.2 Benchmarking Concept Learning
8.3 Extended Metaheuristics for Concept Learning
8.4 Spatial Concept Learning
A Curriculum Vitae
Bibliography
Inde
How Crystallographic Orientation-Induced Fibrinogen Conformation Affects Platelet Adhesion and Activation on TiO2
No full textControl of protein adsorption is essential for successful integration of healthcare materials into the body. Human plasma fibrinogen (HPF), especially its conformation is a key upstream regulator for platelet behavior and thus pathological clot formation at the blood-biomaterial interface. A previous study by the authors revealed that the conformation of adsorbed HPF can be controlled by rutile surface crystallographic orientation. Therefore, it is hypothesized that pre-adsorbed HPF on specific rutile orientation can regulate platelets adhesion and activation. Here, it is shown that platelets exposed to the four low index (110), (100), (101), (001) facets of TiO2 (rutile) exhibit surface-specific behavior. Scanning electron microscopy (SEM) observations of platelets morphology and P-selectin expression measurement revealed that on (110) facets, platelets adhesion and activation are suppressed. In contrast, extensive surface coverage by fully activated platelets is observed on (001) facets. Platelets' behavior has been linked to the HPF conformation and thereby availability of platelet-binding sequences. Atomic force microscopy (AFM) imaging supported by immunochemical analysis shows that on (110) facets, HPF is adsorbed in trinodular conformation rendering the γ400-411 platelet-binding sequence inaccessible. This research has potential implications on the bioactivity of different materials crystal facets, reducing the risk of pathological clot formation and thromboembolic complications