Dissertations and theses


Digital dissertations and theses from the University of Helsinki. Contains both open access theses as well as metadata records.

Collections in this community

Recent Submissions

  • Poutanen, Heidi (Helsingin yliopisto, 2022)
    The thesis focuses on representations of organic production and consumption in content marketing in the grocery and alcohol trade, adopting a social semiotic approach. Accordingly, the study examines how meanings related to organic production and consumption are constructed in editorial texts in Finland Swedish customer magazines, and how the senders and readers, constructed by the text, are positioned in terms of ecological practices. The study also contributes new knowledge about texts in customer magazines. Methodological tools are taken from Systemic-Functional Linguistics, legitimation analysis and social semiotic multimodality research. The first three sub-studies concentrate on written discourse, and the fourth sub-study concentrates on both written and visual discourse. The material consists of articles from three Finland Swedish customer magazines, published in 2016 by leading companies in the grocery and alcohol trade in Finland. The study shows how representations of ecological individuals and an ecological community live side by side in the texts of customer magazines. Regarding ecological individuals, it is above all celebrities from cultural life who present the consumer’s voice in the customer magazines. In addition, local producers get a say and have active roles in ecological practices. A sense of a community, in turn, is often constructed by the inclusive pronoun vi (we), with the meaning ‘we in Finland’/ ‘we Finns’. Through the collectivization of social actors, the consumption of organic products is constructed as a social practice that everyone does, or at least most do. The Finnish cultural context is frequently thematized. The representations evoke memories of the past and connotations of an authentic and traditional Finnish food culture. The representations of idyllic farms are associated, by multimodal relational processes, with specific products. In customer magazines, the ecological is frequently thematized, but at the same time, readers’ ecological behaviour is reduced to ecological consumption. Consumption of organic products appears in the texts as a simple and worry-free way to achieve an environmentally friendly lifestyle. In customer magazines, consumers are assumed to have great power to influence the world through their consumption patterns. Customer magazines are often seen as a medium for self-presentation. They offer a communication channel for building an ecological company image. The study shows, however, that the most active roles in the texts are given to the producers and consumers. The reseller between them, the publishing company, remains much more invisible in relation to ecological practices.
  • Laine, Hanna (Helsingin yliopisto, 2021)
    Salivary gland cancer (SGC) comprises less than 5% of head and neck malignancies. Adenoid cystic carcinoma (ACC) is denoted as the second most common SGC worldwide. In Finland, ACC is the most common histological subtype according to a nationwide study. ACC is a slow-growing neoplasm and has a tendency for perineural invasion. ACC shows three distinct histological growth patterns: cribriform, tubular, and solid. Surgery is the pivotal treatment modality, but treatment is modified according to tumor site, biological aggressiveness, and stage of the disease, which is determined according to the Union for International Cancer Control (UICC) Tumor-Node-Metastasis (TNM) classification. Postoperative radiotherapy is recommended for all patients, but chemotherapy is used mainly for inoperable, recurrent, or metastatic disease. Recurrencies affect approximately 50% of patients. Especially local and distant recurrent tumors are fairly common, with distant metastasis being more frequent and lungs the most common site. ACC has a good 5-year disease-specific survival (76-88%), but the 10-year survival (34-67%) is clearly worse. Prognostic factors affecting survival are tumor site, TNM classification, histology, surgical margin status, and distant metastasis. In this thesis, the goal was to collect all patient data and tumor samples of the minor salivary gland ACC patients diagnosed between years 1974 and 2012 in the Helsinki University Hospital area. Patient and tumor characteristics, treatment, outcome, and their associations were studied. To evaluate the viral role in ACC samples, the presence of three polyomaviruses were assessed genotyping of 24 human papillomaviruses (HPV) was performed. Furthermore, by immunohistochemistry matrix metalloproteinase (MMP)-7, -8, -9, -15, and -25 and antizyme inhibitor (AZIN) 1 and 2 in ACC were studied. The most common tumor site was the palate. Of patients, 94% were treated with curative intent. Moreover, 53% of patients suffered from recurrent ACC of which 36% were local, 12% regional, and 52% distant. Almost all distant metastases appeared within 10 years. The 5- and 10-year overall survival and disease-specific survival were 70% and 79%, and 42% and 52%, respectively. Stage I ACC patients had better survival than patients with higher stages (II-IV). Of interest, John Cunningham polyomavirus (JCPyV) DNA was found in 10% of the tumor samples. In the immunohistochemical studies on MMPs, abundant MMP-7 and -25 were associated with better survival. High tumorous MMP-9 associated with advanced stage and high MMP-15 immunoexpression with poorer survival. Intriguingly, abundant MMP-9 immunoexpression in inflammatory cells in the vicinity of ACC and in luminal material of pseudocysts of ACC associated with better survival and fewer local recurrent tumors. Immunoexpression of AZIN2 was abundant in well-differentiated tumor tissue (cribriform and tubular), but in the solid pattern the expression was negative or mild. AZIN2 immunoexpression associated with better survival. To conclude, these results show that especially stage II ACC should be considered as advanced disease and patients would benefit from more aggressive treatment. Follow-up time should be prolonged for at least ten years. JCPyV could participate in the pathogenesis of a small proportion of ACC. MMPs could participate in ACC carcinogenesis by tissue modulation, activating different signaling pathways, and by immunomodulation. MMP-7, -9, -15, and -25 are related to prognostic factors. High AZIN2 immunoexpression in well-differentiated ACC could be related to a functioning vesicle transport system of tumor cells that no longer exists in poorly differentiated ACC tissue. AZIN2 could be a prognostic factor for better survival of ACC patients.
  • Wrigley, Sam (Helsingin yliopisto, 2021)
    This thesis looks at whether the GDPR can efficiently and effectively promote its goals and aims when personal data is processed by bots and AI-related technologies, specifically when that data is processed on the basis of legitimate interest under art. 6(1)(f), as interpreted by the principle of fairness per art. 5(1)(a). To do this, I have broken the topic into three research questions: How should we understand the concept of fairness under art. 5(1)(a); can the concept of fairness, as understood under art. 5(1)(a), help to address the shortcomings of legitimate interest processing under art. 6(1)(f); and is the use of legitimate interest processing, as interpreted through the lens of fairness, an efficient and effective tool for supporting the GDPR’s goals and aims when personal data is processed by bots and AI-related technologies? As these questions cover legal, philosophical and technical issues, I embrace a socio-legal approach and incorporate sources from each field in my research. For the first question, I look at the existing interpretation of fairness and find that the concept has (at best) relied on innate or implicit judgements and (at worst) been used as mere window-dressing. I therefore use Rawls’ original-position thought experiment to develop a flexible test that can provide a framework for explicit and open discussions about whether something should be considered fair. I then compare this test to other uses of fairness in law and, finding it compatible, propose a test which can be used for evaluating fairness under the GDPR, art. 5(1)(a). For the second, I examine the existing guidance and find that, among other things, the balancing act required by art. 6(1)(f) struggles to deal with the inherent subjectivity of the interests involved. This then leads to a lack of consistency between guidelines, difficulties in identifying and weighing interests, and uncertainty as to how the balancing act should be calibrated. I argue that by incorporating the fairness test outlined in the first question as an interpretive lens for the legitimate interest balancing test, we can incorporate this subjectivity in a more reliable and structured manner, which helps to resolve or mitigate the issues identified and address the shortcomings of art. 6(1)(f). Finally, I look at how this test might operate when personal data is processed by bots and AI-related technologies. I first consider the nature of these technologies and what existing conversations can tell us about them. This includes, inter alia, an examination of existing discussions on ethical usage of such technologies and what factors may be relevant in a balancing test. I then consider how the legitimate interest fairness test developed above might be applied when personal data is processed by bots and AI-related technologies and how the circumstances relevant to this context might be applied. Finally, I evaluate the test and conclude that, if used properly, it can be used to support the GDPR's goals when bots and AI-related technologies are used to process personal data, noting factors that must be considered and the strengths and weaknesses of the approach described.
  • Silén, Yasmina (Helsingin yliopisto, 2021)
    Background and aims: Eating disorders are severe mental health issues that undermine psychological and physical health and quality of life. This thesis aimed to investigate the occurrence of eating disorders in a community sample of adolescents and young adults. Eating disorders were defined using the Fifth Edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). A second major aim was to describe how often individuals with eating disorders were detected and received treatment and what kind of treatment was offered. Finally, the natural course and outcome of different subcategories of eating disorders were also examined. Methods: The study used two population-based Finnish twin datasets and one clinical dataset. FinnTwin12 followed all twins born in Finland between 1983 and 1987, whereas FinnTwin16 followed all twins born between 1975 and 1979, from adolescence to adulthood. From the FinnTwin12 data, we studied the occurrence, detection and treatment of eating disorders in health care and the natural course of these diseases. From the FinnTwin16 data, we examined the effects of diagnostic changes on the occurrence and prognosis of anorexia nervosa. From the clinical data of the Helsinki and Uusimaa Hospital District (HUS) Adolescent Psychiatry Eating Disorders Unit, we examined the treatment and prognosis of adolescents with typical and atypical anorexia nervosa. In all studies, we analysed the outcome of eating disorders using survival analysis. Results: Eating disorders were common. Up to one in 6 females and 1 in 40 males had suffered from an eating disorder during their development towards adulthood. Changes in diagnostic criteria increased the lifetime prevalence of anorexia nervosa by more than half and increased diagnostic heterogeneity. Overall, anorexia nervosa and related subthreshold symptoms were prevalent among females: One in 10 young women had suffered from a restrictive eating disorder by early adulthood. We also found that eating disorder symptoms in a community setting were diverse, and many reported eating disorder symptoms that could not be clearly labelled. This was particularly true among boys and men. Many individuals with eating disorders also described an unmet need for care; healthcare professionals diagnosed only one-third and even fewer received treatment. In addition, eating disorder symptoms were highly persistent: Five years after disease onset, less than two-fifths of the females and two-thirds of the males had recovered. The likelihood of recovery was similar between those who had and who had not received treatment, but more severe cases were more likely to receive treatment. Conclusions: Overall, this thesis showed that eating disorders are common, and their symptoms are highly diverse among Finnish adolescents and young adults. Considering the magnitude of the problem, detection and treatment approaches for eating disorders are still inadequate and mainly focused on typical presentations of eating disorders. In addition, eating disorder symptoms often persisted for years. Future research should determine how the prevention and detection of eating disorders could be improved in Finland. The threshold for access to treatment should also be lowered, and additional interventions should be developed. Future studies should investigate whether these actions could eventually lead to better outcomes.
  • Zhang, Feng (Helsingin yliopisto, 2021)
    By the virtue of the divers physicochemical properties, nanomaterials have emerged as a powerful platform to improve the pharmacokinetic properties of drug molecules. Furthermore, in view that the created or constructed materials own the same size level of biomacromolecules, and can be endowed with various biochemical functions, nanotechnology has been treated as the most promising technology to develop smart therapeutics with flexible articifal controllability to innovate the current medicine. Generally, the anchor or the breakthrough point of these technologies depends on the nanomaterials-based drug delivery systems (DDS), which has been in tremendous development for more than thirty years. However, clinical transition of DDS are facing great challenges related to the insufficient targeting accumulation in the cells/tissues. The in vivo delivery of nanotherapeutics is a multi-stage process and needs to conquer multiple biological barriers. In this case, conventional DDS is not competent enough to cope with the various biological barriers. Thus, the transformable design of DDS with tunable surface properties in responsive to stimuli-signals at different barriers are in urgent need. In this thesis, the focus was on constructing DDS with different stimulus and functions to achieve the task-oriented NPs’ transformation, including surface charge inversion, sequential antifouling surface, in situ size modulation and multi-stage signal interaction. Firstly, it was fabricated a PSi DDS with receptor-mediated surface charge inversion. The negatively charged surface can convert into positive charged surface in response to cancer micro- environment, driven by the AS1411-nucliolin interaction. Secondly, it was modified the biotin- PEI nanoparticles with acrylates-ortho-nitrobenzyl-PEG5000, which further acted as the primary antifouling surface to prevent the formation of protein corona and avoid off-targeting effect. After UV-irradiation, the PEG surface can be cleaved to generate carboxyl group on the biotin- PEI surface, forming a secondary zwitterionic anti-fouling surface. This dual-antifouling modification can efficiently avoid protein adsorption on the NPs’ surface in human serum. Moreover, the secondary zwitterionic surface can guarantee the effective exposure of active targeting segments for improving cell uptake. Simultaneously, the reduced size facilitates deep tissue penetration of the NPs. Thirdly, it was constrcuted a photo-driven size tunable DDS, which can increase the size after tumor accumulation in situ to prolong the tumor retention time and also to improve cancer cell uptake. Finally, it was developed a multistage signal interactive system on NPs through integrating the Self-peptide and YIGSR peptide into a chimeric form, with a hierarchical signaling interface involving “don’t eat me” and “eat me” signals. This biochemical transceiver can act as both signal receiver for amantadine to achieve NP transformation and signal conversion, as well as the signal source with different signals by reversible self-mimicking. Throughout chemical and biological testing, it was demonstrated these designed DDS have efficient signal-induced transformation behavior and enhanced controllability of NP-cell interactions for improving the cancer therapeutic efficacy. Overall, this dissertation provides a new insight of targeting drug delivery and nano-tools to facilitate clinical transition of nanomedicines.
  • Maarala, Ilari (Helsingin yliopisto, 2021)
    High-throughput sequencing (HTS) technologies have enabled rapid DNA sequencing of whole-genomes collected from various organisms and environments, including human tissues, plants, soil, water, and air. As a result, sequencing data volumes have grown by several orders of magnitude, and the number of assembled whole-genomes is increasing rapidly as well. This whole-genome sequencing (WGS) data has revealed the genetic variation in humans and other species, and advanced various fields from human and microbial genomics to drug design and personalized medicine. The amount of sequencing data has almost doubled every six months, creating new possibilities but also big data challenges in genomics. Diverse methods used in modern computational biology require a vast amount of computational power, and advances in HTS technology are even widening the gap between the analysis input data and the analysis outcome. Currently, many of the existing genomic analysis tools, algorithms, and pipelines are not fully exploiting the power of distributed and high-performance computing, which in turn limits the analysis throughput and restrains the deployment of the applications to clinical practice in the long run. Thus, the relevance of harnessing distributed and cloud computing in bioinformatics is more significant than ever before. Besides, efficient data compression and storage methods for genomic data processing and retrieval integrated with conventional bioinformatics tools are essential. These vast datasets have to be stored and structured in formats that can be managed, processed, searched, and analyzed efficiently in distributed systems. Genomic data contain repetitive sequences, which is one key property in developing efficient compression algorithms to alleviate the data storage burden. Moreover, indexing compressed sequences appropriately for bioinformatics tools, such as read aligners, offers direct sequence search and alignment capabilities with compressed indexes. Relative Lempel-Ziv (RLZ) has been found to be an efficient compression method for repetitive genomes that complies with the data-parallel computing approach. RLZ has recently been used to build hybrid-indexes compatible with read aligners, and we focus on extending it with distributed computing. Data structures found in genomic data formats have properties suitable for parallelizing routine bioinformatics methods, e.g., sequence matching, read alignment, genome assembly, genotype imputation, and variant calling. Compressed indexing fused with the routine bioinformatics methods and data-parallel computing seems a promising approach to building population-scale genome analysis pipelines. Various data decomposition and transformation strategies are studied for optimizing data-parallel computing performance when such routine bioinformatics methods are executed in a complex pipeline. These novel distributed methods are studied in this dissertation and demonstrated in a generalized scalable bioinformatics analysis pipeline design. The dissertation starts from the main concepts of genomics and DNA sequencing technologies and builds routine bioinformatics methods on the principles of distributed and parallel computing. This dissertation advances towards designing fully distributed and scalable bioinformatics pipelines focusing on population genomic problems where the input data sets are vast and the analysis results are hard to achieve with conventional computing. Finally, the methods studied are applied in scalable population genomics applications using real WGS data and experimented with in a high performance computing cluster. The experiments include mining virus sequences from human metagenomes, imputing genotypes from large-scale human populations, sequence alignment with compressed pan-genomic indexes, and assembling reference genomes for pan-genomic variant calling.
  • Stepanova, Polina (2021)
    Parkinson’s disease (PD) and Huntington’s disease (HD) are characterized by loss of function or death of definite cell populations in the basal ganglia. HD is triggered by an expanded polyglutamine tract (glutamine repeats) in the huntingtin protein, leading to misfolding of the protein and subsequent accumulation of mutant huntingtin (mHtt) in the nuclei of different types of neurons. In contrast, in PD, dopamine (DA) neurons in the substantia nigra pars compacta (SNpc) degenerate; moreover, there is also the accumulation of misfolded proteins associated with the neuropathology of PD. Additionally, endoplasmic reticulum (ER) stress has been detected in both of these diseases. Currently, no effective treatment for PD or HD is available to slow, stop, or reverse the progression of neurodegeneration. Cerebral dopamine neurotrophic factor (CDNF) is an evolutionarily conserved protein with neurotrophic properties. CDNF protects and restores the function of DA neurons in preclinical models of PD more effectively than other neurotrophic factors (NTFs), making it a promising drug candidate as a disease-modifying treatment of PD. Additionally, CDNF was safe and well-tolerated, showing therapeutic effects in some PD patients in phase 1/2 clinical trials. This thesis aimed to investigate the potential of CDNF as a drug candidate in cellular and rodent models of both HD and PD. We focused on studying the effect of coadministration of CDNF with another NTF, glial cell line-derived neurotrophic factor (GDNF), in a 6-hydroxydopamine (6-OHDA) rat model of PD. Moreover, we wanted to evaluate the difference in the mode of action between the two factors in this model. We found an additive neurorestorative effect after intrastriatal CDNF and GDNF coadministration in the 6-OHDA toxin model of PD. CDNF alone and in combination with GDNF showed a trend toward an increase in the density of tyrosine hydroxylase (TH)-positive fibers in the lesioned striatal area, and moreover, the combination of the two factors significantly protected TH-positive cells in the SNpc area. CDNF activated the PI3K/AKT pathway, whereas GDNF activated two signaling pathways: PI3K/AKT and MAPK/ERK. We found a delay in activating the prosurvival pathway by CDNF compared to GDNF. Additionally, in comparison with GDNF, CDNF alone significantly enhanced the phosphorylation of ribosomal protein S6 downstream of the PI3K/AKT pathway in lesioned rats. Finally, CDNF, but not GDNF, reduced striatal levels of some ER stress markers in the above model. PD and HD show a wide range of commonalities, and as CDNF has been successful in preclinical trials and demonstrated a positive effect in clinical trials of PD, we expected to observe some beneficial effects of CDNF administration in the experimental HD models. We studied the neuroprotective and neurorestorative potential after different delivery paradigms of CDNF in preclinical models of HD to find the optimal paradigm for possible future application in clinical trials. First, we investigated the protective effects of CDNF in in vitro and in vivo quinolinic acid (QA) toxin models of HD. We demonstrated that CDNF improved motor coordination in QA-lesioned animals, which could be explained by the neuroprotective and neurorestorative effects of CDNF in the cell populations that are vulnerable in HD. Moreover, we found a protective effect after a single intrastriatal injection of CDNF in an in vivo QA-lesion model of HD. Second, we tested the therapeutic efficacy of CDNF in a transgenic mouse model of HD. CDNF was delivered as a chronic intrastriatal infusion using Alzet minipumps for N171-82Q mice. Chronic CDNF administration ameliorated the behavioral deficits and showed a trend toward reduced the nuclear staining and intranuclear inclusions in N171-82Q transgenic mice. Additionally, chronic delivery of CDNF demonstrated a trend toward a decrease in ER stress markers in the striatum in the above model. Furthermore, we found a significant increase in hippocampal BDNF mRNA levels after chronic CDNF administration in N171-82Q mice. In conclusion, CDNF shows a unique beneficial effect in several models of HD independent of the etiology of the disease. Notably, this thesis reports the first beneficial effects of CDNF in different models of HD.
  • Bourgeot, Liisa (Helsingin yliopisto, 2021)
    Gustav Shpet (1879-1937) is one of the most noteworthy thinkers in the history of Russian philosophy. Yet he was executed in Stalin’s Great Terror and removed from philosophy books for nearly half a century. Shpet was rediscovered in the late 1980s, and the scholarship around him has since grown, both in Russia and abroad. Especially strong interest has been directed towards his reading of Husserl’s phenomenology. Shpet was Husserl’s student in Göttingen in 1912–14 and published his own considerations about transcendental phenomenology on his return to Moscow. But while he has come to be seen as a pioneer in Russian phenomenology, Shpet’s unconventional interpretation has also raised concerns. According to many, Shpet’s reading of Husserl was founded on a misinterpretation; he is seen as either unwilling to follow the latter’s idea or incapable of doing so. It is also frequently proposed that Shpet’s unorthodox approach may be explained by a certain Russianness at the foundation of his system. The traditional philosophical standpoint in Russia emphasized ontological questions and the collectiveness of experience. It is suggested that Shpet thereby ‘Russianized’ Husserl by replacing his transcendental subjectivity with a collective consciousness. The dissertation proposes a new reading of Shpet’s interpretation of Husserl. It first discusses the Russian roots of Shpet’s thinking and then presents a novel analysis of his "Iavlenie i smysl" (1914), comparing it not only to Husserl’s "Ideen I" (1913) but also "Logische Untersuchungen" (1900–01) and "Formale und Transzendentale Logik" (1929). The study maintains that Shpet not only followed through the transcendental turn but proposed themes and ideas which Husserl himself arrived at in the 1920s. It is suggested that the similarity between Shpet’s and Husserl’s approaches might be traced back to their discussions in Göttingen. The study’s second principal topic is Shpet’s theory of the inner form of the word. After the October Revolution, his phenomenology evolved into a philosophy of language and art. While his theory was known among the Moscow formalists, it was also deemed archaic and unscientific. The concept of the inner form seemed to carry echoes of Aleksandr Potebnia, against whose romantic theory of the formalists fought. In contrast to this reading, the dissertation suggests that Shpet’s theory was in fact closely related to his earlier phenomenological ideas. It is maintained that especially in his 1922–23 "Esteticheskie fragmenty", the content of Shpet’s concept was in a strict sense phenomenological.
  • Tuomainen, Katja (Helsingin yliopisto, 2021)
    Head and neck cancer (HNSCC) is the 8th most common cancer globally. Despite improved therapy, the 5-year survival rate has remained stagnant, at approximately 50%. Current preclinical in vitro anticancer compound testing features a low predictive value. Traditionally, in vitro cancer cell studies are conducted on a plastic surface or using animal–derived extracellular matrices, thereby overlooking the important interaction between cancer cells and the human tumor microenvironment. To overcome this problem, we have developed a human tumor leiomyoma–derived matrix Myogel. The aim of this PhD project was to develop reliable in vitro methods for anticancer drug testing by providing a human TME for HNSCC cells. The first approach was to investigate whether the use of a Myogel matrix could improve the reliability of drug screening compared to conventional plastic or the typically applied mouse sarcoma–derived matrix, Matrigel. We applied a high-throughput drug screening (HTS) to test 19 anticancer compounds, targeting EGFR, MEK, or PI3K/mTOR on 12 HNSCC cell lines under five different culturing conditions; cells on a plastic surface and on top of or embedded in Myogel or Matrigel. We found that cells cultured in Myogel were less sensitive to the EGFR and MEK inhibitors than cultured in Matrigel or on plastic. Additionally, Myogel cultured cells reflected better the response rates of EGFR inhibitor clinical trials compared to the other cultures. Therefore, we concluded that Myogel improves the predictability of in vitro anticancer drug testing compared to the plastic and Matrigel. Next, we performed a high-throughput drug screen to identify synergistic combinations using a compound library (n=396) and irradiation using HNSCC cell lines cultured on Myogel. The Bcl-2/Bcl-xL inhibitor navitoclax, which emerged as the most promising radiosensitizer, exhibited a synergy with irradiation in 13 HNSCC cell lines. Additionally, the navitoclax–irradiation combination halted the cell cycle and increased the irradiation-induced apoptosis. Overall, we demonstrated that the navitoclax–irradiation combination resulted in strong synergistic antitumor effects in HNSCC cell lines, and thereby it might possess therapeutic potential for HNSCC patients. The third, we manufactured a fully human in vitro microfluidic chip assay to test immunotherapeutic agents for personalized medicine purposes for HNSCC patients. First, the assay was run using the oral tongue cancer cell line (HSC-3) embedded in a Myogel–fibrin mixture and immune cells, isolated from healthy blood samples. After successful testing, the chip was applied for freshly isolated patent cells. Immune checkpoint inhibitors, IDO1 inhibitor (NLG-919) and a PD-L1 antibody were applied to chips and the immune cell migration and cancer cell proliferation were measured over three days. We found that the immune cell migration was associated with cancer cell density. The IDO1 inhibitor increased the immune cell migration towards cancer cells in HSC-3 and in two patient samples. In the future, this assay could possibly be utilized to predict an individual patient’s immunotherapy response against HNSCC.
  • Tapper, Helena (Helsingin yliopisto, 2021)
    The focus of this thesis is the study of the Knowledge Society discourse and policies of international development organisations (the UN, AU, IADB, ITU, UNESCO) and two national governments (South Africa and Finland). The research question is “Why and how the policies of national and international organizations have promoted and developed Knowledge Society policies generally and for development in the Sub-Saharan Africa, in particular?” The data for the thesis traces its origins to the author being immersed for 15 years in Knowledge Society policy implementation and study, in and across five regions (Central America, the Caribbean, South America, Europe, and Africa). This thesis involves an introductory essay, four articles and one book chapter. The author’s framework in this study blends earlier research on the historical progression of scholarly thought on Knowledge Society discourse and policies with public discourse on the content and new directions of what defines Knowledge Society policy. The author has engaged in participant observation and many rounds of analysis, iteration, reflection, and interpretation as it pertains to the subject matter. Since 2015, she has carried out further data collection, analysis and interpretation for purposes of theory building . The data collected, analysed and interpreted in this study thus pertain to the history and evolution of Information and Knowledge Society and its manifestation in national policies of governments and the development policies of international development organisations globally, and more specifically in the context of Africa. Following the methodology of Gioia, Corley and Hamilton (2013), the author categorized her foregoing data on Knowledge Society discourse and policies into 21 first-order codes or policy themes. The identified themes are each contextually embedded in the time of their publication. Some of the policy themes in the essay and publications have remained the same from 2000 to 2015. Other policy themes, like gender and local economy, have only appeared on the agenda more recently. The three developmental ‘waves’, ‘phases’ or ‘dimensions’ of Knowledge Society are: An efficiency-and economically oriented wave starting as discourse in the 1970s, an information and communication technology wave starting in the 1980s and globalization and development wave starting in the 1990s. Within each of these waves and across them, Knowledge Society has either contributed to development of national economies and societies in the Global North (such as those of Europe or the United States) or in the Global South (such as those in Sub-Saharan Africa). Calls for further research include the study of the impact of the COVID 19 pandemic on development in selected countries in the Global North and the Global South.
  • Jagne-Soreau, Maïmouna (Helsingin yliopisto, 2021)
    In this thesis, we analyse the representation of racialized second-generation migrants, mixed-race persons and transnational adoptees in contemporary Nordic literature. The work is an article-based dissertation including a theoretical contextualization. The first article “Half-Norwegian, Real Foreigner” (2017) is an analysis of Maria Navarro Skaranger’s novel Alle utlendinger har lukka gardiner (2015) from a postnational perspective. The second article “Yahya Hassan, a modern Cervantes” (2018) brings us to Denmark, and is a reading of Yahya Hassan’s eponymous debut poetry collection (2013), and proposes an alternative reading of the work in the picaresque tradition. The third article “Adrian Perera and the art of wog poems” (2018) is an analysis of Perera’s poetry suite White Monkey (2017), which opens up discussions on whiteness, non-whiteness, racism and racialization in today’s Finland. The fourth article “To wake up as suedi” (2019) is a literary reading of Erik Lundin’s first EP “Suedi” (2015) and focuses on the theme of betweenship in a Swedish context. Finally, the fifth article “I don’t write about me, I write about you” (2021) presents four major iterating motifs in a dozen works and sheds light on the aspect of postmigration literature being a trans-Nordic phenomenon in contemporary Nordic literature. Our results include reflections around the themes of identity, generation conflict, racism and hegemonic whiteness, the effect of authenticity and the game with the expected reader(s). Based on these findings we propose the acknowledgement of a new literary phenomenon that we call postmigration literature.
  • Johari, Mridul (Helsingin yliopisto, 2021)
    Molecular genetics of late-onset neuromuscular diseases is often challenging to study due to limited patient and family material. This thesis aimed to unravel the genetics of late-onset myopathy patients with inclusion body myositis (IBM) and yet unknown inherited distal myopathy phenotype. IBM is an inflammatory myopathy of idiopathic nature showing characteristic quadriceps and finger flexor muscle weakness and rimmed vacuolar pathology. The molecular etiology of IBM is of key interest due to co-existing cytotoxic T-cell activity. However, the precedence of degenerative pathology and autoimmune features remain unclear, hampering the eventual development of appropriate therapeutic options. Using DNA and RNA sequencing-based data analysis, we aimed at understanding the different components of the molecular pathomechanisms in IBM. Using a limited case-control study, we identified the association of HLA-DQB1 in Finnish IBM patients. Additionally, by performing a comprehensive transcriptomics analysis we observed differential expression and splicing patterns in genes associated with maintaining calcium homeostasis, particularly during different T-cell activity and regulation stages. Disturbed antigen-driven T-cell hyperactivity and eventual loss of T-cell apoptosis could be one of the mechanisms behind the refractoriness of immune therapies in IBM patients. Patients with late-onset rare diseases often experience a long diagnostic journey in multiple and often invasive diagnostic procedures and ineffective symptomatic treatments. Due to limited study material, it is often difficult to ascertain the molecular diagnosis of previously unknown inherited myopathies. Identifying new rare disease-causing genes requires collaborations from different diagnostic centers, sharing deep phenotypic and genomic data. We identified a novel type of slowly progressing late-onset distal myopathy with rimmed vacuoles caused by mutations in the small muscle protein X-linked (SMPX) gene in patients from five different countries. Our genetic analysis revealed four different missense mutations, including two different founder mutations in Europe, indicating that the prevalence of this disease may be higher, and therefore SMPX should be considered in all unsolved male patients with late-onset rimmed vacuolar myopathy.
  • Zanjanizadeh Ezazi, Nazanin (Helsingin yliopisto, 2021)
    Biomaterials science denotes a multidisciplinary science combining materials engineering, biomedical engineering, and biology. Tissue engineering is defined as an engineering system field using biomaterials to regenerate, restore tissue, or improve its function. This thesis focuses on the fabrication of scaffolds and patches for bone and heart regeneration using a conductive polypyrrole (PPy) polymer. PPy is widely used in different tissue engineering applications due to its biocompatibility, high electrical properties and conductivity, ease to synthesize, and environmental stability, such as in water and air. First, the role of this polymer in attracting protein and osteoblast cells was investigated. The bone scaffold’s mechanical properties using PPy were analyzed. The results showed no difference between conductive and non-conductive bone scaffolds, presenting almost the same young modulus between the two systems. In addition, more proteins were adsorbed on the surface of the conductive polymers. The same results were obtained in conductive cardiac patches, which showed higher blood coverage on the conductive 2D patch using PPy compared to the non-conductive one. In addition, cellulose was used to 3D print and fabricate a porous conductive cardiac patch. The in vitro investigations showed that cell attachment and proliferation on the conductive bone scaffold and cardiac patches were higher. Biomineralization was induced in simulated body fluid on the bone scaffold’s conductive surface. PPy was also combined with bacterial cellulose (BC) to make conductive hydrogel for the treatment of myocardial infarction. The positive chains of PPy improved the loading of negative charged drug-loaded nanoparticles (NPs). All four systems were successfully developed, fabricated, and loaded with different drugs with different methods to make multifunctional bioengineered systems for drug delivery applications. The results showed that PPy controlled the release of the vancomycin, an antibiotic, and the GATA4-targeted compound 3i-1000, inside the pores of 3D bone scaffold and 3D printed cardiac patch, respectively, which suitable for long-term therapy. Furthermore, bone scaffold, 2D cardiac patch, and conductive BC were combined with microparticles of silica, NPs of porous silicon, and acetylated dextran, respectively, presenting the high potential of in situ drug delivery system enabling and dual delivery of drugs. Overall, this thesis demonstrated that PPy-based biocomposites are successfully developed showing high biocompatibility and high functionality within hard scaffolds and soft patches, and thus, they are very promising platforms for future bone and heart tissue regeneration applications in vivo studies.
  • Celikkol, Göksu (Helsingin yliopisto, 2021)
    This study investigates reactions to increasing cultural diversity in receiving societies, with a special focus on intergroup relations between the Finnish majority group and Russian-speaking immigrants in Finland. Additionally, this study also utilizes data from multiple immigrant-receiving societies around the world. While, for the most part, the study takes the perspective of national majority groups, it also looks at the experiences of Russian-speaking minority group members in Finland, Estonia and Norway. The study focuses on subjective experiences and perceptions of increasing cultural diversity, intergroup threats, and discrimination, and how these are related to intergroup attitudes and willingness to promote societal equality. These topics are addressed through the Social Identity Theory (SIT; Tajfel & Turner, 1979), Realistic Conflict Theory (RCT; Campbell, 1965; LeVine & Campbell, 1972; Sherif, 1966), and Group Threat theories (Blalock, 1967; Blumer, 1958; Schlueter & Scheepers, 2010), as well as related models such as the Rejection-Disidentification Model (Jasinskaja-Lahti et al., 2009) and Social Identity Model of Collective Action (SIMCA; van Zomeren et al., 2008). The study utilizes two cross-sectional survey datasets. The first includes data from Finland, Estonia, and Norway (N = 2296), gathered within the Determinants of an inclusive integration context (MIRIPS-FI) project and the collaborative DIMA project. This data allowed us to examine and compare intergroup relations between the national majority and Russian-speaking minority groups in these three countries. The second data set consists of university student samples representing national majority groups from 28 countries, including Finland. This international comparative data set was gathered to examine the relationship between perceived economic circumstances and reactions to immigration. Four sub-studies were designed to answer several research questions: how do national majority and minority group members perceive and react to increasing cultural diversity, and perceived disadvantage and the unequal treatment associated with it in receiving societies? How do their reactions explain and how may they be used to promote actions that support immigrants’ equal social standing in society, among both national majority and minority groups? First, the results of this study showed the interactive effects of actual and perceived cultural diversity on intergroup outcomes among national majority group members in Finland. Participants who lived in moderately diverse regions, but perceived high levels of cultural diversity in their neighborhood, reported higher levels of group discrimination and lower levels of trust towards the Russian-speaking immigrants than participants who lived in high and low cultural diversity regions. Regarding the moderating role of objectively assessed context, the home country’s objective wealth was found to moderate the relationship between subjectively perceived personal wealth and intergroup outcomes in 28 countries. Namely, in countries with low Human Development Index (HDI) scores, the more majority group members anticipated their personal wealth to decline in the future, the more they perceived immigration to pose a realistic threat, compared to majority group members who lived in moderate or high HDI countries with relatively more secure economies. Second, this study obtained evidence that among majority and minority group members, perceived consequences of ethnic diversity and intergroup interactions, such as perceived threats, insecurities, and discrimination, are critical determinants of willingness to support immigrants' equal social standing. The majority group members' perceptions of realistic threat regarding immigration and cultural diversity (i.e., perceived discrimination and economic insecurity) were associated with their unwillingness to confront injustice on behalf of the Russian minorities in Finland, Estonia, and Norway – especially when they perceived the threats to be targeted more at their national ingroups than at themselves personally. Perceived ethnic discrimination also had consequences on the attitudes of Russian speakers towards the national majority groups in these three countries. Namely, perceived discrimination was related to greater hostility towards the national majority groups and less willingness to confront injustice through national disidentification and outgroup mistrust that has resulted from perceived disadvantage. The current study contributes to the existing research in several ways. First, it mostly focuses on contexts that do not have a long history of immigration nor host large groups of immigrants, but have relatively recently become receiving societies with the prospect of an increasing immigration rate (Studies I-III). Second, the study acknowledges both objective contextual factors and individuals' subjective perceptions of diversity, wealth, and social disadvantage that jointly modify the outcomes of intergroup encounters (Studies I and IV). Third, this study looks at intergroup attitudes and behavioral intentions among both majority and minority groups (Studies II and III). Specifically, it demonstrates the detrimental effect of perceived discrimination, which increases intergroup hostility and demobilizes both majority and minority group members from engaging in actions to confront inequality. And lastly, this research distinguishes group-level evaluations of perceived disadvantage and intergroup threats from the personal-level evaluations, and investigates their joint impact on the intergroup outcomes of increasing cultural diversity (Studies I, III, and IV). The role of future anticipations, a relatively understudied perspective in intergroup relations research, is also acknowledged in this research (Study IV). Altogether, the results of this study show the importance of recognizing the reciprocal processes that underlie intergroup relations in societies that are on the verge of becoming highly diverse. By obtaining evidence on how contextual and psychological factors can jointly determine the intergroup consequences of increasing cultural diversity, this study highlights the need to incorporate macro-level factors into mainstream social psychological research, which often focuses on the individual-level analysis of intergroup relations.
  • Satokangas, Henri (Helsingin yliopisto, 2021)
    Explanation of terms in nonfiction The thesis examines the explanation of terms in nonfiction, that is, how the meaning of field-specific expressions is constructed with an orientation towards a non-specialist audience. The thesis consists of four sub-studies, and the data comprises mainly popularizations by academic scholars but also includes school textbooks. The sub-studies view the explanation of terms from four different perspectives: 1) the structure of explanation sequences, 2) narratives as tools for illustration, 3) multimodal explanation sequences and 4) social and historical contextualization. The data is approached within the framework of dialogistically oriented discourse analysis and by means of qualitative text analysis. In the first three sub-studies, the organization of term explanations is analyzed with the concept of discourse patterns. Based on the analysis, discourse patterns are categorized into three types according to the strategy of unfolding explanation. In elaboration patterns the writer proceeds from the term to the explanation; in naming patterns the direction is from the description of the term’s referent to the term itself; and in classifying patterns the writer represents a taxonomy as the frame in which terms are introduced. The second sub-study examines how narratives are utilized as means of illustration and how they function as parts of broader term-explanation patterns. The third sub-study analyzes multimodal discourse patterns as the co-ordination of image and written language. In the fourth sub-study, the discursive construction of the socio-historical context for terms is examined from the perspective of critical language awareness. Categories of context construction uncovered in the analysis include, firstly, connecting the term to an individual person or a named community: naming a person, positioning the term inside a discipline, delineating a group of people defined by a specific viewpoint, stance or opinion, or linking the term to a colloquial label. Secondly, the term can be anchored in time and space through demarcating a geographical or cultural context or describing a temporal trajectory for the term and its development. The term’s social context can, however, also be dismissed altogether. On the other hand, a social context can be implicitly visible in the text, via, among other linguistic resources, passive verb forms. The specialization of semiotic resources in the explanation of terms becomes evident. Story genres and naturally oriented image types, for example, have specialized to perform an illustrative and exemplifying function, whereas description and explanation genres as well as scientifically and technologically oriented image types function as resources for constructing conceptual knowledge as parts of term-explanation patterns. The thesis exhibits a discourse analytic approach to examining and theorizing the use of terminology. A text analysis of terms makes visible the strategies with which the meanings of terms are constructed in nonfiction texts and thus produces linguistic insight about the popularized representation of specialized knowledge.
  • Snirvi, Jasmi (Helsingin yliopisto, 2021)
    Since wound healing represents an increasing burden and a clinical challenge for healthcare professionals, there is a continuous need for more efficient therapy methods. Although current methods include wound dressings, skin grafts, and more recently skin substitutes that often offer a working therapy method, lack of transplantation material and time-consuming preparation time have limited their use. The aim of this thesis was to study the interactions between human adipose-derived mesenchymal stromal cells (hASCs) and wood-derived nanofibrillar cellulose (NFC) dressings and further to evaluate the function of NFC dressing as a cell delivery vehicle in vitro and a wound-healing process in vivo. Additionally, the biocompatibility of two different NFC hydrogel types was evaluated in vitro and their effects on wound healing in vivo. First, the interactions between hASCs and three differentially manufactured NFC dressings (Types 1, 3, and 4) were studied by analyzing cell characteristics. Furthermore, the immunological properties of hASCs were evaluated using enzymelinked immunosorbent and monocyte migration assays. Second, Type 4 NFC dressing was studied clinically by comparing it with a commercial wound dressing when treating skin graft donor site wounds. Third, the biocompatibility of native and anionic NFC hydrogels was evaluated in vitro by analyzing hASCs’ viability and morphology and in vivo using a splinted full-thickness mouse wound healing model. Three different NFC dressings demonstrated different effects on hASCs. Based on our studies, Type 3 NFC dressing offered a functional cell culture platform and potential cell delivery vehicle for hASCs in a cell density-dependent manner without affecting hASCs’ gene or protein expression levels. Further studies suggested that immunological properties of hASCs were also maintained regarding secretion of vascular endothelial growth factor and transforming growth factor-b1 as well as their positive effect on monocyte migration capacity when cultured on top of Type 3 NFC dressing. In contrast, Type 1 and Type 4 NFC dressing did not support the cell culture of hASCs. However, when studied clinically, Type 4 NFC dressing performed comparably with a commercial wound dressing. No statistically significant differences were observed regarding wound healing time determined as the detachment day of the dressing or in percentage of skin epithelialization. Moreover, no difference was observed between Type 4 NFC dressing and commercial wound dressing or between Type 4 NFC dressing and healthy skin regarding skin elasticity in terms of viscoelasticity, elastic modulus, and transepidermal water loss six months postoperatively. Importantly, Type 4 NFC dressing also presented a trend toward less pain experienced by the patients during treatment. Similar to NFC dressings, both native and anionic NFC hydrogels were biocompatible in vitro and supported hASC spheroid formation and high cell viabilities during three-dimensional cell culturing. In vivo, neither NFC hydrogel showed statistically significant differences in wound healing compared with untreated control wounds, and presented normal inflammatory response, granulation-tissue maturation, wound closure rate, re-epithelialization, and vascularization with no signs of fibrosis. To summarize, NFC dressing offers a functional cell culture platform for hASCs and potential therapy material for patient wound treatment. Similarly, native and anionic NFC hydrogels appear to be promising materials for wound treatment. This thesis describes the potential biomaterial NFC for wound healing treatment that is easy to prepare, modify, and use in different forms. NFC dressing is applicable both as a cell scaffold and for patient wound treatment, while native and anionic NFC hydrogels appear to be promising materials for wound treatment. This thesis further describes cell-biomaterial interactions between hASCs and NFC that show relevance for future cell therapies. These findings offer insight into different wound therapy effects of NFC-based materials as well as their possible effects on hASCs, which could improve (and have already) the design of new wound healing therapy methods.
  • Storvall, Sara (Helsingin yliopisto, 2021)
    Primary hyperparathyroidism (PHPT) is most commonly caused by benign tumors of the parathyroid glands, parathyroid adenomas (PA). Only about 1% of PHPT cases are due to parathyroid carcinoma (PC). In Finland, 2-3 new PC cases are diagnosed every year. The only curative treatment of PC is surgery. The diagnosis of PC is based on one criterium only: invasiveness, either on histopathological examination, or in the presence of metastases. This means that the diagnosis can only be set after surgery. Profusely high levels of calcium and parathyroid hormone (PTH) in the blood are more common in malignant disease than benign, but there is a significant overlap between the tumor groups. Atypical parathyroid adenomas (APA) are tumors that share histological properties with PC but lack the invasive characteristics that define malignancy. This is a largely unknown disease entity, and it is not entirely sure to which extent APAs behave indolently like PA, and to which extent they cause a more aggressive cause of disease. Recurrence occurs in roughly half of all PC cases. In inoperable or disseminated disease, there are only palliative treatment options, as chemotherapy or radiotherapy are not effective. With this thesis, we aimed to improve the diagnostics of parathyroid carcinoma: We wanted to find tissue markers that could help in the diagnostic process, or markers that could give a hint on the prognosis of the patient. We also wanted to investigate potential treatment options in inoperable PC. Our material consists of a nationwide cohort of all PC tumors diagnosed in Finland between the years 2000-2011, and additional PC tumors from the Helsinki University Hospital district diagnosed after that as well as APA and PA tumors, also from the Helsinki University Hospital District. Apart from tumor material, we also have extensive clinical data on these patients. We investigated the expression of somatostatin receptors (SST) 1-5 in parathyroid tumors by immunohistochemistry and found that all SST:s 1-5 were expressed to varying degree, and that the expression was increased in malignant tumors compared to benign. Furthermore, we investigated expression of the calcium-sensing receptor (CaSR) as well as the calcium-sensing receptor regulating protein Filamin A (FLNA). All tumors showed positive expression of CaSR. High expression of FLNA was significantly more common in PC tumors compared to benign. We combined the FLNA results with expression of the established parathyroid tumor marker parafibromin and found that tumors with low expression of FLNA that are parafibromin positive are highly likely to be benign or have an indolent course of disease. We also describe a patient case; the patient had disseminated PC with metastases in her bones and mediastinum. She was treated with a combination of the chemotherapy drug temozolomide (TEM), as well as surgery, radiotherapy, and calcium-lowering agents. The patient’s symptoms improved, and her disease progress halted. Ten years later, she is still in remission. For this case study, we investigated methylation of the MGMT gene in the patient’s tumor DNA: inactivation of MGMT by methylation is a prerequisite for successful treatment with TEM. We found that this patient’s methylation levels were high, which supports the role of TEM in the excellent treatment result. We then investigated MGMT methylation in a larger cohort of PC and APA patients. The case report patient was the only one with high MGMT methylation status, which indicates that MGMT methylation is rare in PC.
  • Oliveira Martins, João Pedro (Helsingin yliopisto, 2021)
    Managing a chronic disease throughout the lifetime is an enormous challenge, and diabetes mellitus (DM) is no exception. For the time being, DM cannot be cured but only managed, if and when appropriately diagnosed. However, the vast majority of the currently available glucose-lowering interventions provide diabetic patients with only invasive and distressing treatment opportunities, and an undeniably poor compliance to the therapy. Therefore, it is of utmost importance to develop alternative treatment modalities. Success in the oral delivery of anti-diabetic proteins and peptides could represent a paradigm shift in the management of DM and improve the quality of life of millions of patients. Over the years, nanoparticles (NPs) emerged with great potential to carry and deliver drugs in a precisely tuned and controllable manner. Moreover, nanocarrier systems aimed at oral drug delivery have demonstrated the ability to overcome a variety of biochemical, physical, and mechanical barriers that are intrinsic to the physiological functioning of gastrointestinal tract (GIT). Yet, drug absorption at the intestinal microenvironment is still severely mitigated due to the presence of the tightly organized and restrictive cellular barrier of the intestinal epithelium. The exploration of the mechanisms by which molecules presented in the lumen can be naturally transported into the blood circulation has recently drawn attention to the neonatal Fc receptor (FcRn). FcRn is a membrane receptor responsible for maintaining the homeostasis of immunoglobulin G (IgG) and albumin, via mechanisms of recycling and transcytosis. Hence, NPs functionalized with FcRn-targeting ligands could hijack the FcRn transport pathway to promote the transport of drugs across the intestinal cell wall. Therefore, the main aim of this dissertation was to design and develop different FcRn-targeted NPs for efficient oral anti-diabetic drug delivery. Porous silicon (PSi) NPs were used as core drug carriers due to the unprecedented advantages shown as drug nanocarriers. Recombinant human insulin and GLP-1 were loaded into the PSi NPs as model anti-diabetic peptides. Drug-loaded PSi NPs were further encapsulated into different pH-responsive polymers (hypromellose acetate succinate and lignin), via different preparation techniques (emulsification-evaporation, microfluidics and desolvation). The surface of the NPs was functionalized with either albumin or the Fc portion of IgG, to investigate the underexplored potential of the FcRn in increasing the intestinal absorption of orally administered drug-loaded NPs. Overall, the developed NP formulations showed a small size and narrow size distribution. Microscopy images showed the successful encapsulation of the NPs into the pH-sensitive polymeric matrices. Moreover, these pH-responsive matrices remained intact in acidic conditions, dissolving only at specific pH conditions, thereby enabling the controlled release of the drugs. When functionalized with FcRn-targeting ligands, the NPs presented high cytocompatibility and increased levels of interaction with intestinal cells, in which the FcRn expression was also confirmed. Also importantly, the targeted NPs showed augmented drug permeability across in vitro intestinal models. Hence, this dissertation provides new insights on the design and development of FcRn-targeted NPs, which emerge as a toolbox to explore the potential of the FcRn transcytotic capacity for oral anti-diabetic drug delivery.
  • Rantanen, Kari (Helsingin yliopisto, 2021)
    Graphical models are a versatile machine learning framework enabling efficient and intuitive representations of probability distributions. They can be used for performing various data analysis tasks that would not be feasible otherwise. This is made possible by constructing a graph structure which encodes the underlying dependence structure of the probability distribution. To that end, the field of structure learning develops specialized algorithms which can learn a structure that describes given data well. This thesis presents advances in structure learning for three different graphical model classes: chordal Markov networks, maximal ancestral graphs, and causal graphs with cycles and latent confounders. Learning structures for these model classes has turned out to be a very challenging task, with the few existing exact methods scaling to much fewer number of random variables compared to more extensively developed methods for Bayesian networks. Chordal Markov networks are a central class of undirected graphical models. Being equivalent to so-called decomposable models, they are essentially a special case of Bayesian networks. This thesis presents an exact branch-and-bound algorithm and an in-exact stochastic local search for learning chordal Markov network structures. Empirically we show that the branch and bound is at times able to learn provably optimal structures with higher number of variables than competing methods, whereas the local search scales considerably further. Maximal ancestral graphs are a generalization of Bayesian networks which allow for representing the influence of unobserved variables. This thesis introduces the first exact search algorithm for learning maximal ancestral graphs and a framework for generating and pruning local scores for the search. Empirically we show that the proposed exact search is able to learn higher-quality structures than existing in-exact methods. Finally, we introduce an exact branch-and-bound algorithm for learning causal graphs in the presence of cycles and latent confounders. Our empirical results show that the presented algorithm is able to learn optimal structures considerably faster than a recent exact method for the problem. We also extend our branch and bound to support interventional data and σ-separation, and show empirically that the algorithm can handle higher number of experimental datasets than the only competing method supporting σ-separation.
  • Saressalo, Anton (Helsingin yliopisto, 2021)
    Electric discharge is present in various aspects of our everyday lives. Internal combustion engines rely on spark plugs for the running of the motor, fluorescent lighting functions by gas discharge and a lightning bolt strikes somewhere on earth every second. An electrical breakdown is an event where a voltage across two conductive electrodes, separated by an electrically insulating medium, becomes high enough for the insulating properties of the medium to be weakened, subsequently allowing an electric current to pass through the medium. A special type of such an event is a vacuum arc breakdown, where the electrodes are separated by a gap of void, which acts as a good insulator, but will still be breached under sufficiently high voltage. When controlled, the electric arcing can be used as a powerful tool to focus energy to a specific location. However, several applications are also hindered by the occurrence of breakdowns, including particle accelerators, vacuum interrupters and solar panels. A common factor in these applications is the aim to maximize the electric field strength to optimize the operational efficiency and ecological footprint of such a device. The breakdown phenomenon is at the crossroads of many fields of science, including plasma, materials and surface physics. Effort to explain the breakdown origin has been ongoing for more than a hundred years, and, despite of the constant progress, there are only hypotheses on the exact nature of the process. This work presents an experimental approach for studying the breakdown phenomenon between Cu electrodes, separated by a vacuum gap. The breakdowns are generated as a consequence of repeatedly applying high-voltage pulses across the gap. As a result, statistics, such as breakdown frequency, of the events are investigated and any effects on the surface analyzed. It was shown that cleaning the electrode surface, either by the electric pulsing or plasma treatment, improves the breakdown resistance of the system, whereas any idle time between the high-voltage pulses increases the breakdown probability. Furthermore, it was found that the breakdown events can be attributed to distinct classes, suggesting separate processes responsible for the breakdown generation. One set of processes were labeled extrinsic, as they are driven by the external factors responsible of the surface contamination of the electrode surface. The other processes were characterized as intrinsic, as they were defined by inherent material properties and continued affecting the breakdown frequency even when the effect of extrinsic processes was minimized by plasma cleaning of the surface. Understanding the formation mechanisms of a vacuum arc breakdown allows designing applications that can sustain higher electric fields without breakdown events. The results of this work provide insight on how improving the surface state of an electrode can increase its breakdown resistance. Additionally, an algorithm is presented for recovering the pulsing voltage after a previous breakdown to a high level in an optimal way, with a minimal probability of follow-up breakdowns.

View more