Skip to main content
Login | Suomeksi | På svenska | In English

Browsing by Title

Sort by: Order: Results:

  • Saarentaus, Elmo (2017)
    Genetiska varianter har under senaste åren upptäckts påverka en myriad vanliga samt sällsynta sjukdomar. En viktig grupp varianter i neuropsykiatriska sjukdomar är kromosomavvikelser, speciellt förändringar i antalet genkopior (eng. ”copy number variations”, CNVs). CNVs är associerade med ett brett spektrum olika syndrom, som 22q11.2-deletionssyndrom, samt utvecklingsbetingade neuropsykiatriska sjukdomstillstånd och störningar, som intellektuell funktionsnedsättning, autismspektrets störningar, epilepsi, samt skitsofreni och andra psykotiska sjukdomar. För att undersöka CNVs associeration med neurologiska utvecklingssjukdomar analyserade vi CNVs i två olika kohorter: en kohort med utvecklingsstörda patienter som är värvda i Norra Finland (NFID); och FINRISK-populationskohorten. I en assocationsanalys efter att ha tagit kön, härstamning och genetisk varians (PCA) i beaktande, resulterande i 433 patienter och 1100 kontroller. Det upptäcktes CNVs som beaktades som högst sannolikt patogena i ca 5.3 % av patienter Den vanligaste typen av CNV var 22q11.2-deletionssyndrom (8 st); den vanligaste patogena kriterien var storlek (deletion på över 2 Mb). Associationsanalysens resultat visade en hög association med tidigare igenkända regioner och gener, med deletioner av storleken över 1 Mb fortfarande identifierade i populationen.
  • Ahvonen, Onni (2022)
    This thesis is concerned with the intellectual formation of two radical political thinkers in the Black radical and anti-colonial traditions: Walter Rodney and Cedric J. Robinson. Deploying the “Horne biographical method”, it contextualizes the knowledge-formation of Rodney and Robinson vis a vis the broader social, cultural, political, and intellectual environments in which they were situated. Particularly central are the national liberation-, Black power-, and post-colonial movements in the U.S., Jamaica, Tanzania, and Guyana during the 1960s and the 1970s. Furthermore, this thesis discusses the epistemological foundation of their knowledge-praxis through Enrique Dussel’s notion of the “geopolitics of knowledge” and Stuart Hall’s conceptualization of “the politics of location”, and contends that knowledge is to be understood in relational and positional terms. In analyzing the knowledge production of Rodney and Robinson, this thesis focuses on three central themes that emerge in their respective writings: (1) the role of the radical political intellectual, (2) the utility of Marxism to Black, Indigenous, and Third World peoples, and (3) the theorization of the emergence and development of the racial-colonial world-system. Finally, this thesis argues that in combining their deconstructive critiques with a constructive politics of worldmaking, Rodney and Robinson provide an important model for “guerrilla intellectualism”, where radical knowledge production is seen to emerge “from below” and conceived of as inseparable from material praxis and liberatory struggles.
  • Heikkinen, Merja (2012)
    Tutkielmassa tarkastellaan kvalitatiivisesti Markus Kajon pakinoiden geneerisiä persoonailmauksia. Kajon pakinat ovat yleensä sivun mittaisia tekstejä, joissa käsitellään usein arkipäiväisiä ilmiöitä sekä pienistä suuriin mittasuhteisiin kasvavia mitä jos? -ajatuksesta liikkeelle lähteviä skenaarioita. Koska monet pakinoiden kuvaamat ilmiöt esitetään yleisinhimillisinä eli sellaisina, että ne voivat sattua kenelle tahansa, tämä ilmenee myös persoonapronominien tasolla. Aineiston perusteella tutkimuksen tarkastelun kohteeksi ovat valikoituneet nollapersoona, geneerinen yksikön ja monikon ensimmäinen ja toinen persoona sekä substantiivi-ilmaukset kuten ihminen ja henkilö. Tutkimuksessa sivutaan myös niitä keinoja, joilla pakinoiden toimijoita kategorisoidaan ryhmiksi. Aineiston kielellisiä valintoja tarkastellaan kognitiivisen kielentutkimuksen periaatteiden kautta. Havaintoja vertaillaan sekä suomenkieliseen että kansainväliseen geneeristen persoonailmausten teoriaan. Aineiston analyysissä hyödynnetään myös labovilaista kertomusskeemaa, jota vasten pronominivalintoja ja niiden mahdollista vaihtelua tutkitaan silloin, kun pakinan rakenne sopii siihen. Tutkimus on tehty analysoimalla Kajon pakinoiden geneerisiä persoonailmauksia tekstikontekstissaan. Vaikka geneerisiä persoonailmauksia voi pitää periaatteessa samaviitteisinä ne kaikki voisi yleensä korvata tekstikontekstissa parafraasilla kuka tahansa niillä voi tulosten perusteella havaita olevan retorisia eroja. Niiden avulla voidaan esimerkiksi kategorisoida toimijoita sisä- ja ulkoryhmiin tai tuoda ilmi sitä, miten toisinaan affektinen kertojanääni suhteuttaa yleisön itseensä. Myös nollapersoonan rakenteelliset rajoitukset saattavat motivoida muiden geneeristen muotojen käyttöä. Tämän lisäksi pakinoiden tarkastelu kertomusskeeman kautta osoittaa, että kertomusrakenteella näyttää olevan vaikutusta erilaisten geneeristen muotojen variaatioon. Esimerkiksi nollapersoona esiintyy usein johdattelevissa ja arvioivissa osuuksissa, kun taas geneeristen persoonapronominien kuten yksikön toisen persoonan avulla voidaan ikään kuin zoomata lähemmäs toimintaan.
  • Koivunen, Tiia (2016)
    Objectives. The objective of this study was to find out from Helsinki university's master graduates about their experiences on generic skills, approaches to learning and the similarities of those two. The study sought to answer three questions: 1. What kinds of generic skills students feel they learned during their university studies? 2. What kinds of approaches to learning can be found from the graduates' studying descriptions? 3. What kinds of similarities are there in generic skills and approaches to learning? Similarities have only been studied in quantitative researches so there is a need for qualitative analysis. Methods. The data is a part of a dissertation research of Tarja Tuononen (University of Helsinki) and it was collected in the spring of 2013. The data consist of 10 interviews from students who have already graduated. All the interviewees worked while studying and most of them studied in generalistic fields. First two research questions were conducted with theory-bound content analyses. Top categories were created based on theory because full detachment from previous knowledge would have been difficult. The third research question was conducted on the basis of the results from two previous research questions. Results and conclusions. From the students' descriptions of generic skills, six top categories were separated. Some of those skills were divided into subcategories. Top category skills were knowledge processing, reflection, organization skills, thinking skills, argumentation and social skills. Some of the students felt like they had not learnt any generic skills when asked directly but as the interview progressed, some generic skills appeared. In the second face of the research, students' descriptions of approaches to learning were studied. Three top categories were created based on previous theory: deep approach, surface approach and organized studying. Especially deep approach and organized studying were visible in the interviews and descriptions of surface approach appeared in a few interviews. When examining the similarities between generic skills and approaches to learning especially deep approach and organized studying included similarities. Surface approach had little similarities with generic skills. It can be said based on this research that for example organization skill is nearly identical with organized studying, when examining students' descriptions on both dimensions. Additionally, the descriptions of thinking skills and the descriptions of deep approach included many similarities. Social skills and argumentation from the generic skills did not contain similarities with approaches to learning in this study. It can be said on the basis of the analyses that generic skills and approaches to learning have some similarities but this topic needs to be further researched qualitatively underlining individual profiles.
  • Luuri, Tytti (2020)
    Objective of the study. Previous studies have shown that generic skills required in working life are learned during university studies. However, according to studies, some university graduates may lack generic skills or do not recognise them. This study aims to examine what generic skills graduates from generalist fields have learned during their university studies and whether work experience acquired during studies and participating in practical training correlate with generalist skills learned during university studies. Additionally, it was examined what kind of content graduates would have hoped for during their studies to develop their generic skills and the relationship between these hopes and generic skills learned during university studies. Methods. The data for the study were collected part of a broader study of working life skills (Tuononen, 2019). A total of 1023 University of Helsinki graduates responded to the survey. Respondents were master’s degree graduates from the university’s faculties of arts, social sciences and behavioural sciences (n=372) and their responses were analysed with quantitative methods. Answers to open-ended questions (n=127) were analysed with the abductive content analysis method. Results and conclusions. The results of the study show that graduates had, on average, learned the generic skills measured in the study well or quite well. The respondents had especially experienced to have learned critical thinking skills, multiple perspective viewing skills and structuring and analysing information skills. Co-operation and interaction skills were assessed to have been learned the least of the studied skills. In this study, practical training or work experience acquired during university studies did not correlate with generic skills learned during university studies. Hopes for content to develop working life readiness were divided into formal and informal learning environments. Formal learning environments included courses and practical training, whilst informal learning environments included working life events, alumni activities, study guidance and job-hunting training. The respondents especially hoped for practical university courses with tangible applications to working life. Respondents were divided into two groups based on their hopes for more content in formal or informal learning environments. There was no statistical difference in the assessment of generic skills learned between these groups. More attention should be given to developing co-operation and interaction skills during generalist studies and finding more efficient ways of teaching them. More research is needed to show how studies should be developed to better support the relevance of generalist studies to working life.
  • Horn, Janina (2020)
    Tutkielman tavoitteena on esittää, että lukion todennäköisyyslaskennan käsitteet ja teoriat voidaan opettaa ja opiskella oppikirjojen perinteisistä esimerkeistä poikkeavassa kontekstissa. Kontekstiksi valitaan todennäköisyyslaskennan mallien todellinen sovellusalue. Samalla luodaan tehtäväpaketin muodossa ainerajat ylittävä opetusprojekti lukioon. Tutkielmassa perustellaan, miksi evoluutio ja populaatiogenetiikka ovat hyvä asiayhteys lukion todennäköisyyslaskennen opetuksessa ja toisin päin. Opetusprojektin toteutus noudattaa soveltaen tutkivan oppimisen menetelmää. Oppiainekohtaisten tavoitteiden saavuttamisen lisäksi yksi opetusprojektin päätavoitteista on oppilaiden ajattelutaitojen kehittäminen, sillä opiskeltavien asioiden syvällinen ymmärtäminen tapahtuu ajattelun kautta. Tutkielmassa on rakennettu tehtäväpaketti, joka koostuu 15 tehtävästä. Tehtävät liittyvät todennäköisyyslaskentaan, evoluutioon ja populaatiogenetiikkaa. Tehtävät alkavat yksinkertaisista käsitteiden määrittelytehtävistä ja peruslaskutehtävistä muuttuen vähitellen tiedon soveltamista vaativiksi ongelmanratkaisutehtäviksi. Opiskelijalla on aktiivinen rooli tiedonhakijana, ja opiskelijoiden keskinäiset keskusteltu ja yhteistyö ovat keskeisessä roolissa prosessissa, jonka tavoite on uuden tiedon rakentaminen.
  • Oristo, Sini (2019)
    Objectives Autism generally refers to lingual, communicational and behavioral continuous traits, which are determined by genetics as well as environment. The purpose of this study was to examine to which extent the genetic autism risk explains the variation in autistic traits in two-year-old children in the normal population. Additionally the impact of the genetic autism risk was examined in conditions, where the mother of the child suffered from depressive symptoms during pregnancy and/or had alexithymia (trouble identifying feelings). Methods The sample was collected from the Child-Sleep cohort study (n=942). The child’s genetic autism risk was calculated by comparing the child’s genome with the genome of people with an autism diagnosis. All of the other information was collected with questionnaires. The child’s autistic traits were assessed according to the autism scale in BITSEA. The mother’s depressive symptoms during pregnancy were assessed according to the Ces-d (short) questionnaire and her trouble identifying feelings according to the TAS-20-questionnaire. The connections between the variables were examined with Pearson’s correlation coefficient. The autistic traits were examined with linear regression analysis. Results and conclusions The genetic autism risk explained 2% of the variation in the autistic traits of two-year-old children in the normal population. This result is in line with previous studies. Gender (autistic traits were more prevalent in males), maternal adolescence and the degree of maternal alexithymia were the best predictors of an autistic phenotype. Generally maternal maturity is considered a risk factor for the child’s more autistic phenotype, whereas our result possibly points out that the risk is rather created by paternal maturity instead. In addition of identifying the greatest risk factors, interactions between the genetic risk of autism and the psychological factors of the mother (depressive symptoms during pregnancy and trouble identifying emotions) were examined. However, these interactions were not statistically significant in this study. The impact of genes is already a verified fact – the next goal is to identify environments, where the genes interact, creating a more autistic phenotype.
  • Sánez Tähtisalo, Heini (2017)
    Sydän- ja verisuonisairauksien syntyyn vaikuttavat sekä ympäristötekijät että perimä. EKG kuvaa sydämen sähköistä toimintaa. Mitattaviin osiin jaetun EKG:n rinnalla sydänperäiseen äkkikuolleisuuteen vaikuttavien geenitekijöiden tunnistaminen mahdollistaisi riskipotilaiden tunnistamisen ennen vakavia sydänperäisiä häiriöitä sekä ennaltaehkäisevän lääketieteellisen hoidon kohdentamisen. Tässä kandidaattipolymorfismitutkimuksessa selvitettiin onko sydämen toimintaan osallistuvien beeta1 – ja beeta2 –adrenoreseptorien (ADRB1 ja ADRB2) tai angiotensiinikonvertaasientsyymin (ACE) geenien keskeisillä polymorfismeilla vaikutusta EKG:n piirteisiin. Potilasaineistona oli GENRES-tutkimukseen osallistuneet verenpainetautia sairastavat miehet (n=186). Polymorfismien vaikutusta QT-aikaan ja T-aallon muodostumiseen etsittiin nelivaiheisella lineaarisella regressiomallilla. Analyysien keskeisimpinä tuloksina havaittiin tilastollisesti merkitsevä ADRB1 Gly49 –alleelin QT-aikaa lyhentävä vaikutus ja ADRB1 Gly389 –alleelin QT-aikaa pidentävä vaikutus. ADRB2 Gly16Arg- ja Gln27Glu-vaihteluiden yhteys T-aallon muodostumiseen jäi olemattomaksi tai heikoksi. ACE I/D:n deleetio-polymorfismi viittaa tutkimuksen perusteella T-aallon muutoksiin, jotka ovat yhdistettävissä lisääntyneeseen äkilliseen sydänperäiseen kuolleisuuteen. Tämän tutkielman havainnot viittaavat siis useampiin mahdollisiin polymorfismien ja EKG-muuttujien välisiin yhteyksiin ja kannustavat tarkempiin jatkotutkimuksiin yhteyden kliinisen merkityksen arvioimiseksi.
  • Muniandy, Maheswary (2014)
    The study of obesity has drawn wide-spread interest because of its far-reaching consequences. Obesity is on the rise and has been linked to several clinical complications such as type 2 diabetes and hypertension. Obesity is defined as a condition in which body mass index (BMI) is greater than 3m2/kg. BMI itself is highly heritable with the rate of heritability in twin and adoption studies ranging from 45%–85%. The study was designed to analyze the differences in gene expression in MZ twin pairs discordant for BMI. A total of 26 twin pairs were selected based on a within-pair BMI difference of more than 3 kg/m2 when compared to his or her twin. The twins for this study belonged to either FinnTwin16 (birth cohort 1975-1979) or FinnTwin12 (birth cohort 1983-1987). Samples were extracted from the participants and used in microarray experiments. The resulting data was processed using various packages of the Bioconductor software. The quality control process identified one sample as faulty and as a result the sample as well as the sample of the twin were discarded. This resulted in a sample size of 50 twins. Differential analysis carried out using the limma package of Bioconductor revealed 980 genes that were differentially expressed. These genes were then processed further in the Ingenuity Pathway Analysis (IPA) tool as well as the BiNGO tool. This revealed the molecular networks, gene pathways and gene annotations that were pertinent to the genes uploaded. IPA also provided a list of functions and diseases these genes were involved in. As the final step, the prevalent themes across these results were summarized. This analysis provided many leads that should be investigated further in future studies. It is suggested that future studies start with precise, concrete biological questions that narrow down the scope of biological analysis. This is in view of the extensive amount of data available via microarray studies and the myriad of hypothesizes that can be investigated.
  • Liukkonen, Titta (2024)
    Atlantic salmon (Salmo salar) have a complex life history, characterized by a juvenile stage spent in freshwater before migrating to the sea for one to several years and then returning to their natal river to spawn. Sea-age at maturity, which corresponds to the number of years spent at sea, is an important life-history trait influenced by environmental and genetic factors. Variation in sea-age at maturity in Atlantic salmon has been associated with a SNP located in the intron of six6 on chromosome 9 (six6a). This gene is also known to be evolutionary conserved in eye development in vertebrates and has a crucial role in sexual maturation in mammals. In 3-year-old Atlantic salmon, six6a was found to be expressed in the eyes, brain, gills, and testes. In addition, in situ hybridization has shown that six6a is expressed in multiple organ systems in embryos, for example, the eye and brain developing areas. However, quantification of its level of expression has only been carried out on whole individuals so far. Therefore, this thesis aims to quantify the expression level of six6a, and its paralogue, six6b, at different time points (1-9 months post-hatching) and in four different tissues and a body part (brain, eye, gill, hypothalamus, and head) to identify the tissues and developmental stages that are relevant for future functional studies on the association between six6a genotype and variation in sea-age at maturity. Using a 3’ RNA sequencing approach we found that the eyes and hypothalamus had the highest expression level of both paralogues with a higher expression level of six6b. The expression of the two paralogues in gill samples was very low or non-existent and the low level of expression detected in head and whole brain samples could be influenced by the presence of the eyes and the hypothalamus, respectively. Based on these results, the hypothalamus and eyes would be important tissues for further investigation to gain more information on the association between the six6a genotype and variation in sea-age at maturity in Atlantic salmon. In addition, a more in-depth analysis of the transcriptomic data obtained would allow the investigation of the specific roles of each paralogue, for example by studying correlated genes in the different tissues.
  • Rantanen, Frida (2024)
    Essential thrombocythemia (ET) is a clonal hematopoietic disease characterized by an abnormal increase of platelets in the circulation, with increased risk of thrombosis and hemorrhage. Despite megakaryocytes having a central role in the disease, few studies have investigated their gene expression in ET. The aim of this study is to characterize the gene expression profiles of megakaryocytes from ET patients harboring different driver mutations, and increase the knowledge of the molecular mechanisms underlying the pathophysiology of the disease. In this study, samples were obtained from healthy donors and ET patients with JAK2 V617F, CALR Type I, CALR Type II driver mutations and triple-negative patients. Following megakaryocyte culture from peripheral blood and RNA sequencing, the data was pre-processed and analyzed using differential gene expression analysis. The downstream analysis was conducted using pathway enrichment analysis tools. The analysis revealed that all mutants shared common deregulated genes related to processes involving platelets and coagulation. However, it was shown that CALR and JAK2 V617F mutants also have distinct patterns of gene expression. CALR Type I mutants had a unique gene expression signature consisting of genes related to immune response, as well as metabolic, regulatory, proliferative, and inflammatory pathways, while CALR Type II mutants had unique genes related to ribosomes. The CALR mutants also shared a common anti-inflammatory response signature which set them apart from JAK2 V617F mutants. In conclusion, this study shows that the gene expression profiles of ET mutants are heterogeneous. Moreover, the results provide new insights into the gene expression profiles of CALR mutants that distinguish them from the other mutants. Further experiments using single-cell RNA sequencing methods could build upon these findings and uncover the observed gene expression discrepancies between CALR and JAK2 mutants with increased accuracy.
  • Nakane, Elina (2018)
    Objectives. Maternal diabetes during the pregnancy increases the risk of pregnancy complications, but the effects of maternal diabetes on offspring cognition are less understood. Earlier studies have mainly associated the adverse effects of maternal diabetes with slight deficits in general cognitive and verbal functions in young children. Despite the earlier studies, it is unclear, does maternal diabetes per se affect cognitive development in children and adult offspring. The offspring with several developmental risks seem to be more prone to the adverse effects of maternal diabetes than offspring without the other concomitant risks. The aim of this study was to examine is maternal diabetes associated to lowered offspring general cognitive function in childhood and midlife, when the other concomitant perinatal risks occurred or either did not occur. A hypothesis was that maternal diabetes is associated to lowered general cognitive function only in children who had the other concomitant perinatal risks. Another aim was to explore is there a time related change in the risk groups. Methods. This study is a part of a prospective birth-cohort study originating in Helsinki region that follows 1971 to 1974 born risk group offspring. Out of 22,359 consecutive deliveries at the Institute of Midwifery during that time, 93 offspring had mother's diabetes obtained during the pregnancy or before it. Of the offspring with maternal diabetes, 59 had maternal diabetes as the only risk, and 34 had at least another predefined concomitant risk. General cognitive function in the subjects and controls was assessed by Wechsler Intelligence Scales at 9 and 40 years as a part of the wider neuropsychological examination. Differences between the groups were examined by group and pairwise comparisons. Longitudinal changes in general cognitive function in each group were estimated by fitting the linear multilevel models. Results and conclusions. Findings of the present study were controversial to the hypothesis. Both risk groups, with and without other concomitant risks, had lower general and verbal function in childhood than controls. At midlife, no effect of maternal diabetes was found. The results indicated that general cognitive function and acquired verbal information improved at least in the risk group with the other concomitant risks. Otherwise performance remained relatively same.
  • Säkkinen, Niko (2020)
    Predicting patient deterioration in an Intensive Care Unit (ICU) effectively is a critical health care task serving patient health and resource allocation. At times, the task may be highly complex for a physician, yet high-stakes and time-critical decisions need to be made based on it. In this work, we investigate the ability of a set of machine learning models to algorithimically predict future occurrence of in hospital death based on Electronic Health Record (EHR) data of ICU-patients. For one, we will assess the generalizability of the models. We do this by evaluating the models on hospitals the data of which has not been considered when training the models. For another, we consider the case in which we have access to some EHR data for the patients treated at a hospital of interest. In this setting, we assess how EHR data from other hospitals can be used in the optimal way to improve the prediction accuracy. This study is important for the deployment and integration of such predictive models in practice, e.g., for real-time algorithmic deterioration prediction for clinical decision support. In order to address these questions, we use the eICU collaborative research database, which is a database containing EHRs of patients treated at a heterogeneous collection of hospitals in the United States. In this work, we use the patient demographics, vital signs and Glasgow coma score as the predictors. We devise and describe three computational experiments to test the generalization in different ways. The used models are the random forest, gradient boosted trees and long short-term memory network. In our first experiment concerning the generalization, we show that, with the chosen limited set of predictors, the models generalize reasonably across hospitals but that only a small data mismatch is observed. Moreover, with this setting, our second experiment shows that the model performance does not significantly improve when increasing the heterogeneity of the training set. Given these observations, our third experiment shows that
  • Viitanen, Esa (2017)
    Työni kuuluu filosofisen logiikan alaan. Se on luonteeltaan kirjallisuustutkimus. Se tarkastelee erilaisten episteemisten ja doksastisten logiikkojen semantiikkaa ja totuuskäsitettä. Tarkasteltavat logiikat kattavat koko filosofisen logiikan modernin historian alkaen 1960-luvun Kripke-malleista ja päätyen 2000-luvun uskomusten korjausteorioihin. Työn tavoitteena on selvittää, millaisia semantiikkoja ja totuusmääritelmiä episteemisille ja doksastisille logiikoille on ehdotettu, mihin luonnollisen tai formaalin kielen ongelmiin ne yrittävät vastata, miten hyvin semantiikat onnistuvat mallintamaan ihmisen tai tekoälyn päättelyä ja miten ne onnistuvat välttämään mahdolliset haasteet kuten loogisen kaikki-tietävyyden ongelman. Tutkimuksen aineistona on käytetty laajasti alan tieteellistä kirjallisuutta, mm. Jaakko Hintikan, Ronald Faginin ja Hans van Ditmarschin kirjoittamia kirjoja, sekä eri tutkijoiden tiedelehdissä julkaistuja artikkeleita. Tutkimus kattaa vanhemmat staattiset episteemiset ja doksastiset logiikat, konditionaalien teoriat sekä uudemmat dynaamiset episteemiset logiikat. Staattiset mahdollisten maailmojen logiikat kuvaavat tarkasti yksinkertaisia tiedon totuusarvoja. Ne ovat kuitenkin alttiita loogisen kaikkitietävyyden ongelmalle. Mahdottomien maailmojen käyttöönotto tai naapurustologiikka auttavat välttämään ongelmaa. Loogisten konditionaalien semantiikat eivät kuvaa oikein luonnollisen kielen konditionaalia eivätkä kausaliteettia, vaan johtavat vakaviin ongelmiin, ns. loogisiin paradokseihin. Sen sijaan David Lewisin ehdottama semantiikka antaa hyväksyttävän kuvauksen hypoteettisille konditionaaleille. Uskottavuus-käsitteeseen perustuvat doksastiset logiikat kuvaavat inhimillistä päättelyä jo varsin tarkasti. Uskomusten ja tiedon päivittymistä uuden tiedon ilmetessä kommunikaation seurauksena kuvataan mm. julkisten ilmoitusten logiikalla. Mutta uudemmat tapahtuma- ja toiminta-mallilogiikat tekevät sen tarkemmin mutkikkaimmissakin tilanteissa, esim. korttipeleissä. Episteemiset logiikat mallintavat tarkasti rationaalisten koneiden kuten tekoälyn ja robottien loogista päättelyä. Mutta erehtyväisen inhimillisen päättelyn mallintaminen episteemisiin ja doksastisiin logiikkoihin on edelleenkin haasteellista ja näyttää väistämättä johtavan ongelmiin.
  • Williams Moreno Sánchez, Bernardo (2022)
    The focus of this work is to efficiently sample from a given target distribution using Monte Carlo Makov Chain (MCMC). This work presents No-U-Turn Sampler Lagrangian Monte Carlo with the Monge metric. It is an efficient MCMC sampler, with adaptive metric, fast computations and with no need to hand-tune the hyperparameters of the algorithm, since the parameters are automatically adapted by extending the No-U-Turn Sampler (NUTS) to Lagrangian Monte Carlo (LMC). This work begins by giving an introduction of differential geometry concepts. The Monge metric is then constructed step by step, carefully derived from the theory of differential geometry giving a formulation that is not restricted to LMC, instead, it is applicable to any problem where a Riemannian metric of the target function comes into play. The main idea of the metric is that it naturally encodes the geometric properties given by the manifold constructed from the graph of the function when embedded in higher dimensional Euclidean space. Hamiltonian Monte Carlo (HMC) and LMC are MCMC samplers that work on differential geometry manifolds. We introduce the LMC sampler as an alternative to Hamiltonian Monte Carlo (HMC). HMC assumes that the metric structure of the manifold encoded in the Riemannian metric to stay constant, whereas LMC allows the metric to vary dependent on position, thus, being able to sample from regions of the target distribution which are problematic to HMC. The choice of metric affects the running time of LMC, by including the Monge metric into LMC the algorithm becomes computationally faster. By generalizing the No-U-Turn Sampler to LMC, we build the NUTS-LMC algorithm. The resulting algorithm is able to estimate the hyperparameters automatically. The NUTS algorithm is constructed with a distance based stopping criterion, which can be replaced by another stopping criteria. Additionally, we run LMC-Monge and NUTS-LMC for a series of traditionally challenging target distributions comparing the results with HMC and NUTS-HMC. The main contribution of this work is the extension of NUTS to generalized NUTS, which is applicable to LMC. It is found that LMC with Monge explores regions of target distribution which HMC is unable to. Furthermore, generalized NUTS eliminates the need to choose the hyperparameters. NUTS-LMC makes the sampler ready to use for scientific applications since the only need is to specify a twice differentiable target function, thus, making it user friendly for someone who does not wish to know the theoretical and technical details beneath the sampler.
  • Latvala, Lotta (2024)
    Generatiivinen tekoäly on tehostanut markkinointiyritysten mahdollisuuksia tuottaa innovatiivisia mainosmateriaaleja entistä tehokkaammin asiakasyrityksen toiveiden mukaisesti. Markkinointiyritysten on otettava toiminnassaan huomioon generatiivisen tekoälyyn ja sen tuottamiin teoksiin liittyvät omistus- ja tekijänoikeuskysymykset, sekä niistä johtuvat oikeudelliset riskit. Tutkielman tavoitteena oli selvittää, mitä tekijänoikeuksiin ja sopimuksiin liittyviä riskejä markkinointiyrityksen tulisi ottaa huomioon, ja miten näitä riskejä voisi hallita sopimusten avulla. Generatiivisen tekoälyn kouluttamisessa on voitu käyttää syötedatana teoksia, joiden käyttöön tekoälyn kehittäjällä ei ole oikeutta. Tällöin myös tekoälyn tuotos eli tuotantodata, jonka markkinointiyritys myy eteenpäin, voi loukata kolmannen osapuolen tekijänoikeuksia ja täyttää esimerkiksi tekijänoikeuslain mukaisen tekijänoikeusrikkomuksen tunnusmerkistön. Tekijänoikeusloukkauksen lisäksi markkinointiyrityksen on otettava toiminnassaan huomioon sopimusvelvoitteet, jotka johtuvat joko osto- tai myyntisopimuksista. Vallitsevan oikeustilan mukaan tekoälyllä luotuun teokseen ei ole mahdollista saada tekijänoikeutta, eikä markkinointitoimisto voi näin ollen luvata näitä oikeuksia asiakasyritykselleenkään. Markkinointiyritys voi siis suoran tekijänoikeusloukkauksen lisäksi syyllistyä sopimusrikkomukseen, josta voi aiheutua sekä taloudellisia menetyksiä että mainehaittoja. Markkinointiyrityksen näkökulmasta generatiivisen tekoälyn käytöstä voi aiheutua riskejä eri osista sopimusketjua, ja nämä riskit voivat koskea sekä syöte- että tuotantodataa. Tekoälyn palveluntarjoaja voi pyrkiä rajaamaan vastuutaan tuotetusta palvelusta ostosopimuksella. Ostosopimuksessa voi olla esimerkiksi kerrottu, millaisella syötedatalla tekoälyä on koulutettu, sekä miten palveluntarjoaja rajaa vastuutaan tuotantodatasta. Markkinointiyrityksen kannalta suurimmat riskit seuraavat ostosopimuksista, joilla palveluntarjoaja rajaa vastuutaan mahdollisista rikkomuksista eikä lupaa suojaa tekijänoikeuskanteita vastaan sekä myyntisopimuksista, joilla markkinointiyritys kantaa itse suuren osan tuotokseen ja sen käyttöön liittyvistä riskeistä myydessään tuotoksia asiakkailleen. Esitettyjä tekijänoikeusloukkaukseen ja sopimusrikkomuksiin liittyviä riskejä voidaan kuitenkin hallita erilaisilla sopimusteknisillä keinoilla, jotka voidaan jakaa neljään kategoriaan: riskien tunnistamiseen ja riskinkantokyvyn arviointiin, palveluntarjoajan valintaan, sopimusten hallintaan ja taloudelliseen hallintaan. Yrityksen kannalta tärkeimmät riskienhallinnan keinot ovat oikean palveluntarjoajan valinta sen perusteella, miten ostosopimus suojaa markkinointiyritystä ja oikein laaditut myyntisopimukset, jotka siirtävät tekijänoikeusproblematiikkaa koskevat riskit asiakkaalle tai rajaavat yrityksen kantamaa vastuuta laadullisesti ja määrällisesti.
  • Norrby, Jonathan (2023)
    Tutkielman kohteena on ajankohtainen ja epäselvä tekijänoikeuden ja tekoälyn kouluttamisen ristitulessa syntyvä oikeudellinen kysymys: saako generatiivista tekoälyä kouluttaa tekijänoikeudella suojatulla aineistolla ilman että teosten oikeudenhaltijat ovat antaneet tähän lupaansa? Tutkielmassa selvitetään siten, miltä osin tekijänoikeutta rajoittavia oikeussäännöksiä mahdollisesti voidaan soveltaa tällaiseen tilanteeseen. Tutkimuskysymystä selvitetään pääasiassa Suomen voimassaolevan oikeuden nojalla sekä soveltuvin osin Euroopan unionin oikeuden nojalla, eli tutkielman tutkimusmetodina on lainoppi, minkä lisäksi noudatetaan pääasiallisesti perinteistä oikeuslähdeoppia. Tarkasteltavaan aineistoon kuuluu siten muun muassa tutkimuskysymykseen liittyvät säädökset, säädöksiä valmistelevat asiakirjat kuten hallituksen esitykset sekä tutkimusongelman kannalta merkityksellinen oikeuskäytäntö ja oikeuskirjallisuus. Tutkimuskysymystä tarkastellaan kuitenkin myös valmisteilla olevan EU:n tekoälyasetuksen valossa, jonka osalta tutkimusmetodi poikkeaa puhtaasta lainopista, sillä asetus ei ole voimassaolevaa oikeutta. Tutkielman tulokset paljastavat, että tutkimusongelmaan soveltuvat tietyt tarkastelun kohteena olevat tekijänoikeutta rajoittavat säännökset, joskaan tämä ei tapahdu ongelmitta. Erityisesti tekijänoikeuden tilapäistä kappaleen valmistamista koskeva rajoitussäännös näyttäytyy tutkimustulosten valossa ongelmalliselta. Tekstin- ja tiedonlouhintaa koskeva rajoitussäännös näyttää toisaalta soveltuvan tutkimuskysymyksen mukaiseen asetelmaan melko ongelmattomasti, joskin senkin osalta on hahmotettavissa joitakin epäselvyyksiä. EU:n tekoälyasetusehdotuksen vaikutus tutkimuskysymykseen jää puolestaan melko etäiseksi, vaikkakin tiettyjä sinänsä merkittäviä yhteyksiä asetuksen ja tutkimusongelman välillä ilmenee. Lopuksi tutkielmassa tiivistetään ne tutkimustulokset, joihin tutkielman edeltävissä jaksoissa on tultu. Lisäksi hahmotetaan sitä oikeustilaa, johon tutkimustulokset johtavat. Näkemys on, ettei tämä oikeustila ole hälyttävä, vaan voimassa oleva kotimainen ja EU-normisto vastaa pääosin riittävällä tavalla tutkimusongelman mukaiseen tilanteeseen. Todetaan myös, ettei tekoälyasetusehdotus nykymuodossaan ja voimaantullessaan tuo tutkimuskysymyksen ytimeen liittyen merkittäviä muutoksia.
  • Martins, Beatriz (2020)
    According to the latest estimations, cancer is the second leading cause of death worldwide. Despite the significant advances in the range of drugs and treatment modalities to treat cancer, the number of deaths is estimated to continue rising, posing serious challenges for the patients, their families, and the healthcare systems. Conventional treatments tend to be associated with severe adverse side effects and treatment resistance. Consequently, safer and more efficient therapy options are urgently needed, especially for the treatment of metastatic tumors refractory to conventional treatments. A new and revolutionizing field in oncology is immunotherapy, in which oncolytic viruses are included. Oncolytic viruses have an inherent or acquired selectivity to replicate exclusively in tumor cells, ultimately destroying them. Simultaneously, they also activate the dormant host’s immune system to fight against the tumor. Adenoviruses, particularly, have shown to be safe, inducing only mild adverse side effects in clinical trials, making them a great candidate for further clinical development. Adenoviruses can be genetically modified to increase their infectivity or improve the anti-cancer immune responses induced by the virus, e.g., through the expression of immunostimulatory molecules. The focus of this thesis was to develop and characterize several genetically modified oncolytic adenoviruses expressing either OX40L alone or OX40L and CD40L, two co-stimulatory molecules capable of engaging both the innate and adaptive arms of the immune system to fight the tumor. The insertion of the transgenes into the E3B-14.7k region of the Ad5/3-∆24 adenovector plasmid was performed using Gibson Assembly® cloning approach. After successful cloning, the recombinant viral genomes were transfected into A549 cells for viral amplification, followed by CsCl purification to produce a high titer viral preparation. The expression of the transgenes was studied in vitro by ELISA and functional assays, showing promising expression levels of functional OX40L and CD40L. However, when the infectivity and virus killing potency were analyzed, in vitro by immunocytochemistry and MTS assay; and in vivo using an immunodeficient mouse model, the data showed that the cloned viruses performed sub-optimally when compared to the control unarmed virus (Ad5/3-∆24). These findings suggest that the insertion of the two transgenes in place of the E3-14.7k gene was detrimental to the fitness of the virus.
  • Sjöberg, Madeleine (2018)
    Cancer afflicts an ever-growing number of people globally each year. In part due to a complex pathophysiology where much is still unknown, the need for new cancer treatments has been persistent, fuelled further by the issue of treatment resistance. An emerging field holding much promise in oncology is immunotherapy, a subgroup of which is oncolytic virus treatments. These treatments utilize the inherent or acquired ability of certain viruses to selectively replicate in tumor cells to fight cancer. One of these viruses is the adenovirus. With these viruses it is possible to modulate the immune response e.g. through the expression of certain genes. The thesis focuses on genetically arming an oncolytic adenovirus in an effort to enhance treatment efficacy. The transgene of choice is the CD40 ligand (CD40L), a costimulatory molecule capable of aiding in the development of systemic antitumor immunity. Adenoviruses have previously been designed expressing the CD40L, however, a novel aspect was introduced with the design and incorporation of a soluble a form of the protein. The main aim of the study was the construction of four functional oncolytic adenoviruses, encoded with either the human or mouse variants of the two CD40L proteins (full-length and soluble). Successful completion required protocols for the cloning, bacterial colony screening, and primary virus production to be established. Insertion of the CD40L transgenes into the E3-gp19k region of the chosen Ad5Δ24 backbone was first attempted with the traditional approach of homologous recombination. The method that ultimately proved successful was a one-step Gibson Assembly® reaction. Screening the bacterial colonies with colony polymerase chain reaction, the potential CD40L positive clones underwent restriction analysis to affirm the presence of the transgene in the viral genome, as well as the retainment of critical elements. Two out of three recombined plasmids carrying the full-length CD40L proceeded to transfection and virus propagation in A549 cells, after which the presence of the adenovirus and CD40L expression was confirmed with immunostaining. Finally, a protocol was successfully established by the construction of one of the intended four viruses. The protocol entails all the main steps from cloning until primary virus production, additionally offering the option of applying it to the genetic arming of the Ad5Δ24 with other transgenes of interest. In terms of future perspectives for the project, following construction of the remaining viruses, the intentions are to validate transgene expression and functionality for all constructs, as well as compare the immunogenicity between the full-length and soluble CD40L. In the event of promising results, the project will hopefully proceed to in vivo studies.
  • Sjöberg, Madeleine (2018)
    Cancer afflicts an ever-growing number of people globally each year. In part due to a complex pathophysiology where much is still unknown, the need for new cancer treatments has been persistent, fuelled further by the issue of treatment resistance. An emerging field holding much promise in oncology is immunotherapy, a subgroup of which is oncolytic virus treatments. These treatments utilize the inherent or acquired ability of certain viruses to selectively replicate in tumor cells to fight cancer. One of these viruses is the adenovirus. With these viruses it is possible to modulate the immune response e.g. through the expression of certain genes. The thesis focuses on genetically arming an oncolytic adenovirus in an effort to enhance treatment efficacy. The transgene of choice is the CD40 ligand (CD40L), a costimulatory molecule capable of aiding in the development of systemic antitumor immunity. Adenoviruses have previously been designed expressing the CD40L, however, a novel aspect was introduced with the design and incorporation of a soluble a form of the protein. The main aim of the study was the construction of four functional oncolytic adenoviruses, encoded with either the human or mouse variants of the two CD40L proteins (full-length and soluble). Successful completion required protocols for the cloning, bacterial colony screening, and primary virus production to be established. Insertion of the CD40L transgenes into the E3-gp19k region of the chosen Ad5Δ24 backbone was first attempted with the traditional approach of homologous recombination. The method that ultimately proved successful was a one-step Gibson Assembly® reaction. Screening the bacterial colonies with colony polymerase chain reaction, the potential CD40L positive clones underwent restriction analysis to affirm the presence of the transgene in the viral genome, as well as the retainment of critical elements. Two out of three recombined plasmids carrying the full-length CD40L proceeded to transfection and virus propagation in A549 cells, after which the presence of the adenovirus and CD40L expression was confirmed with immunostaining. Finally, a protocol was successfully established by the construction of one of the intended four viruses. The protocol entails all the main steps from cloning until primary virus production, additionally offering the option of applying it to the genetic arming of the Ad5Δ24 with other transgenes of interest. In terms of future perspectives for the project, following construction of the remaining viruses, the intentions are to validate transgene expression and functionality for all constructs, as well as compare the immunogenicity between the full-length and soluble CD40L. In the event of promising results, the project will hopefully proceed to in vivo studies.