Browsing by Title

Four different Bradyrhizobium sp. (lupin) inoculants were investigated in both greenhouse and field experiments to compare their effects on growth, yield and biological nitrogen fixation. Narrow-leafed and white lupin of the greenhouse experiment compared the strains test in different potting media in a controlled environment, while field experiment tested their performance in field conditions. The greenhouse experiment was conducted with 3 narrow-leafed lupin cultivars (Haags Blaue, Boruta and Sonet) and 1 white lupin cultivar (Ludic). Plants were grown in 3 different potting media (soil, 2 peat : 1 sand and 1 peat : 2 sand) with 5 Bradyrhizobium treatments (uninoculated control, commercial peat inoculant of HAMBI 3118 and liquid cultures of HAMBI 3115, HAMBI 3116 and HAMBI 3118). Plants were grown in a greenhouse unit with average day and night temperature of 22°C and 18°C. Plants were illuminated by using cool white fluorescent tubes maintaining 18 hours day and 6 hours night. In the greenhouse experiment, inoculation significantly increased shoot (117.1-141.9%), root (45.8-64.4%) and nodule (237.0-266.6%) dry weight, plant height (38.3-46%), nodule number (620-659%) and chlorophyll content (29.0-38.5%) over the values found in uninoculated controls. Soil type or potting medium also influenced lupin growth and yield, with better results observed in soil, poorer in 2 peat : 1 sand and poorest in 1 peat : 2 sand. Best performaces were obtaind by inoculating with HAMBI 3115 strain in soil. Uninoculated plants and even inoculated plants grown in peat-sand potting medium, showed relatively poor results, which was more obvious in high-yielding cultivars, Boruta and Ludic, than in low-yielding cultivars, Haags Blaue and Sonet. Inoculation treatments also showed significantly higher shoot (3.15-3.39% N) and root (1.96-2.54% N) nitrogen content. Biological nitrogen fixation rate, measured by the nitrogen difference method, ranged between 87.9 and 90.8% depending on both bacterial strain and host cultivar. The field experiment showed significant increases in shoot (14.4-47.9%), root (11.9-29.1%) and seed (13.8-68.6%) dry weight, plant height (3.6-10.7%), pod plant-1 (10.7-50.6%) and chlorophyll content (5.7-20.7%) following inoculation of the three narrow-leafed lupin cultivars. Uninoculated plants grown in soil in the greenhouse experiment and in the field experiment both produced some nodules, which showed the evidence of presence of indigenous nodule-forming and nitrogen-fixing bacteria. Among the 3 liquid cultures, HAMBI 3115 performed best in terms of lupin growth, yield and biological nitrogen fixation in both greenhouse and field experiments. The performance of the peat-based commercial inoculant of HAMBI 3118 strain exceeded all other inoculants in the field experiment but not in the greenhouse experiment, showing the importance of the carrier. The results indicated that lupin growth and yield are strongly affected by Bradyrhizobium inoculation and soil characteristics. Selection of a suitable Bradyrhizobium strain for inoculation and growing cultivars according to their soil preferences can maximize lupin yield. The suitability of HAMI 3115 for making peatbased inoculants should be tested.

Pichia pastoris and Saccharomyces cerevisiae are two important fungi in both research and industrial applications of protein production and genetic engineering due to the inherent ability. For example, S.cerevisiae can produce important proteins from wide ranged sugar from ligno-cellulose to methanol. Accurate genome-scale metabolic networks (GMNs) of the two fungi can improve biotechnological production efficiency, drug discovery and cancer research. Comparison of metabolic networks between fungi brings a new way to study the evolutionary relationship between them. There are two basic steps for modeling metabolic networks. The first step is to construct a draft model from existing model or softwares such as the pathway tool software and InterProScan. The second step is model simulation in order to construct a gapless metabolic network. There are two main methods for genome-wide metabolic network reconstruction: constraint-based methods and graph-theoretical pathway finding methods. Constraints-based methods used linear equations to simulate the growth under your model with different constraints. Graph-theoretical pathway finding methods use graphic approach to construct the gapless model so that each metabolite can be acquired from either nutritions or the products of other gapless reactions. In my thesis, a new method designed by Pitkänen [PJH+ 14] is used to reconstruct the metabolic networks of Pichia pastoris and Saccharomyces cerevisiae. Five experiments were developed to evaluate the accuracy of the CoReCo method. The first experiment was to analyze the quality of the GMNs of Pichia pastoris and Saccharomyces cerevisiae by comparing with the existing model. The second and third experiments tested the stability of CoReCo constructed under random mutation and random deletion of the protein sequence simulating noisy input data. The next two experiments were done by considering different number of phylogenetic neighbors in the phylogenetic tree. The last experiment tested the effect of the two main parameters (acceptance and rejection thresholds) when CoReCo filled the reaction gaps in the final step.

Simultaneous Localization and Mapping (SLAM) research is gaining a lot of traction as the available computational power and the demand for autonomous vehicles increases. A SLAM system solves the problem of localizing itself during movement (Visual Odometry) and, at the same time, creating a 3D map of its surroundings. Both tasks can be solved on the basis of expensive and spacious hardware like LiDaRs and IMUs, but in this subarea of visual SLAM research aims at replacing those costly sensors by, ultimately, inexpensive monocular cameras. In this work I applied the current state-of-the-art in end-to-end deep learning-based SLAM to a novel dataset comprising of images recorded from cameras mounted to an indoor crane from the Konecranes CXT family. One major aspect that is unique about our proposed dataset is the camera angle that resembles a classical bird’s-eye view towards the ground. This orientation change coming alongside with a novel scene structure has a large impact on the subtask of mapping the environment, which is in this work done through monocular depth prediction. Furthermore, I will assess which properties of the given industrial environments have the biggest impact on the system’s performance to identify possible future research opportunities for improvement. The main performance impairments I examined, that are characteristic for most types of industrial premise, are non-lambertian surfaces, occlusion and texture-sparse areas alongside the ground and walls.

Externally visual characteristics such as hair, skin and eye pigmentation or clothes have always been used for suspect identification. Also the opposite - to link unknown body parts or only DNA to a person - is recently tried to be introduced into forensics. I am testing the feasibility of one such method, IrisPlex, in Finnish population. The IrisPlex method was first published by Walsh et al. (2011). IrisPlex uses six single nucleotide polymorphisms (SNPs) in different genes found to affect the most in eye color variation. SNPs were detected from DNA sample by single base extension method, SNaPshot. Based on this information from DNA of a participant, the prediction of the most probable eye color was generated with a multinomial regression model. Also the genotypic information in the six loci and differences between Eastern and Western Finns were studied. In addition, this study supplements the knowledge of eye color frequencies across Europe. This study revealed that IrisPlex does work appropriately in Finns when detecting the blue and brown eye colors: 80% of the study participants' eye colors were predicted correctly. The biggest weakness of IrisPlex is its incapability to predict the intermediately colored eyes. Prediction probability differences between genders were not detected. In the study population 60% of the participants had blue eyes (28 individuals), 13% had brown eyes (6 individuals) and 28% (13 individuals) intermediately colored eyes. When the eyes were divided into two categories, the portion of blue-eyed participants was 77% (36 individuals) and brown-eyed participants 23% (11 individuals). These results are consistent with previous studies and update the color frequencies. Genetic segmentation of Finnish people in Eastern and Western Finland has been established in multiple studies. In addition to previous ones, this study is consistent with the genetic segmentation theory between Eastern and Western Finland. Darker eyes were observed slightly more frequently in participants with north-east heritage than in participants with south-west heritage. However, the studied populations were small and the result was insignificant. Additionally, the studied population sample represents the narrow gene pool of the Finns; almost half of the participants, 42%, reported all or 3 of their grandparents to have been born in the same village. The allele and genotype frequencies were also studied and compared to another study, in which these SNPs have been studied in the Finns and the results were consistent. Altogether, this study strengthens the evidence that IrisPlex has potential in forensic, archeological and anthropological applications even in genetically isolated populations as the Finns. This study supports the IrisPlex method to be further developed and especially addresses the need for better sensitivity for intermediate eye color.

Central counterparties (CCPs) interpose themselves between the counterparties to contracts traded on financial markets, becoming the buyer to every seller and the seller to every buyer. Over the past two decades, CCPs have grown into some of the world’s most interconnected financial market infrastructures clearing financial instruments worth trillions of dollars a day. A possible default of a CCP has been described as an extremely high-impact event and notable academics have considered that a default of a CCP would require a large-scale tax payer bail-out. CCP-related financial stability concerns have grown especially in the European Union (EU) as the United Kingdom (UK) exited the EU on 31 January 2020 (Brexit). Many of the world’s leading CCPs are located in the City of London and Brexit has meant that these CCPs will be no longer authorised and supervised pursuant to the European Market Infrastructure Regulation (EU) No 648/2012 (EMIR). They are moving out of the EU’s jurisdiction and becoming “third country CCPs” from the perspective of the EU regulation. The EU confronts these growing financial stability concerns by revising EMIR with European Market Infrastructure Regulation (EU) 2019/2099 (EMIR 2.2). EMIR 2.2 has entered into force on 1 January 2020 and it introduces an unconventional approach to regulate third country CCPs. This approach includes a requirement for third country CCPs to accept direct regulation and supervision of the EU authorities despite the CCPs being located outside the EU’s jurisdiction. In addition, the regime enables the controversial “location policy”, a mechanism to compel third country CCPs to relocate into the EU. This research examines the tensions in extraterritorial regulation of systemically important CCPs and how EMIR 2.2 succeeds in its objective of reinforcing the overall stability of the Union financial system in relation to third country CCPs. The first part examines extraterritorial regulation and supervision of systemically important CCPs through the financial trilemma, a theory developed in economics and international governance, to discover the underlying tensions. The aim is to arrive at a framework that enables meaningful evaluation of EMIR 2.2. The second part evaluates whether EMIR 2.2 is capable to achieve its objective.

The exploration of mineral resources is a major challenge in a world that seeks sustainable energy, renewable energy, advanced engineering, and new commercial technological devices. The rapid decrease in mineral reserves shifted the focus to under-explored and low accessibility areas that led to the use of on-site portable techniques for mineral mapping purposes, such as near infrared hyperspectral image sensors. The large datasets acquired with these instruments needs data pre-processing, a series of mathematical manipulations that can be achieved using machine learning. The aim of this thesis is to improve an existing method for mineralogy mapping, by focusing on the mineral classification phase. More specifically, a spectral similarity index was utilized to support machine learning classifiers. This was introduced because of the inability of the employed classification models to recognize samples that are not part of a given database; the models always classified samples based on one of the known labels of the database. This could be a problem in hyperspectral images as the pure component found in a sample could correspond to a mineral but also to noise or artefacts due to a variety of reasons, such as baseline correction. The spectral similarity index calculates the similarity between a sample spectrum and its assigned database class spectrum; this happens through the use of a threshold that defines whether the sample belongs to a class or not. The metrics utilized in the spectral similarity index were the spectral angler mapper, the correlation coefficient and five different distances. The machine learning classifiers used to evaluate the spectral similarity index were the decision tree, k-nearest neighbor, and support vector machine. Simulated distortions were also introduced in the dataset to test the robustness of the indexes and to choose the best classifier. The spectral similarity index was assessed with a dataset of nine minerals acquired from the Geological Survey of Finland retrieved from a Specim SWIR camera. The validation of the indexes was assessed with two mine samples obtained with a VTT active hyperspectral sensor prototype. The support vector machine was chosen after the comparison between the three classifiers as it showed higher tolerance to distorted data. With the evaluation of the spectral similarity indexes, was found out that the best performances were achieved with SAM and Chebyshev distance, which maintained high stability with smaller and bigger threshold changes. The best threshold value found is the one that, in the dataset analysed, corresponded to the number of spectra available for each class. As for the validation procedure no reference was available; because of this reason, the results of the mine samples obtained with the spectral similarity index were compared with results that can be obtained through visual interpretation, which were in agreement. The method proposed can be useful to future mineral exploration as it is of great importance to correctly classify minerals found during explorations, regardless the database utilized.

Large graphs often have labels only for a subset of nodes. Node classification is a semi-supervised learning task where unlabeled nodes are assigned labels utilizing the known information of the graph. In this thesis, three node classification methods are evaluated based on two metrics: computational speed and node classification accuracy. The three methods that are evaluated are label propagation, harmonic functions with Gaussian fields, and Graph Convolutional Neural Network (GCNN). Each method is tested on five citation networks of different sizes extracted from a large scientific publication graph, MAG240M-LSC. For each graph, the task is to predict the subject areas of scientific publications, e.g., cs.LG (Machine Learning). The motivation of the experiments is to give insight on whether the methods would be suitable for automatic labeling of scientific publications. The results show that label propagation and harmonic functions with Gaussian fields reach mediocre accuracy in the node classification task, while GCNN had a low accuracy. Label propagation was computationally slow compared to the other methods, whereas harmonic functions were exceptionally fast. Training of the GCNN took a long time compared to harmonic functions, but computational speed was acceptable. However, none of the methods reached a high enough classification accuracy to be utilized in automatic labeling of scientific publications.

Uusien lääkkeiden sydäntoksisuus ilmenee yleensä sydämen sähköisen toiminnan häiriöinä (rytmihäiriöt) ja supistustoiminnan heikkenemisenä. Tällä hetkellä uusien lääkemolekyylien sydänturvallisuutta arvioidaan lähinnä tutkimalla yhdisteen vaikutusta hERG-ionikanavan aktiivisuuteen sekä tutkimalla molekyylin vaikutusta aktiopotentiaalin kestoon. Nykymenetelmät perustuvat yksittäisiin solumalleissa tehtäviin ionikanavatutkimuksiin ja patch clamp-mittauksiin sekä koe-eläimillä tehtäviin toksisuustutkimuksiin. Nykymenetelmät ovat työläitä ja saadut tulokset eivät aina ennusta luotettavasti lääkeaineen sydänturvallisuutta ihmisen elimistössä. Optogenetiikka on optiikkaan ja geenimuokkaukseen perustuva menetelmä, joka mahdollistaa kohdesolujen tutkimisen ja hallitsemisen valon avulla. Menetelmässä hyödynnetään valoherkkiä proteiineja, opsiineja. Kun opsiineja koodaavat geenit on siirretty kohdesoluihin, voidaan solujen toimintaa arvioida ja ohjata valolla eri aallonpituuksia hyödyntäen. Optogeneettinen menetelmä soveltuu hyvin myös sydänlihassolujen sähköisen toiminnan tutkimiseen ja sydänlihassolujen tahdistamiseen. Sydänlihassolujen sähköfysiologisilla mittauksilla voidaan havaita esimerkiksi lääkeaineiden rytmihäiriövaikutuksia. Sydänlihassolujen tahdistamismahdollisuus kokeiden aikana on tärkeää, koska osa lääkkeen rytmihäiriövaikutuksista ilmaantuu vain normaalia korkeammilla syketaajuuksilla. Tämän tutkimuksen tavoitteena oli pystyttää ja validoida uusi optogeneettinen menetelmä lääkkeiden sydäntoksisuuden arvioimiseen. Valoherkät opsiinit Optopatch ja CaViar vietiin lentivirustransduktiolla ihmisen kantasoluperäisiin (induced pluripotent stem cell, iPSC) sydänlihassoluihin. Transdusoituja sydänlihassoluja tahdistettiin sinisen laservalon avulla. Sydänlihassolujen sähköistä toimintaa arvioitiin mittaamalla aktiopotentiaaleja sekä solunsisäistä kalsiumvirtausta optogeneettisen fluoresenssi- videomikroskopian avulla. Kuvantamisdatan käsittelyyn kehitettiin Matlab-pohjainen automaattinen analyysiohjelmisto. Videomikroskopiadatasta analysoitiin ohjelmiston avulla sähköfysiologisia muuttujia, kuten sydänlihassolujen syketaajuus, aktiopotentiaalien kesto (action potential duration, APD) ja amplitudi (APA) sekä kalsiumvirtauksen kesto (calcium transient duration, CTD). Optogeneettiseen menetelmään yhdistettiin myös videomikroskopia, jonka avulla voitiin arvioida solujen supistuvuutta. Optogeneettisellä menetelmällä saatuja tuloksia verrattiin patch clamp- menetelmällä saatuihin tuloksiin. Lentivirustransduktiot eivät olleet soluille toksisia, eikä virusten todettu aiheuttavan tilastollisesti merkitsevä muutoksia sydänlihassolujen sähköisessä toiminnassa. Optogeneettisella menetelmällä rekisteröidyt sähköfysiologiset muuttujat (syketaajuus, APD, APA) eivät eronneet tilastollisesti merkitsevästi patch clamp- mittauksilla saaduista tuloksista. Tutkimustulostemme perusteella voidaan optogeneettista menetelmää pitää yhtä luotettavana kuin perinteistä patch clamp-menetelmää. Lääkeainekokeissa sydänlihassolut altistettiin E-4031:lle (hERG kaliumkanavan salpaaja). Lääkkeen annosvasteiset tutkimukset tehtiin sekä spontaanisykkeellä että 1 Hz:n ja 2 Hz:n taajuuteen tahdistetuilla rytmeillä. E-4031 pidensi pienillä pitoisuuksilla aktiopotentiaalin kestoa, jota seurasivat suurilla pitoisuuksilla havaitut varhaiset jälkipolarisaatiot (early afterdepolarization, EAD) ja lopuksi sykkeen loppuminen. Patch clamp- ja supistuvuusmittauksissa saatiin esille samanlaiset vasteet E-4031 lääkealtistuksille. Samankaltainen, mutta E-4031:a vaatimattomampi lääkeainevaste saatiin esiin, kun solut altistettiin JNJ-303:lle (IKs kaliumkanavan salpaaja). Sydänlihassolujen optogeneettinen menetelmä soveltuu korvaamaan ja täydentämään perinteistä patch clamp-menetelmää uusien lääkkeiden sydäntoksisuustutkimuksissa. Yksittäisiin ionikanavatutkimuksiin soveltuu edelleen parhaiten patch clamp-menetelmä. Nopeutensa vuoksi optogeneettinen kuvantaminen soveltuu erityisesti lääkeaineaihioiden teho- ja turvallisuusseulontaan. Ei-invasiivisena menetelmänä optogenetiikka mahdollistaa myös pitkäaikaiset lääkeainealtistuskokeet.

Parkinson’s disease is a progressive neurodegenerative disorder which is commonly treated with Levodopa (L-dopa) and Dopa Decarboxylase (DDC)/ Catechol-O-methyltransferase (COMT) inhibitors. The main problem with this treatment is the intestinal conversion of L-dopa to dopamine despite DDC and COMT inhibition which probably occurs by the Tyrosine Decarboxylase (TyrDC) of intestinal bacteria. This study aims to find new inhibitor molecules that would have dual inhibitory effects towards both DDC and TyrDC enzymes. Currently, available DDC inhibitors cannot inhibit the bacterial TyrDC enzyme. A recently found TyrDC inhibitor (S)-α-Fluoromethyltyrosine (AFMT) is not able to inhibit the human DDC enzyme, respectively. The dual inhibition of both decarboxylases could reduce the dosing frequency and side effects related to L-dopa. In addition, the object of this study is to produce the human DDC enzyme by DNA recombinant technique as well as develop and optimize a biochemical DDC inhibition assay to study the effect of selected small molecule compounds towards inhibition of DDC and L-dopa conversion in E. faecalis model by previously developed cell-based assay. The human DDC was successfully produced in a TB medium with a yield of 1.8 mg/mL. The Km value of DDC for L-dopa was found to be 34 μM which indicates a high affinity for L-dopa. In the optimization of the DDC inhibition assay, the sample volume of 80 μL and incubation time of 3 h with detection reagent was found to give the highest fluorometric signal with sufficient robustness. In the initial screening of test compounds, 14 % of the compounds (n=59) were classified as active towards human DDC, while 31 % of the compounds were active towards L-dopa conversion in the E. faecalis model. Of those compounds, five were having dose-dependent dual inhibitory effects, but the IC50 values of them were higher compared to either carbidopa or AFMT. The most effective compounds were 8009-2501 (IC50 37 μM in E. faecalis model and 19 % inhibition at 1000 μM towards DDC enzyme) and 8012-3386 (IC50 248 μM in E. faecalis model and 37 % inhibition at 1000 μM towards DDC enzyme). However, this study confirms the possibility to find dual decarboxylase inhibitors. By optimizing the structures as well as investigating the mechanism of action, selectivity, and structure-activity relationships of the most active compounds, it is possible to find more effective dual inhibitors in the future.

Auctions are in the core on the field of dynamic pricing. Prices alter as a function of time, by either ascending or descending, and the objective of this kind of pricing mechanism is to allocate the good for the one who is willing to pay the winning price. Northwestern University has used a pricing mechanism (Purple Pricing) combining characteristics of descending auction and dynamic pricing for several years in order to sell tickets to the university’s basketball teams’ home matches. The aim of this thesis is to examine and evaluate the functioning of this kind of mechanism. The main sources for the material and content for this literary survey can be categorized to three branches. Models and main theoretic results were provided by using related economic literature, practical model of Purple Pricing was taken as a research topic separately, and some of key facts regarding functioning of the pricing mechanism were gathered through enquiry. Main results were that Purple Pricing can be modelled by using Bellman equation and characterised as a descending auction where agents (the university and the spectators) interact with both sides having own separate maximization problems. The agents can be distinguished into seller and bidders. With certain assumptions regarding the agents, model can be solved. The solution in itself provides optimality conditions which maximize the allocation (and revenue maximization) problem of the organizer (the university). The functioning of the model has caused result concerning both ticket allocation, but also real life side effects. Besides being socially efficient in allocating the tickets, its merits have been likes of the disclosure of demand curve and abolishment of black market. Although due to its special pricing policy it was not able to reach optimum (maximum) revenue, increase in revenue did occur. Similar pricing mechanisms with context-related adjustments have potential to be used in Finnish football.