List of PhD projects for the second periood of admission 01.05.-01.06.2020

Supervisor: Tsipe Aavik

Loss and fragmentation of natural and semi-natural habitats jeopardizes seriously the richness of plant species depending on these habitats. Furthermore, recent landscape changes are most likely triggering the evolutionary response of plant traits to fragmentation, in particular of those traits related to the reproduction and dispersal of plants. Heterostyly, which is a floral polymorphism particularly common in the genus Primula, has evolved to promote outcrossing mediated by pollinating insects and to avoid possible inbreeding. In plant populations at equilibrium, the frequency of individuals with different morphs (in Primula, there are two types of morphologically different flowers) is usually equal. However, contemporary habitat fragmentation as well as the loss of pollinating insects may shift this balance away from equal distribution. In extreme cases, habitat loss could lead potentially to the evolution of homostyly and self-compatibility. The proposed doctoral project will assess patterns of heterostyly of the grassland plant Primula veris in Estonia as well as across Europe in regions characterised by different level of grassland fragmentation. Heterostyly data will be related to landscape variables to assess the potential impact of landscape change on this interesting reproductive characteristic. In addition, the study will evaluate the role of heterostyly in maintaining the genetic diversity and related viability of plant populations in the context of habitat fragmentation. Obtained information can serve as a basis for improving plant conservation and monitoring. Because the project will make use of data collected within the frames of citizen science campaigns, it will offer excellent opportunities for communicating plant science.

Supervisor: Carlos Perez Carmona

Natural communities have undergone dramatic changes in recent centuries, to the point that the world is now facing a sixth mass extinction. Understanding the consequences of the projected extinctions in ecosystem structure and functioning is a key objective of ecology. These consequences are related most strongly with the functional traits of species. However, most measures of global biodiversity are based on species lists, rather than on considering their functional traits. This project aims at evaluating the functional consequences of species extinctions at different spatial scales, from the global scale to the local scale. With this information, we will study what are the main factors that affect the vulnerability to extinctions of biological assemblages. Then, we will attempt to improve our ability to understand the consequences of different drivers of global change on ecosystem functional structure. Determining patterns of extinction risk and their related drivers will improve our understanding on global biodiversity, making biodiversity conservation planning more efficient.

Supervisors: Leho Tedersoo, Sergei Põlme

The main purpose of this project is to evaluate continental-scale biodiversity of fungi and oomycetes in African soil and to detect variables that drive these differences. Fungi and oomycetes are identified using taxon-specific primers and the resulting high-quality sequences are stored in UNITE and GBIF reference databases. From necrotic plant tissues, both pathogens and potential biocontrol agents are isolated, followed by inoculation trials and experiments to discover efficient fungal strains for use in biocontrol.

Supervisor: Aurele Toussaint

Freshwater fish are among the most imperilled taxa by human activities. During the last two centuries, the species composition of almost all rivers has changed under the introduction of non-native species coupled or not with extinction of native species. Nowadays, the climate change can also amplify this global trend and hence dramatically affect the ecosystem functioning. Estonian rivers are not shielded from this situation, moreover they are among the most invaded rivers in Europe. This project will quantify how the Estonian freshwater fish assemblages have changed across time under introduction/extinction but also climate change. For that we will develop a multifaceted approach by coupling the taxonomic with the functional facet of diversity. We will first study the historical patterns and the determinants of taxonomic and functional Estonian freshwater fish assemblages and then study how they have changed across the world. Historical and temporal data are already available and will be completed by fieldwork during the project. By disentangling the determinants of the historical assemblages of species, we will be able to analyse the functional role of each species and its potential consequences in terms of ecosystem functioning. Determining patterns of introduction and extinction will help to point out the species with higher risk of extinction or higher effect on ecosystem functioning and then making biodiversity conservation planning more efficient.

Supervisors: Arvo Tullus, Pille Mänd

It is predicted, that climate change will cause a shift towards wetter conditions in northern Europe. Wetter climate affects both carbon assimilation due to stomatal closure during periods of very high atmospheric humidity as well as accessibility of nitrogen due to different mobility of nitrogen compounds in moist soil, affecting thus significantly tree growth. As tree growth is crucial both from economic and environmental (e.g. CO2 mitigation capacity) points of view, it is important to understand the underlying physiological effects of climate change-induced precipitation raise on key species of the Estonian forests – Norway spruce and Silver birch.
Current doctoral project will address the question, how wetter climate will affect tree growth and carbon and nitrogen emissions in spruce and birch stands. Considering the different resource-supply conditions under intra- and interspecific competition, the effect of wetter climate on tree growth and physiology will be explored comparatively in mixed versus pure forest stands. In addition, it will be clarified whether competitive advantage (overtopping the neighbouring trees) will provide an advantage also in acclimation to wetter conditions and which physiological processes drive the size-dependent acclimation capacity of trees. The project will considerably advance the science-based decision-making in forestry and environmental issues, while unveiling the effect of wetter climate on tree growth and reactive C and N compound emissions to the atmosphere depending on forest type, including pure and mixed stands of Norway spruce and silver birch.

Supervisors: Sten Anslan, Leho Tedersoo

The main purpose of this project is to develop practical tools for bioinformatics analysis of long DNA reads derived from third-generation DNA sequencing platforms. It is critical to develop scripts for custom segmentation of long reads for efficient recognition of chimeric sequences and relatively conserved and unconserved regions for phylogenetic placement and species-level recognition, respectively. The collection of scripts and reference data sets will be integrated into the PipeCraft workflow to enable easy data analysis for non-bioinformaticians.

Supervisors: Hanno Evard, Ivo Leito

The aim of this research is to significantly improve the performance of lateral flow assay (LFA) tests. LFA tests are simple and cheap but their applications are restricted by limited analytical performance. The main problems are sensitivity and repeatability. Better LFA performance can, for example, lead to wider screening of disease markers in the population which is not feasible at the moment due to currently used complex and expensive technologies.
The main goal of  PhD work is to perform research into a fundamentally novel directions to significantly improve the analytical performance of LFA. More specifically, the following research directions will be taken:
(1) use of screen printing and novel materials as the substrate for the LFA. This will allow achieving higher repeatability and produce complex microfluidic channels;
(2) a microfluidic channel system will be developed that allows several sequential and automatic analysis steps to be made. This will allow significant increase in the signal strength provided by the test;
(3) A system will be developed that allows automatic release of solvent to the LFA. This will allow precise sample measurement leading to higher reproducibility.
The proposed solutions are novel, and 4 research articles will be written during the PhD. Research in the direction of improving LFA has already started and will be funded from PUT PSG431. Therefore, necessary knowledge and funding to perform the research are available.

Supervisor: Ivo Leito

NMR has been used for pKa measurements by different groups, but its potential in investigating acid-base equilibria has still been largely underutilised. Ionization information in NMR is typically obtained from signal positions (under high exchange rates), which can be measured with high accuracy, as opposed to signal intensities. Because of this, NMR measurement is typically highly accurate, often tolerates partial decomposition of compounds. In addition to the ionization ratio, NMR enables obtaining additional information – side-equilibria, conformers, etc. Reliable NMR pKa measurement is possible in almost any solvent and contrary to the common perception, non-deuterated solvents can often (mostly) be used, even with 1H NMR.
The main goal and innovativeness of this project is solving a set of “big challenges” in acid-base chemistry using NMR as key technique (supplemented by other techniques as needed). Specific research questions/tasks: Acidity of sulfonic acids in water; Basicity of Carbenes; Acidities of polyfluorinated organic compounds; Experimental realization of the Unified acidity scale in low-polarity solvents.

Supervisors: Kaido Tammeveski, Margus Marandi, Marek Mooste

Currently, one of the limitations for the large-scale commercialisation of the fuel cells is the use of Pt-based oxygen reduction reaction (ORR) catalysts. Therefore, an alternative non-precious metal catalyst material is needed for the replacement of Pt-based cathodes. Recently, the use of conducting polymers for the preparation of platinum group metal-free materials for fuel cell applications has grown due the attractive properties of the conducting polymer-derived ORR catalysts (CPDC) such as low-cost and facile preparation routes. For tuning the properties of CPDC materials towards higher ORR activity, doping with heteroatoms (e.g. S, N, P, B) and transition metals (e.g. Fe, Co, Mn) can be performed. The aim of the present PhD study is to prepare the sulphur and nitrogen codoped catalysts using the pyrolysed polymer blends for this application. To study the influence of the morphology of polymer materials on the ORR activity, chemical and electrochemical polymerisation methods will be both used. The electrochemical polymerisation is generally preferred for better control over the polymer properties. In present work, the environmentally greener solvents (e.g. water) will be preferred as the polymerisation medium. In addition, the inclusion of transition metals into the polymer structure will be employed to prepare highly active M-N4 type ORR-active centres into the pyrolysed polymer material. A thorough electrochemical and physical characterisation of the pyrolysed materials will be carried out together with the fuel cell testing.

Supervisors: Ave Sarapuu, Kaido Tammeveski

For the commercialisation of low-temperature fuel cells and metal-air batteries, development of inexpensive non-precious metal catalysts for electrochemical oxygen reduction reaction (ORR) and oxygen evolution reaction (OER) is crucial. Transition metal-containing nitrogen-doped carbon nanocatalysts for ORR can be prepared by pyrolysis of organic precursors, nitrogen and metal (Fe, Co, Mn, etc.) sources. The metal-air batteries employ bifunctional ORR/OER catalysts, which can be hybrids of ORR-active and OER-active (transition metal oxides, sulphides, nitrides, carbides, etc.) materials. Phenolic compounds, such as alkylphenols derived from oil shale, can be used as inexpensive and abundant organic precursors for ORR and OER catalysts.
The aim of the project is to develop method(s) for preparing active and stable nanocarbon-based non-precious metal catalysts for the low-temperature fuel cells and metal-air batteries. Catalysts for ORR and OER will be synthesised by pyrolysis of phenolic precursors. The composition and porous structure of the catalyst will be tailored for efficient O2 electrocatalysis and mass transfer, using template-based methods. The electrocatalytic activity of the synthesised materials will be evaluated and short-time stability tests carried out. The best performing catalysts will be extensively characterised by physicochemical methods, tested in membrane-electrode assemblies at anion exchange membrane fuel cell conditions and in metal-air batteries.

Supervisor: Erki Enkvist

The aim of the project is design and synthesis of novel organic molecules that have efficient emission from exited triplet state. To achieve the objectives we are going to develop organic compounds with room temperature phosphorescence and conjugate them with bright fluorescent dyes. These kinds of tandem compounds have efficient FRET from exited triplet state to singlet state of the fluorescent acceptor resulting in very efficient overall emission from triplet state. Photophysical properties in the new compounds will be studied and their feasibility to be applied in OLED devices will be studies in cooperation with foreign partners.

Supervisors: Vladislav Ivaništšev, Piret Pikma, Enn Lust

Nowadays, in the development of all electronic devices, one of the priorities is the miniaturisation of the devices without the loss of performance. One of the solutions to achieve this goal is offered by molecular technology. The formation of nanostructures with well-controlled properties and functionality would allow scaling the size of the transistor to one molecule, which in turn would make it possible to save magnitudes of more information to one microchip.
The main aim of this doctoral project is to model the physicochemical properties of single molecules and self-assembled monolayers at the metal electrode–ionic liquid electrolyte interfaces. The results of the planned density functional theory calculations will be matched with the experimental microscopy and spectroscopy data to develop the general molecular theory of the two-dimensional condensation and single-molecule conductivity phenomena. We plan to narrow the gap between experiment and theory by considering the influence of the electrolyte and substrate in our calculations. Furthermore, by focusing on nickel and cadmium as electrode materials for molecular technology, we will look for a cost-effective alternative to noble metals, e.g. gold, and implement green thinking.

Supervisor: Ago Rinken

Total Internal Reflection Fluorescence Microscopy (TIRFM) enables achieve single-molecule sensitivity in the monitoring of events in the vicinity of the surface even in the presence of an excess of fluorophores in solution. This can be achieved only if the nonspecific binding to surfaces has been minimized experimental conditions carefully optimized. The main objective of the project is the development of a corresponding multiwell assay system and implementation this for the characterization of biomolecular interactions. The Neuropeptide Y1 receptors in baculovirus particles with its fluorescent ligands will be used for the validation of the assay system. During the next steps of the project, the developed system will be used for biochemical characterization of extracellular vesicles and receptors of protein hormones.

Supervisor:  Siim Kukk

Electrospray is a method commonly used to ionize compounds in mass spectrometers and produce polymer films and fibres. However, it has been shown to accelerate organic chemistry reactions occurring in the charged droplets primarily by increasing the surface-to-volume-ratio and reactant concentration. Quite recently it has been demonstrated that electrospray could also be used to polymerize monomers to produce polymer films in one step. Currently only polyamides have been shown to polymerize via electrospray. Polyamides have a variety of uses, ranging from very strong fibres to high-performance membranes used to purify or desalinate water. This PhD project aims to further develop methods to produce various other polyamides and polymers of by spraying different monomers with variable electrospray parameters. The possibility of polymerization catalysis in electrospray by metallic copper will be investigated. If monomers could be polymerized and polymer films or fibres could be prepared in a single electrospray step, the production of polymer materials will be more efficient.

Supervisors: Koit Herodes, Ivo Leito

Lignin is the second most abundant renewable organic feedstock – every year plants on Earth produce ca 35 billions of tons of lignin. Lignin’s structure would enable using it for production of a myriad of chemicals and materials. Yet, only a tiny fraction of lignin is actually valorized. In the first steps of all lignin valorization technologies very complex mixtures of organic compounds are obtained, from which the desired components need to be separated. The obtained ligneous mixtures are highly complex, both from the analysis and separation point of view, which strongly hinders utilization of lignin as a resource.
This project will work towards lignin valorization by (1) developing a multiinstrumental (GC-MS, LC-MS, SEC, FT-ICR, NMR and FT-IR, with possible addition of small-scale preparative chromatography) analytical approach for elucidating the composition of ligneous mixtures and (2) developing combined separation approaches for such mixtures, which in the future could be used in industry, involving extraction (with finding suitable solvent systems), distillation and precipitation.

Supervisor: Koit Herodes

Biological samples (biofluids, tissues) contain a vast numbers of small molecules – metabolome. These compounds reflect the processes taking place in the organism. Metabolome studies are important for understanding these processes, either normal or ones caused by stress/disease. Although the most advanced analytical methods, like liquid chromatography-mass spectrometry (LC-MS), are used, larger part of metabolome is still not characterized.
This project will use derivatization as a tool for LC-MS analysis of amino group containing metabolites. Derivatization adds a specific group to the amino group of metabolites thereby improving separation, selectivity and limit of detection of the LC-MS analysis. The approach, termed derivatization-targeted LC-MS analysis, is used for metabolome profiling of extracts of plant and animal origin.

Supervisors: Nadezda Kongi, Vladislav Ivaništšev

This work focuses on the development of the metal-organic frameworks (MOF) based multifunctional catalysts for future sustainable energetics. By utilizing molecular dynamics and density functional tight-binding simulations of oxygen evolution reaction (OER), hydrogen evolution reaction (HER), and oxygen reduction reaction (ORR), we aim to achieve a molecular-level understanding of their mechanisms. Large scale models will be obtained by reproducing the synthesis of the catalyst starting from the original MOF and comparing the results with the experimental data. Adsorption energetics at single-atom and bi-metal catalysts sites will be studied accounting for the electrolyte solvation and electrode electronic structure. Kinetics of the reactions will be modelled based on the data from the electronic structure calculations. Overall, the project aims to develop a workflow for modelling electroactive materials for designing new MOF based multifunctional electrocatalysts.

Supervisors: Nadezda Kongi, Kaido Tammeveski

In this PhD project new classes of electrocatalyst materials for their use in sustainable energy conversion and storage devices will be developed. Motivated by the urgent need for improved, earth-abundant-based (i.e. Pt-free) catalysts, an original set of metal-organic frameworks (MOFs) based electrocatalysts will be designed and developed. This will be accomplished by customization of catalyst structures, involvement of different earth-abundant metals and developing the knowledge-based design of the electrocatalytically active sites. Multifunctional ability to efficiently catalyze simultaneously different oxygen and hydrogen electrochemical reactions will be explored by using state-of-the-art electrochemical techniques combined with physical characterisation methods.

Supervisor:  Radwa Elshawi

The main goal of this doctoral work is to develop an interactive, user-guided and interpretable automated machine learning framework. The framework should empower its users to develop their own satisfying and trusted models. It should reduce the burden on for data scientists and domain experts for going through the time-consuming process of building and deploying scalable machine learning models. The end goal is to democratize the data science process by facilitating it to non-sophisticated technical end-users and enable them to harness the power of their data.

Supervisor: Naveed Muhammad

As human drivers we instinctively use our understanding of other road users’ behaviour for safe and efficient driving. This however is a non-trivial task for autonomous vehicles. This is further complicated by the fact that human road user behaviour changes with varying context (which again we, as humans, have no problem adapting to).
This project focuses on investigating context-aware agent behaviour modelling and prediction techniques. The project begins with studying models that encode agent behaviour and interactions in traffic scenarios. It then investigates behaviour models that can be used for varying contexts, as well as devising a framework for management of context-adaptive behaviour models. The expected outcome of the project comprises of models, algorithms and a framework, that can be employed by autonomous vehicles for efficient yet safe manoeuvres and trajectories.

Supervisor: Kallol Roy

Artificial intelligence systems named predictive policing is used to predict when and where a crime will take place across a city, bringing to the era of data-driven law enforcement. But surprisingly predictive policing shows racial discrimination, as it is trained on data based on a discriminatory source e.g. criminal data, of racial minorities that are disproportionately of police attention. This creates social feedback loops, leading to the police officers being sent repeatedly to certain neighborhoods – typically ones with a high number of racial minorities regardless of the true crime rate in that area. Predictive policing software seems to learn from reports recorded by the police – which may be higher in areas where there are more police – rather than from underlying crime rates. Thus, the crime rate in some neighborhood is overestimated, without taking into account the possibility that more crime is observed there due to the fact that more officers have been sent there. This project aims to develop a mathematical model of predictive policing in Estonian context. The model will explain why feedback loop occurs and demonstrate how to change the inputs to the predictive policing system, so the feedback loop does not occur, and thus allowing the true crime rate to be learned. The project will demonstrate the way in which reported incidents of crime (those reported by residents) and discovered incidents of crime (i.e. those directly observed by police officers) interact in real time. The project will suggest the proper interventions to attenuate the degree of runaway feedback and thus do unbiased policing.

Supervisors: Helle Hein, Karin Täht

This research project aims to better the understanding of study processes of computer science students using specifically built online tools, adapted questionnaires for CS curricula and learning analytics. To detect and promote learnable factors that could predict students’ academic success, might help to develop more data-driven study guides to academic success as well as psychological well-being for future CS students and educators. Also, the data could be used to shape the curricula in a more student-oriented manner.

Supervisor:  Anne Laius

The goal of the research is the reconceptualisation of environmental education into up-to-date education for sustainable development (ESD) in all aspects of environment – ecological, economical, social and cultural – into all subjects of general shooling at all stages in order to achieve the  scientifically  justified and unified conceptualizations of ESD as cross-cutting topic of curriculum (environmental protection and sustainable devlopment). According to the results of PISA 2015 and PISA 2018, the environmental awareness of Estonian students is one of the highest in Europe, but one of the lowest in the field of environmental behavior and  therefore, in order to increase the effectiveness of ESD, it is necessary to bridge the gap in perceptions of ESD. To achieve the goal of this research, the following tasks are needed: (1) metanalyses of the literature of conceptualizations of ESD and the pedagogical methods of promoting  competencies needed for sustainable development (SD); (2) creation of theoretically justified instrument for investigating the different subject teachers’  and different age groups of students’ conceptualizations of SD and its’ components; (3) conducting the large-scale investigation for surveying the different conceptualizations of SD and ESD with the aim to integrate the ESD into all subjects and stages of general education. The data will be collected in cooperation with University of Tartu Natural History Museum, Estonian Museum of Natural History, informal natural science organisations and Estonian schools.

Supervisors: Gholamreza Anbarjafari, Madis Kiisk, Andrea Giammanco

This PhD proposal focuses on conducting deeper research on cosmic ray tomography and utilising advances on machine learning and computer vision in order to develop an AI-enabled autonomous system that will be able to reconstruct scanned objects and as a result conduct object detection and recognition materials. This project will introduce a novel combination of a new conceptual design of a tomographic detection system with a new tomographic object reconstruction method, enabling high detection sensitivity for low-Z materials. This combination enables high spatial detection resolution and high sensitivity for low-momentum spectrum of the cosmic ray flux. Furthermore, it exploites a novel reconstruction method applied to spectrometric filtering of entering lepton ray flux based on their momentum, and selection of a tomographic method least sensitive for variations in particle’s momentum and a novel image processing methodology for enhanced object detection and noise discrimination. Moreover, within this PhD work, a multi-image object detection based on multi-parameter (e.g scattering density and attenuation length) will be introduced. It is expected that this PhD work, at the end, introduces a novel scanning concept - invisible corridor-like scanning systems for simultaneous scanning of humans and their luggage, e.g. all-in-one scanning approach in border and security check gates. This will help customs and security forces by bringing on a non-interrupted, simultaneous scanning of vehicles and passengers, which is able to conduct an automated decision making and clearing procedures.

Supervisor: Andres Merits

Alphaviruses (family Togaviridae) are medically important arboviruses. They replicate their positive-strand RNA genomes using RNA replicase complexes formed in cytoplasm of infected cells; the complexes consist from virus-encoded RNA polymerase, other virus-encoded proteins, RNAs and number of host cell derived components. The list of proposed research topics includes:
1. Analysis of compatibility of replicase proteins using trans-replicases and chimeric viruses.
2. Analysis of functions of replicase proteins using trans-replication assay and functional complementation.
3. Analysis of mechanisms of actions of antiviral compounds and host factors.
4. Development and use of replicase-based biosensors – stable cell lines and transgenic animals.
The project is experimental and includes analysis of RNA replication, interaction of replicase components with host factors, experiments with recombinant viruses, transgenic cell lines and, in collaboration with other research groups, with transgenic mice.

Supervisors: Saoni Banerji, Alvo Aabloo, Jordi Madrenas, Daniel Fernàndez

Otitis Media (OM), known as the second most reason after common cold for paediatric clinic visits, is an inflammatory disease of the middle ear that commonly affects paediatric populations. The existing gaps with the standard diagnostic practice (otoscopy) recognized are improper positioning and imaging leading to misclassification. The main challenge in imaging is designing a smart image sensor that integrates multi-functional integrated circuits and consumes low power. Complementary Metal Oxide Semiconductor (CMOS) fabrication process answers the solution to these challenges. It offers high integration capability, smaller area and lower cost by monolithically integrating the sensor and processing in a single chip. The proposed research aims to overcome the existing gaps of misclassification in diagnosing middle-ear infections. We propose to combat improper imaging in otoscope by development of a System-on-Chip (SoC) CMOS image sensor (CIS) monolithically integrated with novel readout circuits.

Supervisors: Dmitry Yarmolinsky, Hannes Kollist, Ebe Merilo

Stomata are microscopic pores on plant surfaces which play a central role in balancing the uptake of CO2 for photosynthesis and the loss of water through transpiration. The opening and closing of stomata is regulated by swelling and shrinking of two highly specialized guard cells surrounding the tiny stomatal airway pores. Stomatal regulation is involved in determining plant water use efficiency as well as stress resistance. Despite intensive research, there are still many gaps in our understanding of stomatal regulation. This project aims to identify novel regulators in guard cell signalling and characterize their functions in guard cells. The PhD project is based on the collection of novel stomatal mutants isolated in Plant Signal Research Group. The PhD student will carry out functional characterization of some identified stomatal mutants and will perform next-generation sequencing-based mapping and candidate gene analysis of new stomatal regulators. Identified regulators will be comprehensively examined to unravel their biological relevance, using methods of genetics, biochemistry, molecular and cell biology. The PhD project will ensure new knowledge about stomatal regulation in response to environmental factors. This information is required to develop breeding programs for crop varieties with increased water use efficiency.

Supervisor: Taavi Lehto

Antisense oligonucleotides (ASOs) are a class of RNA therapeutics that comprise enormous potential for the treatment of human disease. While nucleic acids, including ASOs, have been generally believed to have a very limited capacity for uptake by cell/tissues, recently some ASO chemistries have been shown to possess ability to do so in their free/naked form and even in quantities where they can be pharmacologically active. This free/naked uptake of ASOs has been termed as gymnosis or gymnotic uptake process of ASOs. However what is gymnosis, what are the factors that govern this process and if they could be modulated to improve the activity of ASOs remain very poorly understood. Consequently, exploring this space is of utmost importance for developing more effective ASO therapeutics in the future.
The main aim of this project is to study the process of free/gymnotic uptake of ASOs and explore the possibilities of targeting and stimulating the free uptake of ASOs. For this, we plan to study which factors do ASOs interact in the extra- and intracellular space and their impact on the cellular transport. Furthermore, we will study underlying mechanism of how ASOs are taken up and how they traffic inside the cells. Finally, we explore the possibilities of inducing the free/gymnotic uptake processes by making use of chemically modified ASO analogs.

Supervisors: Arun Kumar Singh, Karl Kruusamäe

There are two broad classes of approaches for motion planning and control. One based on classical control theories, which in turn draws from linear algebra, optimization theory, etc. The other approach, which is more recent, is to use data driven AI, specifically based on deep learning to enable a robot to self learn motion and control policies through trial and error (reinforcement learning) or by mimicking a human expert behavior (imitation learning). The control theoretic approaches require more assumptions about the problem structure but at the same time comes with strong guarantees on performance and safety. The AI-based approaches on the other hand requires no knowledge about the kinematics or dynamics of the robot and can work with direct vision inputs. However, it is difficult to predict the performance of an AI-based system in novel situations which have not been encountered during the learning process. In this research, we aim to find the sweet spot between the control theoretic and data driven AI based approaches for robot motion planning and control. In particular, we would develop ways to use AI to relax the limiting assumptions in the control theoretic approaches. At the same time, we would use the mathematical foundations of control theory to induce safety and performance guarantees in AI-based motion planning and control.

Supervisor: Tanel Tenson

Antibiotics consumption by humans and animals is creating antibiotic resistance (AMR) and many previously manageable bacterial infections are becoming increasingly hard to treat. Therefor it is important to understand the origin of AMR and circulation in environment. One of the biggest antibiotics consumer is animal husbandry. The resistant microbes and residues of pharmaceuticals are spread into the environment with sludge and manure, when are used as fertilizers to grow crops. The residues of pharmaceuticals and resistant microbes can accumulate in the soil, crops, and therefor in our food.  The most critical attention should be payed for plants what we eat raw: root vegetables. The goal of the current proposal is to characterize soil and vegetables microbiomes during different farming practices. We will also measure the concentration of pharmaceuticals in soils and plants and estimate the effects on stimulating the spread of antibiotic resistance. and vegetables microbiome.

Supervisor: Indrek Must

Soft robotics offers tools for developing robots that interact with and interface to humans as well as the environment in a fashion intrinsic to natural organisms. Stiffness variation is the key component in natural (as well as synthetic) compliant interactions; however, the deficiency of effective in-situ variable-stiffness solutions available today governs localized stiffness variation as a central goal in modern soft robotics. Electroactive materials structured in a bioinspired fashion and combined with a traditional textile platform promises an attractive platform for soft robotics. The cellular structure, inspired by the microstructure of natural skin, offers a safe and scalable platform for designing minimally invasive medical devices as well as wearable healthcare applications.

Supervisor: Mart Loog

Synthetic biology has been mostly focused on the design of signaling circuits based on transcriptional and translational control. However, the eukaryotic cells contain hundreds of protein kinases and phosphatases to enable rapid regulation of protein activities. These vast combinatorial possibilities of protein phosphorylation have not been properly used in synthetic circuits. Our recent studies have uncovered novel principles of multisite phosphorylation of proteins that present a way for linearly encode different input-output functions for kinase switches (Örd et al. (2019) Nature Struct Mol Biol., and Örd et al. (2019) Molecular Cell). This concept forms the basis of multisite phosphorylation toolbox (MPToolbox) for synthetic circuit design with various applications in biotechnology. especially for engineering yeast-based cell factories for production of high-value chemicals using biomass-derived sugars as feedstock (as an alternative to oil-based chemical industry approaches).
One of the biggest bottlenecks in microbial cell factory development is the difficulty to design metabolite sensors to measure the accumulation of metabolites or to mitigate the effects of toxic intermediary metabolites on cell growth via connecting the sensors to regulatory circuits. Current project aims to apply an innovative technique to overcome this design problem. We will apply MS-based quantitative proteomics to measure changes in protein levels occurring upon toxic stresses of selected metabolites and other compounds of interest and connect these changing protein signals via transcriptional- or direct fusion approaches with the synthetic kinase inputs developed previously in our lab. The levels or localization of target proteins that are able to mitigate the effects of stress (e.g. to downregulate metabolic enzymes over-producing the toxic intermediary metabolite) will be regulated via the kinases. For regulatory mechanism we will use previously developed protein fusion tags containing phospho-regulatable degrons and/or NLS/NES sequences. The project will significantly contribute to our ability to design yeast cell factory strains that are resistant to stresses occurring in industrial bioreactors, and that exhibit more efficient biomass growth on growth-inhibiting lignocellulose-derived sugars from local wood industry.

Supervisor:  Andres Merits

Alphaviruses (family Togaviridae) and flaviviruses (family Flaviviridae) are medically important arboviruses with positive strand RNA genomes. They replicate their RNA genomes using RNA replicase complexes formed in cytoplasm of infected cells; the conmplexes consist from virus-encoded RNA polymerase, other virus-encoded proteins, RNAs and number of host cell derived components. Replicases have also poorly understood functions not directly connected to viral RNA synthesis such as targeting antiviral defenses of the host.
The list of proposed research topics includes:
1. The trans-replication system for flaviviruses will be constructed and used to analyze cross-utilization of flavivirus RNA templates
2. Analysis of ability of replicases of 10 different alpha- and flaviviruses to use cellular RNAs for synthesis of PAMP-molecules.
3. Analysis of correlation between PAMP RNA synthesis and neurovirulence.
4. Analysis of connection between efficiency of PAMP RNA synthesis and dependence of virus from host G3BP/Rasputin proteins; we hypothesize that there is functional link between these two properties.
The project is experimental and includes analysis of RNA replication, interaction of replicase components with host factors, experiments with recombinant viruses, transgenic cell lines and, in collaboration with other research groups, with transgenic mosquitoes.

Supervisors: Reet Kurg, Mart Ustav

Malignant melanoma is the rare form of skin cancer with the high mortality rate and increasing incidence in Estonia and worldwide. The goal in melanoma diagnostics is early detection with an ability to differentiate between benign and malignant lesions. One possibility for an early and accurate diagnosis is to use liquid biopsy. The aim of the PhD studies is to find new liquid biopsy diagnostic markers for malignant melanoma.

Supervisor: Hanna Hõrak

Stomata, pores in plant leaves formed by adjacent guard cells, balance CO2 uptake for photosynthesis with water loss through transpiration. This balance can be regulated by controlling the number of stomata that are made during leaf development (stomatal density) or stomatal aperture that is regulated by changes in guard cell volume. Both stomatal development and aperture are regulated by various environmental and endogenous signals, among them the drought-inducible phytohormone abscisic acid (ABA) that suppresses both traits. Although the signal transduction pathways that control stomatal development and aperture changes are well-studied, their potential interactions or coordinated regulation have never been addressed. Here we focus on filling this knowledge gap by generating model plants with simultaneous mutations in key regulators of pathways that control stomatal density and aperture. By this approach, we aim to generate plant lines with a) improved drought tolerance without yield penalty; b) high transpiration and photosynthesis in favourable environmental conditions combined with a sensitive stomatal response to drought stress; and c) maximal transpiration and enhanced CO2 uptake for photosynthesis. Phenotypic analysis of these plant lines will give valuable information on the potential interactions of the signalling pathways that control stomatal numbers and aperture. As stomatal regulation is highly conserved in evolution, the results obtained with the model plant Arabidopsis thaliana will help to identify trait and gene combinations that could be further used in smart crop breeding to improve drought tolerance, water use efficiency or yield in field conditions.

Supervisors: Margus Pooga, Ana Rebane

Modulation of gene expression by exogenous oligonucleotides (ON) that specifically associate with the complementary targeted sequence and regulate transcription, mRNA splicing or translation provide powerful tools for biomedical research and medicine. Chemically modified ON, acting by various mechanisms are in active development as future drugs. Importantly, first ON drugs have been approved by FDA and EDA, and reached clinics very recently (e.g. eteplirsen for DMA, nusinersen for SMA, patisaran for HTMA).
The wide spread application of highly potential nucleic acid (NA) therapeutics in medicine is still hindered by the lack of efficient and side-effect free transporting systems since naked ON are very poorly taken up by cells. Cell-penetrating peptides (CPP) promote cellular uptake of various macromolecules and particles, and provide a vital alternative to viral vectors in NA delivery. CPPs condense NA cargo to nanoparticles (NP), and protect against degradation. CPPs also promote cellular uptake of formed NP by inducing endocytosis, and later initiate escape of NA from endosomes. However, the CPP-induced escape of NP from endosomes is not sufficiently efficient for majority applications. We have recently discovered a class of compounds that substantially increase the escape of NA from endosomes. These compounds increase the biological response evoked by delivered NA in cells by at least ten-fold, as estimated in reporter cell-lines based on splicing switching by ON.
In the current project we aim to elaborate novel highly efficient NAs-delivering NPs that contain endosomal escape promoting excipients. In parallel, the mechanisms that these compounds utilise in cells, will be analysed using a battery biochemical, cell physiology and microscopy methods.
After administration, NP (both synthetic and natural) are in all biological fluids covered with a protein layer, “protein corona” (PC) that “neutralises” NP and marks for elimination from organism. PC also gives NP a new identity for interaction with cells and typically reduces efficiency of delivery. We continue with analysis of PC compositions of CPP-ON NPs and implementation of this knowledge.

Supervisors: Riho Vendt, Joel Kuusk

Vast amounts of Earth Observation (EO) data are being collected from satellites as well as airborne, ground-based, and seaborne (in-situ) measurement systems all over the world as a key resource to support the decision-making in addressing environmental challenges, provide information for service providers, public authorities and other international organisations in improving the quality of life. Studies have confirmed that metrological traceability of measurements to the international system of units (SI) with related uncertainty evaluation is of crucial importance in producing reliable remote sensing data. Analysis of measurement data indicates that overall uncertainty can be reduced significantly by regular calibration and characterisation of radiometric sensors used for remote sensing. However, comparisons in stable laboratory conditions and various real environment conditions show that merely good radiometric calibration of a sensor on its own is insufficient to ensure firm traceability of measurements to SI with expected low uncertainty. The doctoral study has goals to 1) analyse most significant uncertainty sources in radiometric remote sensing measurements of waterbodies with respect to the requirements for expected uncertainty levels; 2) develop a strategy for characterisation of earth observation radiometers addressing the most significant uncertainty sources hindering the achievement of expected requirements; and 3) develop a method for characterisation and in field correction for the effect of variable ambient temperature on the responsivity of in-situ radiometric sensors.

Supervisors: Indrek Must, Kärt Ojavee

In natural organisms, sensing for awareness of the surroundings, physiological regulation, signaling, as well as defense are commonly addressed by a large number of semiautonomous but synchronized individual units that collectively form patterns. Today, the powerful technique of controlled functional pattern formation has attracted undeservedly little attention in soft robotics. In the quest for highest safety and biocompatibility, protein-based biomaterials such as silk are incorporated as the intrinsic part of an ionic electromechanical drive and patterned according to bioinspired cues. The morphing patterns regulate the robot’s exposure to the environment and the wearer’s body, affecting transpiration and enhancing sense, as only selected examples. The particular address is given to societal acceptance of wearable health and care robots. As a novel aspect, the professional applied design combines the competences in biomaterial technology with soft robotics. The ambition of this project is to design pattern-morphing wearable robots that are comfortable to use and are accepted by ordinary people as well as healthcare professionals as products for health benefit. The provided design cues are essential in developing robots for facilitating faster post-injury recovery as well as enhancing the wellbeing of all wearers. 

Supervisors: Riho Mõtlep, Andres Marandi

EU Green Deal likely ends the leading position of oil shale industry in Estonian energy sector but the direct and indirect environmental impacts of the industry will remain. So do the need to eliminate and mitigate these effects will continue for decades. The long-term effects of the oil shale industry are related to the potential environmental risks (including stability) of mining waste and waste depositories of electricity and oil production and the stability of underground mining areas.
There are scarcity of studies explaining the temporal and spatial distribution of already existing or spreading groundwater pollution from these sources. Therefore it has been extremely difficult to assess the gravity of potential groundwater pollution in areas where waste have been deposited. Particularly complex is the spreading of such potential pollution in areas where old underground mining areas exist that affect and control the movement of groundwater. Mine backfilling is an effective solution to fight land subsidence and at the same time is one of the most important and large scale operation of waste beneficial reuse.
The PhD project focuses on two related sub-topics: development of spatial models for the groundwater pollution as point and surface pollution from oil shale waste disposal site and the effect of old mining areas on groundwater dynamics; and geotechnical stability of mined areas by backfilling, assessing backfilling also as a barrier for groundwater transported pollutants.

Supervisors: Krista Alikas, Susanne Kratzer

Remote sensing of highly absorbing waters such as the Baltic Sea and boreal lakes in Sweden and Estonia has always been a difficult task. However, the satellite method can generate products from space-based sensors that offer insight into the dynamic nature of these water bodies and allow for the analysis of environmental parameters over long periods of time with substantially improved spatial and temporal resolution. This project will investigate if one the most frequent space-borne ocean colour sensors, Sentinel-3/OLCI, and if it can accurately derive water quality products such as chlorophyll a, turbidity, and colour dissolved organic matter by comparing values derived by the sensors to values measured by ship and mooring. The validation of these products is necessary to develop better processors and calibration methods for the advancement of remote sensing of optically complex waters. Many of the water quality products generated by these sensors are based on water-leaving reflectance so therefore it is also necessary to validate the water-leaving reflectance produced by Sentinel-3/OLCI. Once the validity of the water quality and water-leaving reflectance has been established, it will be possible to begin using these products also for environmental analysis. The first environmental analysis will be a long term trend study using integrated data from several space sensors to examine changes in SPM from both organic and inorganic sources, and Secchi depth in highly absorbing waters. This study will produce an insight on how to correctly merge data produced by space-borne sensors with different spatial and spectral resolution while generating a consistent products that show long term change in water parameters perhaps indicative of human-made climate change. A secondary analysis will be done by merging data derived from several sensors and space-based thermal data (SLSTR also based on S3; 1 km resolution) to determine if there are any long-term trends in the ocean colour products chlorophyll a and CDOM for Baltic Sea coastal regions and Swedish and Estonian lakes. Trends in these two varibles might be the result of human-induced climate change resulting in eutrophication and higher amounts of yearly runoff from rivers draining into lakes and the Baltic Sea basin.

Supervisors: Anders Eriksson, Tõnis Org, Georgi Hudjašov, Irene Gallego Romero

The aim of the current PhD project is to find associations between the genetic adaptation and environment-cultural changes in West Eurasia during the last 20,000 years - a time period that includes major shifts in climate, diet and pathogen exposures in this region. The main aim of the project is to investigate how recent genetic evolution has shaped variation in immune system and inflammation signalling and the way these systems interact with environmental factors such as biomolecules from food, pathogens and modern artificial substances.
Through the use of existing data from the Estonian Biobank, as well as experimental functional genomics work, we aim to dissect how key genetic variants affect the cells' biological responses to different types of exposures. Ultimately, this project seeks to identify pleiotropic effects in immune and inflammation regulatory pathways, including gene-gene and gene-environment interactions, in order to resolve individual differences in susceptibility to chronic inflammation and associated medical conditions. In addition, the project will dissect the roles of different dietary and environmental (lifestyle) factors in the susceptibility for metabolic and cardiovascular diseases in present-day populations, to better inform the use of genetic information to personalise the prediction, prevention and treatment of these diseases.

Supervisors: Michael Dannemann,  Jon Anders Eriksson

This project focuses on genetic adaptations to environmental and cultural changes in human populations of West Eurasia during the last 20,000 years, a time period that covers major shifts in climate, diet and pathogen exposures in this region.
The project will take advantage of high-quality genetic and biomedical datasets, including ancient DNA from dated human fossils, whole genome sequences from modern populations, and case-control and biobank datasets with associated genetic, biomarker and health information (such as the Estonian Biobank).
The main aim of this project is to systematically map how natural selection has shaped human genetic variation during this time period and the consequences of these adaptations for health in present-day populations. With some of the genetic variation in present-day non-Africans being inherited from admixture with Neandertals ~55,000 years ago, the project will further investigate which role this admixture has played in shaping these adaptive processes.
The outcome of this project will (1) provide important information about the role of selection and admixture in the interplay between genetics and environments over long time scales, which are difficult to obtain from contemporary populations and environments.
The project will (2) help resolving the role of different dietary and environmental (life-style) factors in the susceptibility for metabolic and cardiovascular diseases in present-day populations, which are important when using genetic information to personalise the prediction, prevention and treatment of these diseases.

Supervisors: Lili Milani, Kelli Lehto

Patients with severe mental disorders such as schizophrenia (SCZ), bipolar disorder (BIP) and major depressive disorder (MDD) show significantly higher mortality than the general population, mainly due to somatic comorbidities, particularly CVD. Studies of CVD risk factors in mental disorders show considerable association to: medication side effects, lifestyle factors such as unhealthy diets, sedentary habits and smoking. Large genome-wide association studies (GWAS) of individual phenotypes have provided extensive evidence for a large genetic susceptibility for both mental disorders and CVD. However, these studies that suggest overlapping genetic susceptibility do not point at specific mechanisms underlying CVD and mental disorder comorbidity.The aim of this PhD project is to use genomic methods to improve the understanding of the mechanisms that are underlying CVD and mental disorder comorbidity and implement this new knowledge to improve treatment outcomes in individuals with CVDs or severe mental disorders. The thesis work includes GWAS of CVDs, mental disorders, and CVDs among individuals with mental disorders using data from the Estonian and UK biobank (numbers of cases >10,000). The results of the studies will be further explored by linkage-disequilibrium score regression to investigate the extent of genetic overlap between CVDs in a psychiatric population and CVDs in the general population. In addition, the causal direction between CVDs and mental disorders will be explored using Mendelian Randomization. The treatment outcomes of the study subjects will be investigated using pharmacogenetic and genome-wide approaches. A final task, in collaboration with project partners, will be developing prediction models to target CVD risk in mental disorder populations (and vice versa) including cardiometabolic side effects of medication.
The work is part of a European collaboration project “CoMorMent”, which has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 847776.

Supervisors: Jaan Pärn, Ülo Mander

{C}{C}{C}{C}{C}{C}{C}{C}{C}{C}{C}{C}{C}{C}{C}{C}Peatlands store as much carbon as all the world’s vegetation or nearly as much as the atmosphere. Peatlands contain one-tenth of the world’s nitrogen. Drainage releases a large share of the nitrogen as N2O, a potent greenhouse gas (GHG) and the main driver of stratospheric ozone depletion. However, mires are a sizable source of methane. Thus the potential of mires as one of the greatest allies and most efficient contributors in the fight against climate change remains unclear and contested. Few regional studies include tropical peatlands, where GHG production is probably the most intensive. The current IPCC guidelines for GHG inventories are based on land-use categories while direct factors of GHG production such as soil temperature and moisture, organic matter quality and functional genes are not considered. University of Tartu holds a unique organic-soil library with accompanying GHG measurements. The proposed doctoral thesis will build landscape-level process-based GHG models that will be applicable on mires, peat grasslands and bare peatlands of the world.

Supervisors: Anto Aasa, Tiit Tammaru

Mobile and GPS positioning have been acknowledged as valuable data sources for mobility studies already for several decades. Both methods have been proved to be high quality and cost-effective alternatives to the traditional data collection methods. Using mobile positioning data as input to produce official tourism statistics is a well-known success story. Therefore, are governmental statistical offices mapping possibilities to use mobile positioning data as one possible data source to replace traditional census. Statistics covering the entire population is usually not very detail. To gather information in more detail level (household, individual) about the target population surveys are used. Conducting surveys is expensive. Therefore, the sample size is relatively small and studied period very short. Among other surveys Eurostat is organizer of Harmonised European Time Use Surveys. The Harmonised European Time Use Surveys (HETUS) are national surveys conducted in European countries to quantify how much time people spend on various activities. The main survey instruments are a household questionnaire, an individual questionnaire and a time-use diary in which respondents are asked to record their daily activities in 10-minute time slots (Eurostat). All this could be done with help of mobile phones. Mobile phones can be used as sensors to collected more detail data for surveys. Most of mobile phones are equipped with GPS and other sensors, which are used to collect locational information about phone carrier. GPS data can be used to analyse mobility patterns and detect places visited by the respondent. Overlaying derived mobility statistics with spatial context (land use, population register, buildings register, business register, services etc) can give answers to question: why the specific place was visited? And what was the activity there? As the starting platform the data collection application “MobilityLog” will be used (link).
The key innovation of this PhD thesis is a systematic development of methodological framework based on GPS-data to get answers to the questions asked during time use surveys.

Supervisors: Evelyn Uuemaa, Tiit Tammaru, Alexander Kmoch

The PhD project focusses on the nexus of changes in sea-level as a result of global warming, and the world-wide patterns in population redistribution that have brought along rapid population and urban growth taking place in coastal areas. The global forecasts of mean sea level rise by as much as 1 m by 2100 according to IPCC. Many studies have estimated the potential risks of sea level rise at local and global levels, but were mostly focussed on estimating population size that is affected by sea level rise. Less studies have analysed the land development and land reclamation in the coastal areas and the vulnerability of different sociodemographic groups in these areas. In this study, a novel approach will be developed and tested, that uses Google Street View images for training machine learning models to identify socio-demographic characteristics and status from the remote sensing images. It is challenging to estimate the socio-demographic groups living in the coastal areas because of the lack of data from traditional sources such as censuses. The specific aims for this study are 1) to assess the land use changes in coastal areas globally by focussing on built up areas from 2000 to 2020; 2) to estimate how much new land has been reclaimed from the sea from 2000 to 2020; and 3) to evaluate which socio-economic groups are growing in the coastal zone and estimate their vulnerability to sea level rise both globally and in focussed case study areas.

Supervisors: Kaido Soosaar, Ülo Mander

Managed nutrient-rich organic soils are one of the largest key sources of greenhouse gas (GHG) emissions in boreal and temperate cool & moist climate regions in Europe and also worldwide. Drained forests are usually considered as a sink of CH4 and a source of N2O. Fluxes of CH4 and N2O between forests and the atmosphere have been mostly calculated only based on GHGs exchange at the soil–atmosphere interface, excluding exchange activity of other ecosystem compartments as trees. The majority of performed field measurements on trees focuses on CH4 and N2O exchange from stems only. The fluxes from leaves and shoots are very rarely studied and completely ignored in the ecosystem GHGs studies because of technical/methodological difficulties. However, our first results show that CH4 and N2O emissions by tree foliage can significantly exceed those of stems. We therefore expect that trees might contribute even more significantly to the forest CH4 and N2O exchange than currently expected based on stem flux measurements only.
The project is based on simultaneous flux measurements in different ecosystem levels in recently established Agali Birch Forest Research Station. Within a larger research framework, these GHG flux data will be joined with detailed information of soil biogeochemistry and microbial dynamics to further improve process-based modeling of peatland GHG emissions.
The general goal of this PhD project is to improve our understanding of GHG exchanges in drained nitrogen-rich peatlands and to separate component fluxes in response to climate and vegetation controls. The main objectives of the doctoral thesis are: (1) to analyse simultaneously measured GHG fluxes from different level of ecosystem: in the soil-tree-canopy continuum; (2) to test the applicability of different soil screening methods in explaining the variation of GHG exchange rates between sites; (3) to analyse the relations between the GHG fluxes and soil microbial communities.

Supervisors: Kadri Leetmaa, Kadi Kalm, Tiit Tammaru

The intergenerational transmission of inequality and segregation as its spatial expression is an important topic in research on inequality worldwide. In major cities, real estate prices have risen to levels that make it difficult for younger generations to enter the city housing market (both in cities where owner-occupation as well as rental housing is dominating). While modern, well-paid jobs are primarily located in the same economic cores (major cities), wealth reached during the lifetime is often not enough to make the steps in the real estate market, far from choosing a place of residence according to preferences. Increasingly, the residential trajectory of young households in cities depends on the extent to which they have been able to inherit the wealth of their parents – parents' support during their education as well as direct financial capital in the real estate market. The project analyses the differences between young households depending on their parents' relationship to the real estate market in different regions of Estonia, the interethnic differences in the intergenerational transmission of wealth and segregation, and compares the same phenomenon in Tallinn and Stockholm. Data analysis is based on the linked individual-level census databases (1989, 2000, 2011) and recent registry data, and longitudinal registry data in other countries.

Supervisors: Enn Karro, Andres Marandi, Alar Rosentau

Current PhD project aims at developing new GIS tools and methodologies for modelling the groundwater resources and its vulnerability under the present and future climates in Estonia. The increasing amount of data shared internationally creats a need to transfer them into complex and regionally variable hydrogeologal context. Multicriteria approach, where dynamic modelling is integrated with GIS analysis provides the tools for holistic treatment of data used for hydrogeological studies. The proposed PhD project involves four major research topics. Firstly, the development of DRASTIC groundwater vulnerability mapping method. Primary due to the thick Quaternary cover, the DRASTIC method requires significant modification to be successfully applied for the confined bedrock aquifers in Estonia. Secondly, the developing of grid-based calculation tools to estimate the groundwater resources, fluxes and quality in 3D structure of European aquifers. The topic is connected with the Horizon 2020 project GeoERA. Thirdly, the development of tools for the GIS raster analyses to predict the possible consequences of climate change to the hydrogeological system. Within the topic grid-based GIS methods will be updated by calibrated and verified groundwater modelling results. Finally, the developed scripts and GIS toolboxes will be published.

Supervisor: Kalle Kirsimäe

This project aims at evaluation the geochemistry of authigenic Mn-Fe and P in sediments across the times of the major environmental disturbances in Earth history in order to constrain the mechanisms driving the phosphate dynamics. The specific objective of the study is to reveal at high resolution the dynamics and secular changes in nutrients availability and their carrying phases in marine sediments at the Archean-Proterozoic transition during the Great Oxygenation Event and in its aftermath.

Supervisors: Päärn Paiste, Kärt Paiste

Sulfide minerals, particularly pyrite (FeS2), are ubiquitous in various geological environments and provide an accessible proxy for investigating geological processes.Sulfide trace element and sulfur isotopic composition are powerful tools in investigations of various geochemical processes such as sulfide-ore formation and finding evidence for life on early Earth.The objective of this project is to develop and validate a procedure for in situ sulfurt isotope and trace element analyses of different sulfide minerals (mainly pyrite), and to demonstrate its applicability in relevant geological situations.The project will also include development and validation of matrix-matching sulfide standards.

Supervisors: Vambola Kisand, Angela Ivask, Rainer Pärna

The aim of the present PhD project is to study further antimicrobial coatings based on ZnO/Ag nanocomposite particles. Such coatings are planned to be applied in hospitals as well as in public spaces on high-touch surfaces, which would also be exposed to UV-A or solar radiation at some stage. Compared to the products already on the market, the advantage of our solution is three simultaneously acting antimicrobial functionalities: zinc ions, silver ions and reactive oxygen radicals. In addition, on such coatings organic residues from dead microbes are photocatalytically decomposed. During this project, we plan to study the ZnO/Ag nanocomposite material fixation on surfaces, photocatalytic degradation of different fixation matrices and to investigate wearability, stability and antimicrobial efficacy of the ZnO/Ag based surfaces in real-life conditions.

In statistical modelling it is essential to find a model that is in some sense most suitable for describing reality. Most model selection methods are related to some information criterion. Information criteria such as Akaike or Bayes select the best model among the candidates but they cannot measure whether the assigned model class is suitable. Maximum spacing method (MSP) which is originally a parameter estimation method for continuous distributions, enables to decide about suitability of the assigned model class for given data. The MSP method can be studied for different information measures which is essential in model validation context. One aim of the project is to study properties of MSP estimators for different information measures in the case of multivariate independent and identically distributed (iid) observations.  The second goal is to study extensions of MSP method to more general models than the iid-case, and explore its possibilities as a model validation tool for model classes of different complexity.

Supervisors: Angela Ivask, Maia Kivisaar, Vambola Kisand

The goal of this project is to reveal the most potent targets during bacterial biofilm formation that could be attacked by antimicrobial coatings. The study will focus on several mechanisms playing a crucial role in biofilm formation – amount of extracellular DNA, extracellular polymeric substances including amyloids that all play a crucial role in biofilm formation and stabilization. Additional focus will be on quorum sensing, a central process initiating biofilm formation. In the first stages of the project, antimicrobial surfaces that have been developed in earlier studies and consist of nanosize ZnO and silver composites will be studied for their anti-biofilm properties against Staphylococcus aureus, Staphylococcus, epidermidis, Pseudomonas aeruginosa, Escherichia coli and other biofilms. Based on earlier indications, we hypothesize that these antimicrobial coatings inhibit the formation of bacterial extracellular amyloids and decrease the amount of extracellular DNA but here, the molecular mechanisms behind this process will be studied. Methodologies that will be used consist of microscopy, standard staining and molecular methods. This work will be conducted in collaboration with Prof. Pu-Chun Ke at Zhongshan Hospital, Fudan University in China. Prof. Ke is an expert in amyloidosis inhibition and protein fouling in nanomedicine.
The effects and mechanism of action of the studied antimicrobial coatings on biofilms will be compared with earlier proven anti-biofilm compounds. The final goal of the project is to propose effective and mechanistically proven components for anti-biofilm coatings. Thus, after the selection of the most potent compounds, their applicability in coatings will be evaluated in close collaboration with Institute of Physics of the University of Tartu.

Supervisors: Jaak Truu, Marika Truu

Worldwide marine coastal areas and sediments are exposed to the oil spills occurring as a result of accidents or illegal practices. Among the many environmental challenges related to the accidental release of oil to sea, the management and remediation of contaminated marine sediments is perhaps the most difficult one.  Bioelectrochemical systems (BES) have recently emerged, in which an electrical current serves as either electron donor or acceptor for organic pollutant bioremediation. Field experiments about bioremediation of stranded oil will be initiated and monitored for assessment natural attenuation as a remediation approach. Microbial community structure in lab-scale BES will be studied using amplicon-based, shotgun metagenomic and metatranscriptomic approaches in order to model the BES performance and increase the system treatment efficiency. Based on obtained information better predictions for intrinsic biodegradation capacity of oil fractions in sea water, sediments and coastal material can be obtained for the Baltic Sea and improved oil remediation technologies could be developed.

Supervisors: Marika Truu, Jaak Truu, Ivika Ostonen-Märtin

Understanding and predicting how climate and land-use change affect the structure and functioning of ecosystems is a key challenge of the 21st century. Qualitative or quantitative changes in rhizodeposition (root litter, metabolites exudation) will cause a shift in the structure and function of soil and rhizosphere microbiome responsible for stabilization of SOM. The soil and plant microbiomes mediating nutrient foraging and affecting the adaptation capacity of plants and ecosystems in global change. The overall goal of this doctoral project is to advance our understanding of the effects of higher temperature and changed water conditions (drainage, irrigation, higher relative air humidity) on the structural and functional adaptation of soil and plant microbiomes in subarctic grasslands and boreal forest ecosystems, both on short and long time scales by using the geothermal soil temperature gradients (FORHOT), and different water level gradients (drainage gradients in peatland forests and FAHM facility) in Estonia.
The overall approach of the project will be assessing interactions between plant root structural, physiological and biochemical traits, as well as the plant and soil microbiome components (bacteria, archaea and fungi) in relation to increasing soil temperature and changes in soil water regime. The functional potential of soil and plant-related microbial communities and the expression of the genes at different environmental conditions will be analysed by metagenomes and metatranscriptomes, phylogenetic diversity of microbial communities and plant microbiomes will be assessed using 16S rDNA/ITS amplicon-based high-throughput sequencing; abundances of soil bacteria and archaea will be assessed using16S rRNA gene abundance by quantitative PCR method.

Supervisor: Uko Maran

Solubility in intestinal fluids is a key biopharmaceutical property that determines the absorption of orally administered drugs in the human gastrointestinal tract. Effective development of oral drugs requires that potential problems with intestinal absorption due to poor or variable solubility be identified at an early stage of drug development. Therefore, computational modeling of the solubility of drug substances in human intestinal fluids is of great importance. This thesis focuses on the systematic analysis of intestinal solubility data for drug substances to understand the mechanisms of molecular interactions during intestinal solubility and to detect patterns in chemical data. Emphasis is on cheminformatics, machine learning (ML), and artificial intelligence (AI) based methods that help develop descriptive and predictive quantitative structure-activity relationship (QSAR) models for evaluating solubility of drug substances in intestinal fluids.

Supervisors: Uko Maran, Sulev Sild, Jaan Leis

Ionic liquids have gained strong interest in chemical industry and research to replace traditional volatile organic solvents as safer or better alternatives in synthesis and catalysis, electro chemistry, separation and extraction chemistry, analytical chemistry, biological and material engineering applications, including electrospinning of polymer mats. The proposed doctoral research is about effective modelling of ionic liquids in very diverse datasets with focus on multicomponent systems. Particularly it is dedicated on the modelling of ionic liquids with the aid in understanding physicochemical interaction mechanism when chemicals enter the ionic liquids and how to model and predict such interactions. For this the in silico approaches, such as chemo-informatics analysis, including machine learning and artificial intelligence, will be used. Such methods allow bridging the lack of knowledge about chemicals when little or no experimental data are available. The qualitative and quantitative structure-property relationships (QSPR) methodology development faces new challenges that are related with the increasing diversity of chemical compounds measured and the new ones that need predictions for decision support. The methodology also allows fine tuning the properties of ionic liquids for specific applications by introducing small changes to the chemical structure of its components. In summary, this doctoral thesis will investigate complex relationships between the properties of ionic liquids and their chemical structure, eventually leading to the development of novel approaches for designing new ionic liquids.

Supervisors: Luca Marzola, Martti Raidal

The nature, origin and properties of dark matter (DM) of the Universe have remained obscured and all attempts to detect the DM interactions beyond gravity have failed. This result naturally leads to the paradigm that the DM is not a manifestation of some particle physics concept but of extended gravity. The aim of this project is to perform a comprehensive study of this possibility and, using multi-messenger approach, to work out its possible experimental and cosmological signatures. We are going to study macroscopic DM candidates, new spin-0, 3/2, 2 particles of extended gravity, and their connection to inflation and other phenomena in particle physics and cosmology.

Supervisors: Madis Kiisk, Andrea Giammanco

Cosmic ray tomography is a rapidly growing research area. It relies mainly on muon particles, which hit the earth surface at a rate of about 1 muon/m2/min. Cosmic ray tomography uses the scattering of muons in the Coulomb field of atomic nuclei. Scattering deflection is dependent on the charge of the nucleus (Z) as a function of Z2. Therefore, the method has been so far applied for detection of special nuclear materials in the security and nuclear industry. Though detection of low-Z materials has been proposed by several research groups, no clear evidences exist that detection of light materials on small volumes and within a minute measurement time is feasible. Our research group has recently developed a new tomographic reconstruction method that has been proved to be able to detect low-Z materials, such as explosives, narcotics, other light materials. The main aspects for achieving such a sensitivity is the ability to analyse and discretize the usable cosmic ray flux on the sea level, have sufficient detector spatial resolution as well applying appropriate reconstruction methods specifically adapted to detection of low-Z materials.
This doctoral project is going to study above-mentioned aspects and will focus on some of the crucial parameters in efficient object reconstruction procedure, such as scattering centres (vertices), combination of independent interaction effects in the object reconstruction procedure and momentum estimation accuracy of the leptons.

Supervisors: Jüri Raud, Lauri Aarik

Low-temperature plasma is an efficient source of highly reactive plasma species which have used in various applications. Usually, the plasma is used as a “black box“ having certain effect on the plasma treatment target and only external plasma parameters (e.g. plasma source power, composition of gases, pressure) are related with the desired effect. Proposed study aims to develop and validate a plasma-chemical model which enables to determine RONS concentration and composition in the reactor and helps to find optimal plasma source working parameters in respect to application. The study investigates the RONS influence in two plasma sources used for specific applications: plasma medicine (RONS influence on cancer cells) and plasma enhanced atomic layer deposition. In both applications a radio frequency discharge in Ar/He feed gas at pressure range 1-760 Torr will be used as plasma source. N2 and/or O2 will be added to the feed gas in order to produce RONS. As a first step, methodology for RONS composition and concentration detection will be investigated. Second step will be the development of plasma-chemical model which allows to calculate RONS concentration and distribution near the surface of treated material. The results from RONS measurements as well as from analysis of plasma treated samples will be used for model validation. In the final part of the work the plasma-chemical model will be used in order to find optimal experimental condition beneficial for specific application.

Supervisors: Tomi Koivisto, Stefan Groote, Luca Marzola

The unification is approached from an algebraic perspective. We will take advantage of the facts that the Standard Model can be elegantly obtained from the properties of the four division algebras, and that the previously considered unifications of the Standard Model such as SU(5) can be represented with differential forms that also describe the elements of spacetime geometry. The paramount role of complex numbers in quantum mechanics and in gauge theories is well appreciated, and there is compellingly accumulating evidence for the structure of the division algebras underlying the totality of the physical reality. An ultimate aim is to arrive at a new understanding of space, time and matter as aspects of a single unified structure, and to reduce the number of free parameters in the standard model. A particular task to be undertaken is the description of the Higgs mechanism in the electroweak sector and a spontaneous symmetry breaking giving rise to spacetime as a single phenomenon. During the project, we will also investigate the implications of the new spacetime theory we are constructing. We plan to address some issues related to dark matter, to the origin of the universe, and black holes.

Supervisors: V. Kiisk, R. Jaaniso

Novel directions in the development of semiconductor gas sensors involve application of 2D (incl. graphene-based) materials and/or optical response/stimulus. This allows implementing miniature multiresponse sensor elements and their arrays, which are inexpensive, sensitive and energy efficient. The main goal of the project is to introduce contemporary machine learning software (such as TensorFlow) and develop optimized protocols of data processing that would utilize both short-term and long-term trends of the sensor array responses to counteract drift, decrease response time, and improve selectivity towards several hazardous gases (NO2, O3, NH3, …).

Supervisors: Mihkel Pajusalu, Mart Noorma

Offline navigation is increasingly based on cameras, but the use of camera images to navigate in an unfamiliar environment still requires a lot of development work. The purpose of this thesis is to develop a fault-tolerant visual odometry system for localization and navigation on different celestial bodies and on different landscapes on Earth, based on multi-camera image streams. The aim of the development is to reach level 3 of autonomy, thus assisting the operator of the self-driving vehicle as much as possible.
Local visual odometry for safe maneuvering is already being developed in collaboration with Milrem Robotics in the framework of the project "Applied Research for the Development of a System for Sensors and Software Algorithms for Remote Control and Assistance in Security Features".
The project will complement existing methods, such as optical flow, neural network-based monocular depth perception, edge processing with algorithm optimization and acceleration. In addition, autonomous calibration of the cameras and reconstruction and simulation of the vehicle environment are being dealt with.

Supervisors: Heikki Junninen, Madis Kiisk, Urmas Hõrrak

Atmospheric ions play an important role in atmospheric clusters and new particle formation and cloud activation. Detailed understanding on formation of atmospheric ions is a necessity for understanding whole biosphere-atmosphere-climate interconnections. In this work a new instrumentation will be developed to detect ionization rate of air, physical and chemical properties of air ions and characterization of ionizing radiation in atmosphere. Ionizing radiation is originated from galactic cosmic rays and from soil radiation, through direct emission and also through radon gas outgassing. The new instrumentation is deployed to field station for atmospheric research where closure on ion sources and sinks will be estimated. Work will contain radiation modelling, development of new instruments, experimental work in laboratory and in field station, data collection and statistical data analysis of the collected data together with all relevant data measured in the field station. Depending on personal capabilities and interests of candidate the work can be shifted more towards modelling, instrument development or data analysis.

Supervisors: R. Jaaniso, M. Kodu

The aim of the project is to integrate the gas sensors based on functionalised single-layer graphene onto CMOS chips, micro-hotplates and -lightplates. Methods to combine multiple functionalisations on a single chip will be investigated. The project is supported by Graphene Flagship.

Supervisor: Marco Kirm

The goal of research is to develop functional nanomaterials for medical applications based on the synergistic effect of X-ray irradiation and other therapeutic and diagnostic approaches. These heavy nanoparticles, delivered into the tumor tissue, create UV scintillation photons under X-ray excitation for the increase of efficiency of killing cancer cells during radiotherapy treatment. Pr3+ doped with orthophosphates emitting in UV-C will be studied to understand the interaction of different rare earth dopants upon high energy excitation and provide a detailed clarification of energy relaxation pathways subsequent to the absorption of X-rays. Wise selection of the rare earth co-dopants (e.g. Nd3+, Gd3+, Er3+, or Yb3+) in LuPO4, LaPO4 and YPO4 hosts allows to create novel materials for multifunctional imaging, diagnosis and treatment applications in medicine.

Supervisors: Mihkel Pajusalu, Andris Slavinskis

In several scientific use cases, such as fly-bys of Small Solar System Bodies (SSSBs), on orbit inspection, or fast pace terrain traversal, measurements of a target can be made only once and in a very limited time window, while the data budget per target can also be highly constrained. An example of this is the ESA F-class mission Comet Interceptor, which will perform a single fly-by of a dynamically new object (a comet or possibly even an interstellar object) to gather data in order to constrain the conditions in the early solar system. In the case of fast traversals using planetary rovers, science output could be significantly increased if objects of interest could be automatically characterized while driving and a minimal data set could be saved for transmission, possibly after a long communication blackout. In the terrestrial case, this situation can happen during drone fly-bys or when objects of interest pass by fixed cameras.
The main focus of this thesis project will be analyzing small spacecraft fly-by missions of small solar system bodies that use imaging cameras to map targets and to constrain spacecraft trajectory and attitude, especially the case of the Comet Interceptor mission for which The Optical Periscopic Imager for Comets (OPIC) is being developed in Tartu Observatory. Due to heavily constrained mission architecture, OPIC requires development of automatic imaging, image processing, and image transmission strategy to maximize scientific output.
To achieve this goal, we will use image simulation using physics based rendering methods and feed these images through image and trajectory reconstruction frameworks and see how image pre-processing and capturing strategies affect the final result and which method will lead to the optimal balance between the scientific output and resources used.
We will also explore the use of these strategies for autonomous driving on Earth and on the Moon.

Supervisors: Randel Kreitsberg, Margit Heinlaan, Angela Ivask

Plastic pollution is recognized as one of today’s major environmental challenges yet substantial knowledge gaps prevent creation and implementation of effective policy responses. Effect data on microplastic (MPL) exist yet these are often for environmentally irrelevant concentrations and such data may lead to misleading conclusions about the risks posed to aquatic biota. According to 1199 scientific publications in German Litterbase database, 2249 marine species are affected by plastic pollution. MPL biological effects have been shown to be complex and mostly sublethal, induced by physical/chemical toxicity/effects, food limitation, and their combinations. The fate of MPL upon uptake in aquatic organisms is an open question. We aim to fill the gaps on MPL abundance in Estonian coastal sea area as well as oceanic environments and use these abundance data to perform environmentally relevant MPL effect analysis using zebrafish Danio rerio as model organism. To ensure successful fulfilling of the aims, we have united the best current know-how on MPL (toxicity) research in Estonia and planned the project as cooperation between UT, NICPB and EULS.

Supervisors: Toomas Tammaru, John Clarke

Despite the key significance of body size (BS) for the biology of any organism, our understanding of the evolution of BS is far from complete. The PhD project applied for will focus on research on the evolution of insect body size in a phylogenetic comparative framework. Taking advantage of the advances in phylogenetic reconstruction of the moth family Geometridae, we will perform a global analysis focusing on determinants of BS, and rates of BS evolution. Applying the most modern methods of phylogenetic analysis, the research will address the questions about long-term evolutionary trends in BS, and constraints on the adaptive evolution of BS. Such approaches are unprecedented in the context of evolutionary ecology of invertebrate animals and will thereby substantially advance our understanding of the evolution of BS size in general.

Supervisor: Asko Lõhmus

The project explores how extensive heart-rots develop in trees, and to what extent these serve a dynamic supply of key structures for large cavity-using animals in forest landscapes across Europe. It is known that, in old-growth forests, a significant part of trees can be hollow (with implications on carbon storage assessment); that large tree cavities (entrance diameter min. 10 cm) are among the most vulnerable microhabitats to production forestry; and that already formed cavity trees can be effectively retained at harvests. However, the ecological feedback links and functioning of the tree-hollow supply are not known sufficiently enough to address the current impoverishment of forest landscapes and to plan ahead for decades and centuries. Most previous studies on hollows have focused on small cavities and small cavity-nesters because these are more abundant and easier to study.
The current project comprises three parts: (1) the pools of large tree cavities and hollowing trees (prone to cavity development) will be surveyed across natural temperate forests in Europe using standard methods. The ultimate aims is to describe the (natural) reference level of large tree cavity supply and its main factors; (2) decadal dynamics of the large-cavity supply (including loss) will be directly assessed in a subset of plots, which have been studied for more than 10 years; (3) in a subset of plots and cavities the patterns of large-cavity use by animals will be described using camera-traps, installed in front of hole entrances to get year-round data on all visits divided into breeding, roosting, feeding and social interactions.

Supervisor: Raivo Kolde

The centralized collection of electronic healthcare data in Estonia creates a number of opportunities for developing applications that utilize the data for public health goals. One such goal is chronic disease management, as complications from such diseases put a huge burden on the healthcare system. Complications, however, can be prevented, by increasing patient engagement with their treatment plan and ensuring regular follow-up with the patients.
Electronic health data can be used to tackle both of these aspects. Prediction models can be built to identify patients with higher risk, to prioritize their follow-up. Patient engagement can be increased by visualizing the risks and the effect of behavioural choices in a more comprehensible and personalized manner. This can be achieved through creating personalized cohorts of similar patients and summarizing their medical histories as a prediction.
The goal of the PhD project is to explore the viability of such these ideas and implement prototype systems to test them in practice. The candidate is expected to have skills to work with healthcare data, build prediction models, create novel visualisations and, importantly, have interest in medical domain.

Supervisors: Kuldar Taveter, Ishaya Peni Gambo

Research on smart cities from the perspective of requirements engineering is currently unorganized and fragmented. Recent research papers on requirements engineering for smart cities advocate defining the requirements for a smart city using the bottom-up approach. With this kind of approach, the emphasis is on the importance of requirements for a system architecture that could scale up to systems of any size and accommodate applications of any kind. This effectively means that smart cities are defined by technological needs and possibilities rather than by functional and emotional needs of citizens met with acceptable quality. Because of the considerations mentioned, this proposal is concerned with how to continuously elicit requirements for socio-technical systems of smart cities. The project is dealing with incorporating emotions into the applications of smart cities and describing emotions in the context of smart
cities to develop privacy-aware affective systems. This PhD project is particularly concerned with how a smart city could improve the opportunities for older adults in a community. In this context, the project will explore how the satisfaction levels with smart city applications could be increased by studying how people react and behave under different circumstances and creating a method for grouping the reactions and behaviours of people within a community. This kind of grouping enables to generate a large dataset for analysis of people’s behaviours and predictive privacy modelling. The dataset will serve as a basis for designing and implementing a learning algorithm that determines if the data is safe or unsafe to share. The results of the PhD project will be validated in an EU project dealing with older adults and in other projects.

Supervisor: Jaan Aru

How does the human mind come to creative ideas? How to implement these algorithms of the human mind in artificial intelligence (AI)? The main goals of the present proposal are to experimentally dissect the process of creative problem solving and to take the first steps in developing AI systems that can do that. To that end we will study problem solving with virtual reality (VR) and build new AI algorithms based on these experimental insights.
VR allows us to create novel and otherwise impossible tasks that require innovative solutions. In particular, we will design a new “escape room” type of a problem in VR that requires finding a non-orthodox solution. With our novel approach we can study which processes are required for creativity and which factors negatively affect creativity and problem solving.
These experimental insights will allow us to create a computational model of the creativity process. We will then implement this model on top of modern AI algorithms. Our hypothesis is that AI agents equipped with a human-like creativity process can solve tasks that AI agents without such module cannot. By studying and answering these questions our research will significantly contribute to understanding the processes underlying human intelligence and take important steps toward implementing these algorithms in AI.

Supervisor: Jaan Aru

Computer vision has been extremely successful in solving object detection and classification tasks. Nevertheless, researchers have realized that these processes implemented in today’s computer vision systems only capture one aspect of human visual competences.
In addition to fast and automatic processes, well studied in today’s artificial intelligence (AI) systems, human thinking has another part, a slower and deliberate process, that is crucial in situations that are ambiguous. How can we study these latter processes in a manner that would enable us to implement these processes in AI? In particular, while computer vision benchmarks have revolutionized modern AI algorithms, a benchmark is lacking that would engage such deliberate processing.
In this project we will create a novel computer vision benchmark about images that require deliberate processing in humans. We will systematically study and collect data about human behavior on this benchmark. Moreover, we will create the first AI algorithms that can solve this benchmark. Together, this work will provide an unprecedented but much needed window into the cognitive flexibility that defines us as a species.

Supervisor: Riccardo Tommasini

The main goal of this doctoral work is to identify what are the open challenges that prevent BigSPEs to substitute full-fledged data management systems. To this extent, the candidate should survey the state of the art of stream processing, big data, and data management systems. After the literature review, the candidate will proceed to investigate the most promising open challenges. Examples of relevant research directions include but are not limited to (i) expressive query processing; (ii) integrity constraints, and (iii) transaction management on top of Stream Processors.

Supervisor: Huber Flores

Plastic pollution is a growing worldwide concern, affecting the health of marine ecosystems, human health, climate change, and weather patterns. Currently, obtaining accurate information about the extent of plastic pollutants is difficult as existing measurement solutions are highly laborious or costly, limited to specific pollutants, and have limited spatial and temporal resolution. In this work, we envision a large-scale autonomous marine pollution monitoring solution that uses coordinated swarms of autonomous underwater vehicles (AUV)s to monitor extent and characteristics of plastic
pollutants. Our envision solution combines optimal sensing, data science and autonomous robotics to identify and classify plastic pollutants in oceans – or other aquatic environments such as lakes or rivers.

Supervisor: Marlon Dumas

Business processes are the operational backbone of modern organizations. Their continuous improvement is crucial to achieve organizational objectives, be it with respect to efficiency, quality, compliance or agility. Accordingly, a common task for business process analysts is to discover and assess business process improvement opportunities.
Current approaches to discover business process improvement opportunities are manual and expert-driven. In these approaches, execution data are used to assess opportunities derived from experience and intuition rather than to discover these opportunities in the first place.
This PhD project will develop methods to automatically discover business process improvement opportunities from execution data. Specifically, the project will develop methods to discover opportunities to alter the resource allocation rules and the decision rules used in a business process, in order to improve its performance with respect to one or more performance measures. To achieve this, the project will apply, adapt, and compare existing techniques from the field of rule mining and cost-sensitive learning.
The resulting techniques will be evaluated using real-life business process execution data. The techniques will also be integrated into a toolset for automated process improvement, which will be evaluated via case studies in cooperation with industry partners.

Supervisors: Joris Vermeesch (KU Leuveni Ülikool) ja Ants Kurg (Tartu Ülikool, TÜMRI);

The goal of the project is to develop novel single-cell technologies to understand mechanistic origin of aneuploidy and mosaicism during preimplantation embryo development. This project will specifically aim: (1) To improve single-cell haplotyping and genotyping technologies and to enable highly accurate single-cell sequencing for concomitant transcriptome and genome analysis of single cells, using appropriate cell-lines and (2) To unravel the incidence and mechanisms underlying chromosomal rearrangement during early embryo development, by exploiting the developed single-cell technologies in bovine IVF embryos.

The PhD student will mostly be hosted by the KU Leuven, and will be seconded for a shorter period to the University of Tartu. The position also includes a training period in European company focusing on reproductive health.

Supervisors: Ants Kurg/Tõnis Org (Tartu Ülikool, TÜMRI) ja Juha Kere (Karolinska Instituut).

The goal of the project is to study the effects of in vitro environment on epigenetic regulation of oocyte maturation and embryonic development by comparing in vivo and in vitro generated bovine embryos. This will be achieved by comparing the in vivo conceived embryos and in vitro embryos produced by ovum pick-up (OPU) with ovarian stimulation (OPU-IVF). The more specific focus is to study the effects of in vitro manipulations (embryo in vitro culturing at different conditions and embryo vitrification/thawing) on the epigenetic regulation of embryonal cells’ development.

This will be done through ChIP-seq technology from single or low number of embryonic cells allowing to investigate histone modifications, associated with different genomic features, such as promoters of active genes, active enhancers or silenced chromatin. In addition, ATAC-seq technology will be used to study opened/closed chromatin. Critical regions identified by above methods will be tested with other methodologies.

The PhD student will be mostly hosted by the University of Tartu, and will be seconded for a shorter period to the Karolinska Institute. The position also includes a training period in European company focusing on reproductive health.

Supervisors: Triin Laisk (TÜ), Reedik Mägi (TÜ), Terhi Piltonen (Oulu Ü)

The goal of the project is to (1) identify novel genetic causes of reproductive health related phenotypes (such as polycystic ovary syndrome, pregnancy complications, reproductive aging and others) with the possibility to take promising findings further for functional validation; (2) develop polygenic risk scores (PRS) for reproductive traits to improve the predictive power of PRSs for stratification of diseases, their subtypes and comorbidities while accounting for possible population structure. This will be done using large-scale data with rich health information at the Estonian Biobank. During the studies, the research fellow will also spend time at Medical Research Center (MRC) Oulu. MRC Oulu located at the Kontinkangas campus, that has wide range of infrastructures to conduct high-quality research, including facilities of Faculty of Medicine, Biochemistry and Molecular Medicine, the Biocenter core facilities, Northern Finland Birth Cohort, the Northern Finland Biobank Borealis and Oulu University Hospital.

Supervisors: Andres Salumets/Tõnis Org (Tartu Ülikool, TÜMRI/Genoomika instituut) ja Taneli Raivio (Helsinki Ülikool)

The aim of the project is to elucidate the molecular mechanisms of female infertility and endocrinological diseases such as hypogonadotropic hypogonadism, primary amenorrhea and premature ovarian insufficiency. Candidate genes from previous genome-wide association studies or based on the previous clinical expertise of the partners will be selected for studies. The functional characterization of the candidate genes will be performed using human induced pluripotent or embryonic stem cell lines, harbouring the disease-causing gene mutations, and using the stem-cell differentiation protocols supporting the differentiation of hormone-producing neuronal cells and follicular granulosa cells. Studies on the cellular differentiation and functional characterization of the stem-cell derived neurohormonal and follicular cells provide information on the pathology mechanisms of female infertility-associated endocrinological diseases, which is of great importance in understanding the causes and development of these diseases.