ERC FUNDED PROJECTS

This interdisciplinary proposal has the objective to greatly enhance our understanding of fundamental biomineralization processes involved in the formation of calcium carbonates by marine organisms, such as corals, foraminifera and bivalves, in order to better understand vital effects. This is essential to the application of these carbonates as proxies for global (paleo-) environmental change. The core of the proposal is an experimental capability that I have pioneered during 2008: Dynamic stable isotopic labeling during formation of carbonate skeletons, tests, and shells, combined with NanoSIMS imaging. The NanoSIMS ion microprobe is a state-of-the-art analytical technology that allows precise elemental and isotopic imaging with a spatial resolution of ~100 nanometers. NanoSIMS imaging of the isotopic label(s) in the resulting biocarbonates and in associated cell-structures will be used to uncover cellular-level transport processes, timescales of formation of different biocarbonate components, as well as trace-elemental and isotopic fractionations. This will uncover the origin of vital effects. With this proposal, I establish a new scientific frontier and guarantee European leadership. The technical and scientific developments resulting from this work are broadly applicable and will radically change scientific ideas about marine carbonate biomineralization and compositional vital effects.

2 182 000 €

Start date: 2010-07-01, End date: 2015-06-30

This project is directed towards development of new laser diagnostic techniques and a deepened physical understanding of more established techniques, aiming at new insights in phenomena related to combustion processes. These non-intrusive techniques with high resolution in space and time, will be used for measurements of key parameters, species concentrations and temperatures. The techniques to be used are; Non-linear optical techniques, mainly Polarization spectroscopy, PS. PS will mainly be developed for sensitive detection with high spatial resolution of "new" species in the IR region, e.g. individual hydrocarbons, toxic species as well as alkali metal compounds. Multiplex measurements of these species and temperature will be developed as well as 2D visualization. Quantitative measurements with high precision and accuracy; Laser induced fluorescence and Rayleigh/Raman scattering will be developed for quantitative measurements of species concentration and 2D temperatures. Also a new technique will be developed for single ended experiments based on picosecond LIDAR. Advanced imaging techniques; New high speed (10-100 kHz) visualization techniques as well as 3D and even 4D visualization will be developed. In order to properly visualize dense sprays we will develop Ballistic Imaging as well as a new technique based on structured illumination of the area of interest for suppression of multiple scattering which normally cause blurring effects. All techniques developed above will be used for key studies of phenomena related to various combustion phenomena; turbulent combustion, multiphase conversion processes, e.g. spray combustion and gasification/pyrolysis of solid bio fuels. The techniques will also be applied for development and physical understanding of how combustion could be influenced by plasma/electrical assistance. Finally, the techniques will be prepared for applications in industrial combustion apparatus, e.g. furnaces, gasturbines and IC engines

2 466 000 €

Start date: 2010-02-01, End date: 2015-01-31

Cells use circuits of interacting proteins to respond to their environment. In the past decades, molecular biology has provided detailed knowledge on the proteins in these circuits and their interactions. To fully understand circuit function requires, in addition to molecular knowledge, new concepts that explain how multiple components work together to perform systems level functions. Our lab has been a leader in defining such concepts, based on combined experimental and theoretical study of well characterized circuits in bacteria and human cells. In this proposal we aim to find novel principles on how circuits resist fluctuations and errors, and how they can be controlled by drugs: (1) Why do key regulatory systems use bifunctional enzymes that catalyze antagonistic reactions (e.g. both kinase and phosphatase)? We will test the role of bifunctional enzymes in making circuits robust to variations in protein levels. (2) Why are some genes regulated by a repressor and others by an activator? We will test this in the context of reduction of errors in transcription control. (3) Are there principles that describe how drugs combine to affect protein dynamics in human cells? We will use a novel dynamic proteomics approach developed in our lab to explore how protein dynamics can be controlled by drug combinations. This research will define principles that unite our understanding of seemingly distinct biological systems, and explain their particular design in terms of systems-level functions. This understanding will help form the basis for a future medicine that rationally controls the state of the cell based on a detailed blueprint of their circuit design, and quantitative principles for the effects of drugs on this circuitry.

2 261 440 €

Start date: 2010-03-01, End date: 2015-02-28

Allostery is a fundamental concept Nature uses to regulate the affinity of a certain substrate to an active site of a protein by binding a ligand to a distant allosteric site. We will design experimental tools to gain an atomistic understanding of the conformational transitions that give rise to allostery. We will approach the problem from two distinctively different directions. First, we will initiate conformational transitions of proteins that per se are not photoswitchable, by cross-linking two sites of an allosteric protein with a photo-switchable azobenzene-moiety to initiate a conformational transition similar to ligand binding. We will use ultrafast infrared spectroscopy to time-resolve the conformational transition. Second, we will experimentally verify a frequently expressed hypothesis that allosteric and active site communicate by exchange of vibrational energy. To that end, we will design a versatile approach that allows us to locally deposit vibrational energy at essentially any site in a protein (e.g. through pumping of an optical chromophore that undergoes ultrafast internal conversion), and to detect its appearance at any other site by using vibrational transitions as local thermometers. Thereby, we will map out a network of connectivity in a given protein. Both approaches will applied both to one and the same protein family. One concrete example are PDZ domains, which are among the smallest allosteric proteins, and for which the connection between allostery and vibrational energy flow has been made explicit, based on computer simulations. We will eventually test this hypothesis experimentally, and provide the foundation for a description of allostery that is on an equal footing as our current understanding of protein folding.

2 400 000 €

Start date: 2010-02-01, End date: 2015-07-31

The advent of the Internet and the increased power of modern day computing have dramatically changed the economic landscape. Billions of dollars worth of goods are being auctioned among geographically dispersed buyers; online brokerages are used to find jobs, trade stocks, make travel arrangements, etc. The architecture of these online (trading) platforms is typically rooted in their pre-Internet counterparts, and advances in the theory of market design combined with increased computing capabilities prompt a careful re-evaluation. This proposal concerns the creation of novel, more flexible institutions using an approach that combines theory, laboratory experiments, and practical policy. The first project enhances our understanding of newly designed package auctions by developing equilibrium models of competitive bidding and measuring the efficacy of alternative formats in controlled experiments. The next project studies novel market forms that allow for all-or-nothing trades to alleviate inefficiencies and enhance dynamic stability when complementarities exist. The third project concerns the design of market regulation and procurement contests to create better incentives for research and development. The fourth project addresses information aggregation properties of alternative voting institutions, suggesting improvements for referenda and jury/committee voting. The Internet has also dramatically altered the nature of social interactions. Emerging institutions such as online social networking tools, rating systems, and web-community Q&A services reduce social distances and catalyze opportunities for social learning. The final project focuses on social learning in a variety of settings and on the impact of social networks on behavior. Combined these projects generate insights that apply to a broad array of social and economic environments and that will guide practitioners to the use of better designed institutions.

1 797 525 €

Start date: 2010-01-01, End date: 2015-12-31

Nanotechnology research has been directed mostly to the design and synthesis of (a) materials with passive nanostructures (e.g. coatings, nanoparticles of organics, metals and ceramics) and (b) active devices with nanostructured materials (e.g. transistors, amplifiers, sensors, actuators etc). Little is known, however, about how well the unique properties of nanostructured materials are reproduced during their large scale synthesis, and how such manufacturing can be designed and carried out. A key goal here is to fundamentally understand synthesis of surface-functionalized, nanostructured, multicomponent particles by flame aerosol reactors (a proven scalable technology for simple ceramic oxide nanopowders). That way technology for making such sophisticated materials would be developed systematically for their efficient manufacture so that active devices containing them can be made economically. Our focus is on understanding aerosol formation of layered solid or fractal-like nanostructures by developing quantitative process models and systematic comparison to experimental data. This understanding will be used to guide synthesis of challenging nanoparticle compositions and process scale-up with close attention to safe product handling and health effects. The ultimate goal of this research is to address the next frontier of this field, namely the assembling of high performance active devices made with such functionalized or layered nanoparticles. Here these devices include but not limited to (a) actuators containing layered single superparamagnetic nanoparticles and (b) ultraselective and highly sensitive sensors made with highly conductive but disperse nanoelectrode layers for detection of trace organic vapors in the human breath for early diagnosis of serious illnesses.

2 500 000 €

Start date: 2010-05-01, End date: 2015-04-30

Sometimes in the sciences there are different yet complementary descriptions for the same object. This extends to the particle-wave duality of quantum mechanics; one mathematical analog of this duality is the Fourier transform. Questions that are difficult when formulated in one language of science may become simple when interpreted in another. The Langlands conjecture posits the existence of a correspondence between problems in arithmetic and in Representation Theory. The Langlands conjecture has only been proven for a limited number of cases, but even this has solved problems such as the famous Fermat conjecture. The aim of this project is to continue study of the "classical" aspects of the Langlands conjecture and to extend the conjecture to the quantum geometric Langlands correspondence, higher-dimensional fields, Kac-Moody groups (with D.Gaitsgory: quantum Langlands correspondence; D.Gaitsgory and E. Hrushevsi: groups over higher-dimensional fields; A. Braverman: Kac-Moody groups; R. Bezrukavnikov, S.Debacker, Y.Varshavsky: classical aspects of the correspondence; A. Berenstein: geometric crystals and crystal bases). The quantum case is much more symmetric than the classical case and can lead in the limit q->0 to new insights into the classical case. The quantum case is also related to the multiple Dirichlet series. New results in the quantum case would lead to progress in understanding important Number Theoretic questions. Extending the Langlands correspondence to groups over higher-dimensional fields could substantially enlarge its applicability. Studying Kac-Moody groups would provide tools for the new important class of L-functions. This progress could lead to a proof of the existence of the analytic continuation of classical L-functions. The geometric Langlands correspondence is closely related to T-symmetry in 4-dimensional gauge theory and the understanding of this relation is important for both Mathematics and Physics.

1 277 060 €

Start date: 2010-01-01, End date: 2014-12-31

In recent years, as first established in some 6 papers in Science and Nature from the PI s group, a new paradigm has emerged. This reveals the remarkable and unsuspected - role of hydration layers in modulating frictional forces between sliding surfaces or molecular layers in aqueous media, termed hydration lubrication, in which the lubricating mode is completely different from the classic one of oils or surfactants. In this project we address the substantial challenges that have now arisen: what are the underlying mechanisms controlling this effect? what are the potential breakthroughs that it may lead to? We will answer these questions through several interrelated objectives designed to address both fundamental aspects, as well as limits of applicability. We will use surface force balance (SFB) experiments, for which we will develop new methodologies, to characterize normal and frictional forces between atomically smooth surfaces where the nature of the surfaces (hydrophilic, hydrophobic, metallic, polymeric), as well as their electric potential, may be independently varied. We will examine mono- and multivalent ions to establish the role of relaxation rates and hydration energies in controlling the hydration lubrication, will probe hydration interactions at both hydrophobic/hydrophilic surfaces and will monitor slip of hydrated ions past surfaces. We will also characterize the hydration lubrication properties of a wide range of novel surface systems, including surfactants, liposomes, polymer brushes and, importantly, liposomes, using also synchrotron X-ray reflectometry for structural information. Attainment of these objectives should lead to conceptual breakthroughs both in our understanding of this new paradigm, and for its practical implications.

2 304 180 €

Start date: 2010-05-01, End date: 2015-04-30

Metabolic engineering is the development of new cell factories or improving existing ones, and it is the enabling science that allows for sustainable production of fuels and chemicals through biotechnology. With the development in genomics and functional genomics, it has become interesting to evaluate how advanced high-throughput experimental techniques (transcriptome, proteome, metabolome and fluxome) can be applied for improving the process of metabolic engineering. These techniques have mainly found applications in life sciences and studies of human health, and it is necessary to develop novel bioinformatics techniques and modelling concepts before they can provide physiological information that can be used to guide metabolic engineering strategies. In particular it is challenging how these techniques can be used to advance the use of mathematical modelling for description of the operation of complex metabolic networks. The availability of robust mathematical models will allow a wider use of mathematical models to drive metabolic engineering, in analogy with other fields of engineering where mathematical modelling is central in the design phase. In this project the advancement of novel concepts, models and technologies for enhancing metabolic engineering will be done in connection with the development of novel cell factories for high-level production of different classes of products. The chemicals considered will involve both commodity type chemicals like 3-hydroxypropionic acid and malic acid, that can be used for sustainable production of polymers, an industrial enzyme and pharmaceutical proteins like human insulin.

2 499 590 €

Start date: 2010-01-01, End date: 2014-12-31

Over the last decade epigenetic gene regulation has become a major focus of scientific research as it was shown to play an important role in normal plant and animal development, but also in the ontogeny of human disease. A role of epigenetic processes in evolution, however, has found little general support to date. The goal of this project is to understand the complex interplay of epigenetic mechanisms in plant development and evolution. Many of the approaches we use rely on the recent advances in sequencing technologies, which allow the analysis of molecular characters at an unprecedented level and speed. To achieve our goal, we will focus on two epigenetic paradigms. In Program A, we will focus on dissecting the mechanisms of genomic imprinting at the MEDEA (MEA) locus in Arabidopsis, which we will investigate using genetic, molecular, and innovative biochemical approaches to gain a comprehensive picture of the complex interplay of various epigenetic pathways. In program B, we will analyze the role of epigenetic change in adaptation and evolution using (i) an experimental selection approach in Arabidopsis, where genome-wide analyses of epigenetic modifications have become possible, and (ii) a stable, heritable, epigenetic change occurring in Mimulus populations. In this system, an epigenetic switch of the pollinator syndrome leads to reproductive isolation and, therefore, has an effect on population structure and thus the evolutionary trajectory. These experimental systems each offer unique opportunities to shed light onto the underlying mechanisms controlling epigenetic states. In combination with the new methodologies used, these analyses promise to provide step change in our understanding of epigenetic processes at the level of genes, organisms, and populations.

2 496 641 €

Start date: 2010-04-01, End date: 2015-12-31

Small RNA-mediated regulation of gene expression is a recent addition to fundamental gene regulatory mechanisms that directly or indirectly affect possibly every gene of a eukaryotic genome. The predominant sources of small RNA in somatic tissues are microRNA genes that encode short dsRNA hairpins of evolutionary conserved sequence. Disorders of metabolism, such as obesity and type 2 diabetes are poorly understood at a molecular level. In this application we propose to explore if miRNA regulatory networks play a role in these diseases. We will employ state of the art methods for identification of small RNAs and their regulated targets and use biochemical, cell and animal model systems to study the detailed molecular mechanisms of metabolic gene regulation by miRNAs. In addition, we will investigate the underlying principles of how RNAs are taken up by cells and develop methods that will improve delivery of miRNA mimetics or inhibitors through cell-specific uptake. The specific aims of this study are: Aim 1: To define the small regulatory miRNA content of liver, muscle and adipose tissue that are associated with abnormal glucose and lipid homeostasis and to dissect the underlying molecular pathways that govern their expression. Aim 2: To characterize the functions of miRNAs in insulin resistance, glucose uptake and production, fatty acid oxidation and lipogenesis. Aim 3: To identify factors and dissect the pathways that regulate RNA uptake by cells and to develop novel pharmacological treatment strategies to manipulate miRNA-expression. Together, this proposal will shed light on the function that miRNA regulatory networks play in metabolism and in the pathophysiology of obesity/type 2 diabetes. In addition, these studies will contribute to the development of new RNA delivery technologies that are urgently needed as experimental tools as well as for novel therapeutic strategies.

2 021 235 €

Start date: 2010-07-01, End date: 2015-06-30

If we are ever to unravel the mysteries of brain function at its most fundamental level, we will need a precise understanding of how its component neurons connect to each other. Furthermore, given the many recent advances in genetic engineering, viral targeting, and immunohistochemical labeling of specific cellular structures, there is a growing need for automated quantitative assessment of neuron morphology and connectivity. Electron microscopes can now provide the nanometer resolution that is needed to image synapses, and therefore connections, while Light Microscopes see at the micrometer resolution required to model the 3D structure of the dendritic network. Since both the arborescence and the connections are integral parts of the brain's wiring diagram, combining these two modalities is critically important. In fact, these microscopes now routinely produce high-resolution imagery in such large quantities that the bottleneck becomes automated processing and interpretation, which is needed for such data to be exploited to its full potential. We will therefore use our Computer Vision expertise to provide not only the necessary tools to process images acquired using a specific modality but also those required to create an integrated representation using all available modalities. This is a radical departure from earlier approaches to applying Computer Vision techniques in this field, which have tended to focus on narrow problems. State-of-the-art methods have not reached the level of reliability and integration that would allow automated processing and interpretation of the massive amounts of data that are required for a true leap of our understanding of how the brain works. In other words, we cannot yet exploit the full potential of our imaging technology and that is what we intend to change.

2 495 982 €

Start date: 2010-04-01, End date: 2016-03-31

Nanosystems are integrated systems exploiting nanoelectronic devices. In particular, this proposal considers silicon nanowire and carbon nanotube technologies as replacement/enhancement of current silicon technologies. This proposal addresses high-risk, high-reward research, unique in its kind. The broad objective of this proposal is to study system organization, architectures and design tools which, based on a deep understanding and abstraction of the manufacturing technologies, allow us to realize nanosystems that outperform current integrated systems in terms of capabilities and performance. Thus this proposal will address modelling of technological aspects, synthesis and optimization of information processing functions from high-level specifications into the nanofabric, and new design technologies for specific aspects of nanosystems including, but not limited to, sensing and interfacing with the environment. This proposal will address also cross-cutting design goals such as ultra-low power and high-dependability design, with the overall objective of realizing nanosystems that are autonomous (w.r. to energy consumption) and autonomic (i.e., self healing). The scientific novelty of this proposal stems from the use of a nanofabric, where computation, sensing and communication are supported by a homogeneous means as well as from the study of algorithmic tools for mapping high-level functions onto the nanofabric. The intrinsic benefit of this research is to provide a design flow that extends both the technological basis and the capabilities of integrated systems, thus strengthening the industrial European position in a key sector where disruptive innovation is key for survival. The extrinsic benefit of this research is to broaden the use of nanosystems to new domains, including mobile/distributed embedded systems, health/environment management, and other areas that are critical to our lives.

2 499 594 €

Start date: 2010-04-01, End date: 2015-12-31

Global concerns regarding the economy, environment and sustainable energy resources dictate an urgent need for the design of novel catalytic reactions. We have recently discovered novel, environmentally benign reactions catalyzed by pincer complexes, including an entirely new reaction, namely the direct coupling of alcohols with amines to produce amides and H2 (Science, 2007, 317, 790). We believe that the mechanisms of these reactions involve a new concept in catalysis: metal-ligand cooperation by aromatization-dearomatization of the ligand. Such cooperation can play key roles also in the activation of H2, C-H, and other bonds. Remarkably, we have very recently discovered a new strategy towards light-induced water splitting into H2 and O2, also based on metal-ligand cooperation in a pincer system, and have observed an unprecedented O-O bond formation process (Science, in press). The design of efficient catalytic systems for splitting water into hydrogen and oxygen, driven by sunlight, and without use of sacrificial reagents, is among the most important challenges facing science today, underpinning the potential of hydrogen as a clean, sustainable fuel. In this context, it is essential to enhance our understanding of the fundamental chemical steps involved in such processes. We plan to (a) explore the scope of bond activation and catalysis based on the new concept of metal ligand cooperation by aromatization-dearomatization (b) study the mechanism and scope of the newly discovered novel approach towards water splitting by light (c) develop novel environmentally benign catalytic reactions involving O-H, C-H and other bonds, such as anti-Markovnikov hydration of alkenes (d) develop unprecedented asymmetric catalysis using chiral cooperating ligands (e) develop new CO2 chemistry, including its hydrogenation to methanol and photolytic splitting to CO and O2. The research is expected to lead to novel catalysis, of importance to environment and sustainable energy.

1 912 018 €

Start date: 2010-04-01, End date: 2015-03-31

Oxidation reactions are of fundamental importance in Nature and are key transformation in organic synthesis. There is currently a need from society to replace waste-producing expensive oxidants by environmentally benign oxidants in industrial oxidation reactions. The aim with the proposed research is to develop novel green oxidation methodology that also involves hydrogen transfer reactions. In the oxidation reactions the goal is to use molecular oxygen (air) or hydrogen peroxide as the oxidants. In the present project new catalytic oxidations via low-energy electron transfer will be developed. The catalytic reactions obtained can be used for racemization of alcohols and amines and for oxygen- and hydrogen peroxide-driven oxidations of various substrates. Examples of some reactions that will be studied are oxidative palladium-catalyzed C-C bond formation and metal-catalyzed C-H oxidation including dehydrogenation reactions with iron and ruthenium. Coupled catalytic systems where electron transfer mediators (ETMs) facilitate electron transfer from the reduced catalyst to molecular oxygen (hydrogen peroxide) will be studied. Highly efficient reoxidation systems will be designed by covalently linking two electron transfer mediators (ETMs). The intramolecular electron transfer in these hybrid ETM catalysts will significantly increase the rate of oxidation reactions. The research will lead to development of more efficient reoxidation systems based on molecular oxygen and hydrogen peroxide, as well as more versatile racemization catalysts for alcohols and amines.

1 722 000 €

Start date: 2010-01-01, End date: 2015-12-31

I shall pursue two projects both of which belong to the fields of partial differential equations, geometric analysis and mathematical physics. The first project, ``the shock development problem", belongs also to the field of fluid dynamics and aims at a full understanding of how, in the real world of 3 spatial dimensions, hydrodynamic shocks evolve, my previous work having analyzed in detail how they form. The second project, ``the formation of electromagnetic shocks in nonlinear media" aims at establishing how electromagnetic shocks form by the focusing of incoming electromagnetic wave pulses in a nonlinear medium. The case of an isotropic nonlinear dielectric will be studied first, to be followed by the case of a general isotropic medium. The methods of geometric analysis introduced in my previous work shall be employed, in particular the ``short pulse method" introduced in my work on the formation of black holes by the focusing of incoming gravitational waves in general relativity. The application of these methods to the problem for a general isotropic medium will require the development of new geometric structures. My three Ph. D. students shall purse the following three projects, belonging also to the fields of partial differential equations, geometric analysis and mathematical physics. The first project is in nonlinear elasticity. It is the study of the equilibrium configurations, in free space, of a crystalline solid in which a continuous distribution of dislocations is present, and aims at analyzing the relationship between the dislocation distribution and the resulting internal stress field. The second is in general relativity and aims at a theoretical understanding of the phenomena discovered by M. Choptuik in his numerical study of the gravitational collapse of a self-gravitating scalar field in spherical symmetry. The third is the study of hydrodynamic shock interactions and focusing in spherical symmetry.

1 278 000 €

Start date: 2010-03-01, End date: 2015-02-28

Atomistic computer simulation has established itself as one of the most important methodologies in modern science, its impact being felt in areas as diverse as physics, chemistry, geophysics, materials science and biophysics, to name but a few. Yet in spite of great progress in computer power and algorithms, we are still not able to simulate such important phenomena as nucleation, phase transitions or protein folding. The purpose of this proposal is to strongly push back the limits of length, time scale and accuracy of present-day methods, greatly enhancing the scope of atomistic simulations. We expect the impact of a successful outcome of this proposal to revolutionize the field. We shall make use of three technical innovations: an extension of Langevin-type equations to include correlated (colored) noise; the use of h-matrices to speed up electronic structure calculations; and an intelligent use of neural networks. Our strategy will be complex. We plan to speed up ab-initio molecular dynamics calculations considerably and also to generate new and highly accurate effective potentials based on electronic structure calculations. A large part of our effort will be devoted to the time scale problem. In this respect we shall improve metadynamics so that its implementation becomes as general and as automatic as possible, and we shall also introduce methods for reconstructing the real dynamics from metadynamics. Finally, highly innovative and powerful sampling methods based on specially designed colored noise Langevin equations will be developed.

2 499 600 €

Start date: 2010-08-01, End date: 2015-07-31

The general theme of this research proposal involves random walks and percolation theory. The research proposal aims at exploring in depth the surprising links between on the one hand a class of problems directly pertaining to random walks (in particular related to disconnection, or the creation of large vacant components and separating interfaces), and on the other hand questions pertaining to a non-conventional model of percolation based on random interlacements. Traditional methods of percolation typically do not apply to this model. The present research program if successfull ought to uncover new paradigms and lead to the development of new methods.

583 092 €

Start date: 2010-04-01, End date: 2014-03-31

Signal representations with Fourier and wavelet bases are central to signal processing and communications. Non-linear approximation methods in such bases are key for problems like denoising, compression and inverse problems. Recently, the idea that signals that are sparse in some domain can be acquired at low sampling density has generated strong interest, under various names like compressed sensing, compressive sampling and sparse sampling. We aim to study the central problem of acquiring continuous-time signals for discrete-time processing and reconstruction using the methods of sparse sampling. Solving this involves developing theory and algorithms for sparse sampling, both in continuous and discrete time. In addition, in order to acquire physical signals, we plan to develop a sampling theory for signals obeying physical laws, like the wave and diffusion equation, and light fields. Together, this will lead to a sparse sampling theory and framework for signal processing and communications, with applications from analog-to-digital conversion to new compression methods, to super-resolution data acquisition and to inverse problems in imaging. In sum, we aim to develop the theory and algorithms for sparse signal processing, with impact on a broad range of applications.

1 839 174 €

Start date: 2010-05-01, End date: 2015-04-30

The present project addresses numerical analysis and algorithmic realization of sparse, adaptive tensor product discretizations of partial differential equations (PDEs) in high dimensions with stochastic data. The aim of the project is to develop mathematically founded adaptive algorithms which are based on sparse tensorization of hierarchic Riesz bases or frames. These will be hierarchic multilevel bases in the physical domain, either Finite Element wavelet type bases or hierarchical, multilevel bases. In the parameter domains corresponding either to random inputs or to phase spaces in transport problems, spectral type representations of ``polynomial chaos'' type shall be employed. Mathematical aim is to analyzed for a classes of elliptic and parabolic PDEs on high or possibly infinite dimensional parameter spaces adaptive, deterministic and dimension independent solution methods with convergence rates superior to those afforded by Monte Carlo Methods, in terms of accuracy vs. complexity. Algorithmic work will address design of data structures with minimal overhead for the efficient realization of the sparse tensor approximations. Applications include space-time adaptive solvers for elliptic, parabolic and certain parametric hyperbolic PDEs, nonlinear approximate spectral representations of nonstationary random fields, scale-resolving solvers of elliptic and parabolic problems with multiple scales with complexity independent of the number of scales, and sparse, adaptive numerical solvers for parametric transport problems. The project will be in collaboration with coworkers in France, Germany, UK, The Netherlands. The project involves mentoring postdocs and predocs who will be actively involved in all aspects of the research, as well as a teaching component.

1 349 564 €

Start date: 2010-01-01, End date: 2014-12-31

This proposal envisages a research program in the field of systems and signals that will lead to innovative and agile mathematical modeling as well as model-based signal processing and control tools for applications in biology and medicine. The project's goal is to bridge the gap between systems biology, on one hand, and medical signal processing and control engineering, on the other. Mathematical models of systems biology will be used to devise algorithms for biological data processing and computerized medical interventions. Experimental biological and clinical data will be utilized to estimate and characterize the parameters of mathematical models derived for biological phenomena and mechanisms. An extensive collaboration network of medical researchers from Sweden and abroad will provide the project team with necessary experimental data as well as with access to medical competence. The envisaged tools are expected to be applicable more generally but their efficacy will be demonstrated in two main application areas. These areas are endocrinology and neurology for which the use of formal control engineering and signal processing methods is currently deemed to be most promising. The proposed program will result in novel systems and signals tools for medical research and health care enabling multi-input multi-output modeling and analysis of endocrine regulations and providing model-based algorithms for individualized drug dose titration.

2 379 000 €

Start date: 2010-04-01, End date: 2015-03-31

The main objective of this interdisciplinary research project is to elucidate regulatory mechanisms through which the circadian timing system coordinates temporal physiology. This system has a hierarchical architecture, in that a master clock in the brain s suprachiasmatic nucleus synchronizes subsidiary oscillators in nearly all body cells. The establishment of phase coherence is obviously of utmost importance in the coordination of circadian physiology. While recent studies have identified feeding cycles, hormone rhythms, and body temperature oscillations as timing cues for peripheral clocks, the molecular makeup of the involved signalling mechanisms is largely unknown. Using liver and cultured cells as model systems, we will employ two innovative strategies for the elucidation of relevant signalling pathways. (1) STAR-Prom (Synthetic TAndem Repeat-PROmoter display), a technique developed in our laboratory, will hopefully identify most if not all immediate early transcription factors activated in cultured cells by rhythmic blood-borne and temperature-dependent signals. (2) A transgenic mouse model with conditionally active liver clocks will be explored in the genome-wide identification of coding and non-coding transcripts whose rhythmic accumulation is system-driven. The in vivo significance of the components emerging from these approaches will be assessed via RNA interference. Thus, relevant siRNAs will be injected into the tail vein of mice, and their effect on the phase of circadian liver gene expression will be monitored in freely moving mice by using whole body fluorescence imaging. Physiologically important components will serve as entry points for the identification of upstream and downstream constituents in the corresponding signal transduction cascades.

2 360 136 €

Start date: 2010-04-01, End date: 2015-12-31

Multidimensional nuclear magnetic resonance (nD NMR) plays a unique role in Science as a primary tool for the characterization of biomolecules, as part of drug-discovery processes, and in clinical imaging (MRI). Further progress in NMR is hampered by this spectroscopy s low sensitivity, arising from the weak interactions that it involves. The prospects of solving this problem by continuing with incremental bigger machines approaches are poor, given the high maturity reached by existing technologies. The present Project deals with this issue by departing from traditional concepts, and relying on two incipient but highly promising developments in the field. One of these pertains ex situ dynamic nuclear hyperpolarization, an approach capable of eliciting liquid state NMR signals that surpass those afforded by the highest-field spectrometers by factors e10,000. While capable of providing super-signals hyperpolarization has the drawback of involving irreversible changes in the physical state of the sample. This makes it incompatible with nD NMR technologies, requiring the collection of multiple scans identical to one another except for systematic delay variations. As second component in this high-risk/high-gain Project we propose merging hyperpolarization with "ultrafast" methods that we have recently developed for completing arbitrary nD NMR/MRI acquisitions within a single scan. The resulting synergy could increase sensitivity by orders of magnitude, while demanding negligibly small amounts of spectrometer/scanner time to complete nD acquisitions. This should provide an ideal starting point for the analysis of a variety of organic and structural biology problems, and provide new tools to explore in vivo metabolism focusing on cancer biomarkers.

2 499 780 €

Start date: 2010-03-01, End date: 2015-02-28