Project acronym ABLASE
Project Advanced Bioderived and Biocompatible Lasers
Researcher (PI) Malte Christian Gather
Host Institution (HI) THE UNIVERSITY COURT OF THE UNIVERSITY OF ST ANDREWS
Country United Kingdom
Call Details Starting Grant (StG), PE3, ERC-2014-STG
Summary Naturally occurring optical phenomena attract great attention and transform our ability to study biological processes, with “the discovery and development of the green fluorescent protein (GFP)” (Nobel Prize in Chemistry 2008) being a particularly successful example. Although found only in very few species in nature, most organisms can be genetically programmed to produce the brightly fluorescent GFP molecules. Combined with modern fluorescence detection schemes, this has led to entirely new ways of monitoring biological processes. The applicant now demonstrated a biological laser – a completely novel, living source of coherent light based on a single biological cell bioengineered to produce GFP. Such a laser is intrinsically biocompatible, thus offering unique properties not shared by any existing laser. However, the physical processes involved in lasing from GFP remain poorly understood and so far biological lasers rely on bulky, impractical external resonators for optical feedback. Within this project, the applicant and his team will develop for the first time an understanding of stimulated emission in GFP and related proteins and create an unprecedented stand-alone single-cell biolaser based on intracellular optical feedback. These lasers will be deployed as microscopic and biocompatible imaging probes, thus opening in vivo microscopy to dense wavelength-multiplexing and enabling unmatched sensing of biomolecules and mechanical pressure. The evolutionarily evolved nano-structure of GFP will also enable novel ways of studying strong light-matter coupling and will bio-inspire advances of synthetic emitters. The proposed project is inter-disciplinary by its very nature, bridging photonics, genetic engineering and material science. The applicant’s previous pioneering work and synergies with work on other lasers developed at the applicant’s host institution provide an exclusive competitive edge. ERC support would transform this into a truly novel field of research.
Summary
Naturally occurring optical phenomena attract great attention and transform our ability to study biological processes, with “the discovery and development of the green fluorescent protein (GFP)” (Nobel Prize in Chemistry 2008) being a particularly successful example. Although found only in very few species in nature, most organisms can be genetically programmed to produce the brightly fluorescent GFP molecules. Combined with modern fluorescence detection schemes, this has led to entirely new ways of monitoring biological processes. The applicant now demonstrated a biological laser – a completely novel, living source of coherent light based on a single biological cell bioengineered to produce GFP. Such a laser is intrinsically biocompatible, thus offering unique properties not shared by any existing laser. However, the physical processes involved in lasing from GFP remain poorly understood and so far biological lasers rely on bulky, impractical external resonators for optical feedback. Within this project, the applicant and his team will develop for the first time an understanding of stimulated emission in GFP and related proteins and create an unprecedented stand-alone single-cell biolaser based on intracellular optical feedback. These lasers will be deployed as microscopic and biocompatible imaging probes, thus opening in vivo microscopy to dense wavelength-multiplexing and enabling unmatched sensing of biomolecules and mechanical pressure. The evolutionarily evolved nano-structure of GFP will also enable novel ways of studying strong light-matter coupling and will bio-inspire advances of synthetic emitters. The proposed project is inter-disciplinary by its very nature, bridging photonics, genetic engineering and material science. The applicant’s previous pioneering work and synergies with work on other lasers developed at the applicant’s host institution provide an exclusive competitive edge. ERC support would transform this into a truly novel field of research.
Max ERC Funding
1 499 875 €
Duration
Start date: 2015-06-01, End date: 2021-11-30
Project acronym ACCORD
Project Algorithms for Complex Collective Decisions on Structured Domains
Researcher (PI) Edith Elkind
Host Institution (HI) THE CHANCELLOR, MASTERS AND SCHOLARS OF THE UNIVERSITY OF OXFORD
Country United Kingdom
Call Details Starting Grant (StG), PE6, ERC-2014-STG
Summary Algorithms for Complex Collective Decisions on Structured Domains.
The aim of this proposal is to substantially advance the field of Computational Social Choice, by developing new tools and methodologies that can be used for making complex group decisions in rich and structured environments. We consider settings where each member of a decision-making body has preferences over a finite set of alternatives, and the goal is to synthesise a collective preference over these alternatives, which may take the form of a partial order over the set of alternatives with a predefined structure: examples include selecting a fixed-size set of alternatives, a ranking of the alternatives, a winner and up to two runner-ups, etc. We will formulate desiderata that apply to such preference aggregation procedures, design specific procedures that satisfy as many of these desiderata as possible, and develop efficient algorithms for computing them. As the latter step may be infeasible on general preference domains, we will focus on identifying the least restrictive domains that enable efficient computation, and use real-life preference data to verify whether the associated restrictions are likely to be satisfied in realistic preference aggregation scenarios. Also, we will determine whether our preference aggregation procedures are computationally resistant to malicious behavior. To lower the cognitive burden on the decision-makers, we will extend our procedures to accept partial rankings as inputs. Finally, to further contribute towards bridging the gap between theory and practice of collective decision making, we will provide open-source software implementations of our procedures, and reach out to the potential users to obtain feedback on their practical applicability.
Summary
Algorithms for Complex Collective Decisions on Structured Domains.
The aim of this proposal is to substantially advance the field of Computational Social Choice, by developing new tools and methodologies that can be used for making complex group decisions in rich and structured environments. We consider settings where each member of a decision-making body has preferences over a finite set of alternatives, and the goal is to synthesise a collective preference over these alternatives, which may take the form of a partial order over the set of alternatives with a predefined structure: examples include selecting a fixed-size set of alternatives, a ranking of the alternatives, a winner and up to two runner-ups, etc. We will formulate desiderata that apply to such preference aggregation procedures, design specific procedures that satisfy as many of these desiderata as possible, and develop efficient algorithms for computing them. As the latter step may be infeasible on general preference domains, we will focus on identifying the least restrictive domains that enable efficient computation, and use real-life preference data to verify whether the associated restrictions are likely to be satisfied in realistic preference aggregation scenarios. Also, we will determine whether our preference aggregation procedures are computationally resistant to malicious behavior. To lower the cognitive burden on the decision-makers, we will extend our procedures to accept partial rankings as inputs. Finally, to further contribute towards bridging the gap between theory and practice of collective decision making, we will provide open-source software implementations of our procedures, and reach out to the potential users to obtain feedback on their practical applicability.
Max ERC Funding
1 395 933 €
Duration
Start date: 2015-07-01, End date: 2020-12-31
Project acronym ALKENoNE
Project Algal Lipids: the Key to Earth Now and aNcient Earth
Researcher (PI) Jaime Lynn Toney
Host Institution (HI) UNIVERSITY OF GLASGOW
Country United Kingdom
Call Details Starting Grant (StG), PE10, ERC-2014-STG
Summary Alkenones are algal lipids that have been used for decades to reconstruct quantitative past sea surface temperature. Although alkenones are being discovered in an increasing number of lake sites worldwide, only two terrestrial temperature records have been reconstructed so far. The development of this research field is limited by the lack of interdisciplinary research that combines modern biological and ecological algal research with the organic geochemical techniques needed to develop a quantitative biomarker (or molecular fossil) for past lake temperatures. More research is needed for alkenones to become a widely used tool for reconstructing past terrestrial temperature change. The early career Principal Investigator has discovered a new lake alkenone-producing species of haptophyte algae that produces alkenones in high abundances both in the environment and in laboratory cultures. This makes the new species an ideal organism for developing a culture-based temperature calibration and exploring other potential environmental controls. In this project, alkenone production will be manipulated, and monitored using state-of-the-art photobioreactors with real-time detectors for cell density, light, and temperature. The latest algal culture and isolation techniques that are used in microalgal biofuel development will be applied to developing the lake temperature proxy. The objectives will be achieved through the analysis of 90 new Canadian lakes to develop a core-top temperature calibration across a large latitudinal and temperature gradient (Δ latitude = 5°, Δ spring surface temperature = 9°C). The results will be used to assess how regional palaeo-temperature (Uk37), palaeo-moisture (δDwax) and palaeo-evaporation (δDalgal) respond during times of past global warmth (e.g., Medieval Warm Period, 900-1200 AD) to find an accurate analogue for assessing future drought risk in the interior of Canada.
Summary
Alkenones are algal lipids that have been used for decades to reconstruct quantitative past sea surface temperature. Although alkenones are being discovered in an increasing number of lake sites worldwide, only two terrestrial temperature records have been reconstructed so far. The development of this research field is limited by the lack of interdisciplinary research that combines modern biological and ecological algal research with the organic geochemical techniques needed to develop a quantitative biomarker (or molecular fossil) for past lake temperatures. More research is needed for alkenones to become a widely used tool for reconstructing past terrestrial temperature change. The early career Principal Investigator has discovered a new lake alkenone-producing species of haptophyte algae that produces alkenones in high abundances both in the environment and in laboratory cultures. This makes the new species an ideal organism for developing a culture-based temperature calibration and exploring other potential environmental controls. In this project, alkenone production will be manipulated, and monitored using state-of-the-art photobioreactors with real-time detectors for cell density, light, and temperature. The latest algal culture and isolation techniques that are used in microalgal biofuel development will be applied to developing the lake temperature proxy. The objectives will be achieved through the analysis of 90 new Canadian lakes to develop a core-top temperature calibration across a large latitudinal and temperature gradient (Δ latitude = 5°, Δ spring surface temperature = 9°C). The results will be used to assess how regional palaeo-temperature (Uk37), palaeo-moisture (δDwax) and palaeo-evaporation (δDalgal) respond during times of past global warmth (e.g., Medieval Warm Period, 900-1200 AD) to find an accurate analogue for assessing future drought risk in the interior of Canada.
Max ERC Funding
940 883 €
Duration
Start date: 2015-04-01, End date: 2021-03-31
Project acronym CatHet
Project New Catalytic Asymmetric Strategies for N-Heterocycle Synthesis
Researcher (PI) John Forwood Bower
Host Institution (HI) UNIVERSITY OF BRISTOL
Country United Kingdom
Call Details Starting Grant (StG), PE5, ERC-2014-STG
Summary Medicinal chemistry requires more efficient and diverse methods for the asymmetric synthesis of chiral scaffolds. Over 60% of the world’s top selling small molecule drug compounds are chiral and, of these, approximately 80% are marketed as single enantiomers. There is a compelling correlation between drug candidate “chiral complexity” and the likelihood of progression to the marketplace. Surprisingly, and despite the tremendous advances made in catalysis over the past several decades, the “chiral complexity” of drug discovery libraries has actually decreased, while, at the same time, for the reasons mentioned above, the “chiral complexity” of marketed drugs has increased. Since the mid-1990s, there has been a notable acceleration of this “complexity divergence”. Consequently, there is now an urgent need to provide efficient processes that directly access privileged chiral scaffolds. It is our philosophy that catalysis holds the key here and new processes should be based upon platforms that can exert control over both absolute and relative stereochemistry. In this proposal we outline the development of a range of N-heteroannulation processes based upon the catalytic generation and trapping of unique or unusual classes of organometallic intermediate derived from transition metal insertion into C-C and C-N sigma-bonds. We will provide a variety of enabling methodologies and demonstrate applicability in flexible total syntheses of important natural product scaffolds. The processes proposed are synthetically flexible, operationally simple and amenable to asymmetric catalysis. Likely starting points, based upon preliminary results, will set the stage for the realisation of aspirational and transformative goals. Through the study of the organometallic intermediates involved here, there is potential to generalise these new catalytic manifolds, such that this research will transcend N heterocyclic chemistry to provide enabling methods for organic chemistry as a whole.
Summary
Medicinal chemistry requires more efficient and diverse methods for the asymmetric synthesis of chiral scaffolds. Over 60% of the world’s top selling small molecule drug compounds are chiral and, of these, approximately 80% are marketed as single enantiomers. There is a compelling correlation between drug candidate “chiral complexity” and the likelihood of progression to the marketplace. Surprisingly, and despite the tremendous advances made in catalysis over the past several decades, the “chiral complexity” of drug discovery libraries has actually decreased, while, at the same time, for the reasons mentioned above, the “chiral complexity” of marketed drugs has increased. Since the mid-1990s, there has been a notable acceleration of this “complexity divergence”. Consequently, there is now an urgent need to provide efficient processes that directly access privileged chiral scaffolds. It is our philosophy that catalysis holds the key here and new processes should be based upon platforms that can exert control over both absolute and relative stereochemistry. In this proposal we outline the development of a range of N-heteroannulation processes based upon the catalytic generation and trapping of unique or unusual classes of organometallic intermediate derived from transition metal insertion into C-C and C-N sigma-bonds. We will provide a variety of enabling methodologies and demonstrate applicability in flexible total syntheses of important natural product scaffolds. The processes proposed are synthetically flexible, operationally simple and amenable to asymmetric catalysis. Likely starting points, based upon preliminary results, will set the stage for the realisation of aspirational and transformative goals. Through the study of the organometallic intermediates involved here, there is potential to generalise these new catalytic manifolds, such that this research will transcend N heterocyclic chemistry to provide enabling methods for organic chemistry as a whole.
Max ERC Funding
1 548 738 €
Duration
Start date: 2015-04-01, End date: 2020-03-31
Project acronym COEVOLUTION
Project Black holes and their host galaxies: coevolution across cosmic time
Researcher (PI) Debora Sijacki
Host Institution (HI) THE CHANCELLOR MASTERS AND SCHOLARS OF THE UNIVERSITY OF CAMBRIDGE
Country United Kingdom
Call Details Starting Grant (StG), PE9, ERC-2014-STG
Summary Galaxy formation is one of the most fascinating yet challenging fields of astrophysics. The desire to understand
galaxy formation has led to the design of ever more sophisticated telescopes which show a bewildering variety
of galaxies in the Universe. However, the degree to which an interpretation of this wealth of data can succeed
depends critically on having accurate and realistic theoretical models of galaxy formation. While cosmological
simulations of galaxy formation provide the most powerful technique for calculating the non-linear evolution of
cosmic structures, the enormous dynamic range and poorly understood baryonic physics are main uncertainties
of present simulations. This impacts on their predictive power and is the major obstacle to our understanding of
observational data. The objective of this proposal is to drastically improve upon the current state-of-the-art by i)
including more realistic physical processes, such as those occurring at the sphere of influence of a galaxy’s central
black hole and ii) greatly extending spatial dynamical range with the aid of a novel technique I have developed.
With this technique I want to address one of the major unsolved issues of galaxy formation: “How do galaxies and
their central black holes coevolve?” Specifically, I want to focus on three crucial areas of galaxy formation: a) How
and where the very first black holes form, what are their observational signatures, and when is the coevolution with
host galaxies established? b) Is black hole heating solely responsible for the morphological transformation and
quenching of massive galaxies, or are other processes important as well? c) What is the impact of supermassive
black holes on galaxy clusters and can we calibrate baryonic physics in clusters to use them as high precision
cosmological probes? The requested funding is for 50% of the PI’s time and three postdoctoral researchers to
establish an independent research group at the KICC and IoA, Cambridge.
Summary
Galaxy formation is one of the most fascinating yet challenging fields of astrophysics. The desire to understand
galaxy formation has led to the design of ever more sophisticated telescopes which show a bewildering variety
of galaxies in the Universe. However, the degree to which an interpretation of this wealth of data can succeed
depends critically on having accurate and realistic theoretical models of galaxy formation. While cosmological
simulations of galaxy formation provide the most powerful technique for calculating the non-linear evolution of
cosmic structures, the enormous dynamic range and poorly understood baryonic physics are main uncertainties
of present simulations. This impacts on their predictive power and is the major obstacle to our understanding of
observational data. The objective of this proposal is to drastically improve upon the current state-of-the-art by i)
including more realistic physical processes, such as those occurring at the sphere of influence of a galaxy’s central
black hole and ii) greatly extending spatial dynamical range with the aid of a novel technique I have developed.
With this technique I want to address one of the major unsolved issues of galaxy formation: “How do galaxies and
their central black holes coevolve?” Specifically, I want to focus on three crucial areas of galaxy formation: a) How
and where the very first black holes form, what are their observational signatures, and when is the coevolution with
host galaxies established? b) Is black hole heating solely responsible for the morphological transformation and
quenching of massive galaxies, or are other processes important as well? c) What is the impact of supermassive
black holes on galaxy clusters and can we calibrate baryonic physics in clusters to use them as high precision
cosmological probes? The requested funding is for 50% of the PI’s time and three postdoctoral researchers to
establish an independent research group at the KICC and IoA, Cambridge.
Max ERC Funding
1 975 062 €
Duration
Start date: 2015-09-01, End date: 2021-08-31
Project acronym CogSoCoAGE
Project Tracking the cognitive basis of social communication across the life-span
Researcher (PI) Heather Ferguson
Host Institution (HI) UNIVERSITY OF KENT
Country United Kingdom
Call Details Starting Grant (StG), SH4, ERC-2014-STG
Summary A vital part of successful everyday social interaction is the ability to infer information about others (e.g. their emotions, visual perspective, and language). Development of these social skills (termed Theory of Mind, ToM) has been linked to improvements in more general cognitive skills, called Executive Functions (EF). However, to date very little is known of how this link varies with advancing age, and no model exists to explain the relationship. Thus, the key aim of the proposed research is to systematically explore the cognitive basis of social communication and how this changes across the life-span. The research will address three complementary objectives: (1) to what degree can variations in ToM ability across the life-span be accounted for by changes in EF skills, (2) how do ToM ability and EF skill change over time in different age groups (using longitudinal methods, i.e. test-retest of the same participants), and (3) can ToM ability be enhanced through training specific EF skills, and how do these training effects differ across the life-span. Contrary to traditional studies of social communication, I will employ an interdisciplinary approach that links theory and practice from cognitive, social, developmental, and clinical (neuro)psychology to study the relationship between ToM and EF across a broad and dynamic age range (10 to 80+ yrs old). I will use cutting-edge combinations of techniques (eye-tracking and EEG) and paradigms, alongside sophisticated statistical methods to track the timecourse of social understanding, and model how it relates to EF and more general cognitive/social skills (eg. IQ, language) within and between individuals. This research will open up new horizons in ToM research by developing an intervention programme to enhance the quality of social communication in older adults (thus improving their mental health and well-being), which has the potential to be applied in other individuals with social communication deficits (eg. autism).
Summary
A vital part of successful everyday social interaction is the ability to infer information about others (e.g. their emotions, visual perspective, and language). Development of these social skills (termed Theory of Mind, ToM) has been linked to improvements in more general cognitive skills, called Executive Functions (EF). However, to date very little is known of how this link varies with advancing age, and no model exists to explain the relationship. Thus, the key aim of the proposed research is to systematically explore the cognitive basis of social communication and how this changes across the life-span. The research will address three complementary objectives: (1) to what degree can variations in ToM ability across the life-span be accounted for by changes in EF skills, (2) how do ToM ability and EF skill change over time in different age groups (using longitudinal methods, i.e. test-retest of the same participants), and (3) can ToM ability be enhanced through training specific EF skills, and how do these training effects differ across the life-span. Contrary to traditional studies of social communication, I will employ an interdisciplinary approach that links theory and practice from cognitive, social, developmental, and clinical (neuro)psychology to study the relationship between ToM and EF across a broad and dynamic age range (10 to 80+ yrs old). I will use cutting-edge combinations of techniques (eye-tracking and EEG) and paradigms, alongside sophisticated statistical methods to track the timecourse of social understanding, and model how it relates to EF and more general cognitive/social skills (eg. IQ, language) within and between individuals. This research will open up new horizons in ToM research by developing an intervention programme to enhance the quality of social communication in older adults (thus improving their mental health and well-being), which has the potential to be applied in other individuals with social communication deficits (eg. autism).
Max ERC Funding
1 488 028 €
Duration
Start date: 2015-09-01, End date: 2021-02-28
Project acronym complexNMR
Project Structural Dynamics of Protein Complexes by Solid-State NMR
Researcher (PI) Jozef Romuald Lewandowski
Host Institution (HI) THE UNIVERSITY OF WARWICK
Country United Kingdom
Call Details Starting Grant (StG), PE4, ERC-2014-STG
Summary Multidrug resistant bacteria that render worthless the current arsenal of antibiotics are a growing global problem. This grave challenge could be tackled by polyketide synthases (PKSs), which are gigantic modular enzymatic assembly lines for natural products. PKSs could be developed for industry to produce chemically difficult to synthesize drugs, but cannot be harnessed until we understand how they work on the molecular level. However, such understanding is missing because we cannot easily investigate large complexes with current structural biology and modeling methods. A key puzzle is how the function of these multicomponent systems emerges from atomic-scale interactions of their parts. Solving this puzzle requires a holistic approach involving measuring and modeling the relevant interacting parts together.
Our goal is to develop a multidisciplinary approach rooted in solid and solution state NMR that will make possible studies of complexes from PKSs. The two main challenges for the NMR of PKSs are increasing sensitivity and resolution. Recent innovations from our lab allow application of solid-state to study large complexes in 2–10 nanomole quantities. Building on this approach, with a protein-antibody complex as a test case, we will develop new NMR methods that will enable a study of structure and motions of domains in complexes. We will probe, for the first time, the structural dynamics of PKSs of enacyloxin and gladiolin, which are antibiotics against life-threatening multidrug resistant hospital-acquired Acinetobacter baumannii infections and tuberculosis. These studies will guide rational engineering of the PKSs to enable synthetic biology approaches to produce new antibiotics.
If successful, this project will go beyond the state of the art by: enabling studies of unknown proteins in large complexes and providing unique insights into novel mechanisms for controlling biosynthesis in PKSs, turning them into truly programmable synthetic biology devices.
Summary
Multidrug resistant bacteria that render worthless the current arsenal of antibiotics are a growing global problem. This grave challenge could be tackled by polyketide synthases (PKSs), which are gigantic modular enzymatic assembly lines for natural products. PKSs could be developed for industry to produce chemically difficult to synthesize drugs, but cannot be harnessed until we understand how they work on the molecular level. However, such understanding is missing because we cannot easily investigate large complexes with current structural biology and modeling methods. A key puzzle is how the function of these multicomponent systems emerges from atomic-scale interactions of their parts. Solving this puzzle requires a holistic approach involving measuring and modeling the relevant interacting parts together.
Our goal is to develop a multidisciplinary approach rooted in solid and solution state NMR that will make possible studies of complexes from PKSs. The two main challenges for the NMR of PKSs are increasing sensitivity and resolution. Recent innovations from our lab allow application of solid-state to study large complexes in 2–10 nanomole quantities. Building on this approach, with a protein-antibody complex as a test case, we will develop new NMR methods that will enable a study of structure and motions of domains in complexes. We will probe, for the first time, the structural dynamics of PKSs of enacyloxin and gladiolin, which are antibiotics against life-threatening multidrug resistant hospital-acquired Acinetobacter baumannii infections and tuberculosis. These studies will guide rational engineering of the PKSs to enable synthetic biology approaches to produce new antibiotics.
If successful, this project will go beyond the state of the art by: enabling studies of unknown proteins in large complexes and providing unique insights into novel mechanisms for controlling biosynthesis in PKSs, turning them into truly programmable synthetic biology devices.
Max ERC Funding
1 999 044 €
Duration
Start date: 2015-05-01, End date: 2020-10-31
Project acronym CoPS
Project Coevolutionary Policy Search
Researcher (PI) Shimon Azariah Whiteson
Host Institution (HI) THE CHANCELLOR, MASTERS AND SCHOLARS OF THE UNIVERSITY OF OXFORD
Country United Kingdom
Call Details Starting Grant (StG), PE6, ERC-2014-STG
Summary I propose to develop a new class of decision-theoretic planning methods that overcome fundamental obstacles to the efficient optimization of autonomous agents. Creating agents that are effective in diverse settings is a key goal of artificial intelligence with enormous potential implications: robotic agents would be invaluable in homes, factories, and high-risk settings; software agents could revolutionize e-commerce, information retrieval, and traffic control.
The main challenge lies in specifying an agent's policy: the behavioral strategy that determines its actions. Since the complexity of realistic tasks makes manual policy construction hopeless, there is great demand for decision-theoretic planning methods that automatically discover good policies. Despite enormous progress, the grand challenge of efficiently discovering effective policies for complex tasks remains unmet.
A fundamental obstacle is the cost of policy evaluation: estimating a policy's quality by averaging performance over multiple trials. This cost grows quickly with increases in task complexity (making trials more expensive) or stochasticity (necessitating more trials).
To address this difficulty, I propose a new approach that simultaneously optimizes both policies and the manner in which those policies are evaluated. The key insight is that, in many tasks, many trials are wasted because they do not elicit the controllable rare events critical for distinguishing between policies. Thus, I will develop methods that leverage coevolution to automatically discover the best events, instead of sampling them randomly.
If successful, this project will greatly improve the efficiency of decision-theoretic planning and, in turn, help realize the potential of autonomous agents. In addition, by automatically identifying the most useful events, the resulting methods will help isolate critical factors in performance and thus yield new insights into what makes decision-theoretic problems hard.
Summary
I propose to develop a new class of decision-theoretic planning methods that overcome fundamental obstacles to the efficient optimization of autonomous agents. Creating agents that are effective in diverse settings is a key goal of artificial intelligence with enormous potential implications: robotic agents would be invaluable in homes, factories, and high-risk settings; software agents could revolutionize e-commerce, information retrieval, and traffic control.
The main challenge lies in specifying an agent's policy: the behavioral strategy that determines its actions. Since the complexity of realistic tasks makes manual policy construction hopeless, there is great demand for decision-theoretic planning methods that automatically discover good policies. Despite enormous progress, the grand challenge of efficiently discovering effective policies for complex tasks remains unmet.
A fundamental obstacle is the cost of policy evaluation: estimating a policy's quality by averaging performance over multiple trials. This cost grows quickly with increases in task complexity (making trials more expensive) or stochasticity (necessitating more trials).
To address this difficulty, I propose a new approach that simultaneously optimizes both policies and the manner in which those policies are evaluated. The key insight is that, in many tasks, many trials are wasted because they do not elicit the controllable rare events critical for distinguishing between policies. Thus, I will develop methods that leverage coevolution to automatically discover the best events, instead of sampling them randomly.
If successful, this project will greatly improve the efficiency of decision-theoretic planning and, in turn, help realize the potential of autonomous agents. In addition, by automatically identifying the most useful events, the resulting methods will help isolate critical factors in performance and thus yield new insights into what makes decision-theoretic problems hard.
Max ERC Funding
1 480 632 €
Duration
Start date: 2015-10-01, End date: 2021-09-30
Project acronym CRYOMAT
Project Antifreeze GlycoProtein Mimetic Polymers
Researcher (PI) Matthew Ian Gibson
Host Institution (HI) THE UNIVERSITY OF WARWICK
Country United Kingdom
Call Details Starting Grant (StG), PE5, ERC-2014-STG
Summary Fish living in polar oceans have evolved an elegant, macromolecular, solution to survive in sub-zero water: they secrete antifreeze (glyco)proteins (AFGPs) which have several ‘antifreeze’ effects, including ice recrystallization inhibition (IRI) - they slow the rate of ice crystal growth. Ice crystal growth is a major problem in settings as diverse as oil fields, wind turbines, road surfaces and frozen food. Analysis of the process of cryopreservation, whereby donor cells are frozen for later use, has revealed that ice recrystallization is a major contributor to cell death upon thawing. Enhanced cryopreservation methods are particularly needed for stem cell storage to maximize the use of this currently limited resource, but also to enable storage of clinically transfused cells such as platelets and red blood cells. AFGPs have thus far not found application in cryopreservation due to their low availability from natural sources, extremely challenging synthesis, indications of cytotoxicity, but more importantly they have a side effect of shaping ice crystals into needle-shapes which pierces cells’ membranes, killing them. The aim of this ambitious project is to take a multidisciplinary approach to develop synthetic polymers as tunable, scalable and accessible bio-mimetics of AFGPs, which specifically reproduce only the desirable IRI properties. Precision synthetic and biological methods will be applied to access both vinyl- and peptide- based materials with IRI activity. The bio-inspired approach taken here will include detailed biophysical analysis of the polymer-ice interactions and translation of this understanding to real cryopreservation scenarios using blood-borne cells and human stem cells. In summary, this ambitious project takes inspiration from Nature's defense mechanisms that have evolved to allow life to flourish in extreme environments and will employ modern polymer chemistry to apply it to a real clinical problem; cryopreservation.
Summary
Fish living in polar oceans have evolved an elegant, macromolecular, solution to survive in sub-zero water: they secrete antifreeze (glyco)proteins (AFGPs) which have several ‘antifreeze’ effects, including ice recrystallization inhibition (IRI) - they slow the rate of ice crystal growth. Ice crystal growth is a major problem in settings as diverse as oil fields, wind turbines, road surfaces and frozen food. Analysis of the process of cryopreservation, whereby donor cells are frozen for later use, has revealed that ice recrystallization is a major contributor to cell death upon thawing. Enhanced cryopreservation methods are particularly needed for stem cell storage to maximize the use of this currently limited resource, but also to enable storage of clinically transfused cells such as platelets and red blood cells. AFGPs have thus far not found application in cryopreservation due to their low availability from natural sources, extremely challenging synthesis, indications of cytotoxicity, but more importantly they have a side effect of shaping ice crystals into needle-shapes which pierces cells’ membranes, killing them. The aim of this ambitious project is to take a multidisciplinary approach to develop synthetic polymers as tunable, scalable and accessible bio-mimetics of AFGPs, which specifically reproduce only the desirable IRI properties. Precision synthetic and biological methods will be applied to access both vinyl- and peptide- based materials with IRI activity. The bio-inspired approach taken here will include detailed biophysical analysis of the polymer-ice interactions and translation of this understanding to real cryopreservation scenarios using blood-borne cells and human stem cells. In summary, this ambitious project takes inspiration from Nature's defense mechanisms that have evolved to allow life to flourish in extreme environments and will employ modern polymer chemistry to apply it to a real clinical problem; cryopreservation.
Max ERC Funding
1 496 439 €
Duration
Start date: 2015-06-01, End date: 2020-05-31
Project acronym DEEP TIME
Project Dynamic Earth Evolution and Paleogeography through Tomographic Imaging of the Mantle
Researcher (PI) Karin Sigloch
Host Institution (HI) THE CHANCELLOR, MASTERS AND SCHOLARS OF THE UNIVERSITY OF OXFORD
Country United Kingdom
Call Details Starting Grant (StG), PE10, ERC-2014-STG
Summary DEEP TIME will unearth a record of geological time that is buried thousands of kilometres deep. The seafloor that covers two-thirds of the earth's surface is a tiny fraction of all seafloor created during its history – the rest has sunk back into the viscous mantle. Slabs of subducted seafloor carry a record of surface history: how continents and oceans were configured over time and where their tectonic plate boundaries lay. DEEP TIME will follow former surface oceans as far back in time as the convecting mantle system will permit, by imaging subducted slabs down to the core with cutting-edge seismological techniques. Current tectonic plate reconstructions incorporate little if any of this deep structural information, which probably reaches back 300+ million years; they are based on present-day seafloor, which constrains only the past 100-150 million years.
DEEP TIME will match deep slab structure to the geological surface record of subduction – volcanic arcs and other crustal slivers that stayed afloat, survived collisions, and form the world’s largest mountain belts. Integrating these two direct records of subduction, the project will
* Add paleo-trenches to existing plate reconstructions and extend them 2-3 times longer into the past.
* Produce a 3-D atlas of the mantle that matches subducted seafloor with paleo-oceans inferred by land geology.
* Rigorously test the hypothesis of vertical slab sinking, which may yield an absolute mantle reference frame.
Tomographic models and geological land records will be synthesized into quantitative and testable paleogeographic reconstructions that complement and extend existing ones, especially in paleo-oceanic areas. This is likely to transform our understanding of the earth’s physical surface environment and biosphere during Mesozoic times, as well as the formation of natural resources. It also will put observational constraints on elusive mantle rheologies. Nearly every subdiscipline of the earth sciences could benefit.
Summary
DEEP TIME will unearth a record of geological time that is buried thousands of kilometres deep. The seafloor that covers two-thirds of the earth's surface is a tiny fraction of all seafloor created during its history – the rest has sunk back into the viscous mantle. Slabs of subducted seafloor carry a record of surface history: how continents and oceans were configured over time and where their tectonic plate boundaries lay. DEEP TIME will follow former surface oceans as far back in time as the convecting mantle system will permit, by imaging subducted slabs down to the core with cutting-edge seismological techniques. Current tectonic plate reconstructions incorporate little if any of this deep structural information, which probably reaches back 300+ million years; they are based on present-day seafloor, which constrains only the past 100-150 million years.
DEEP TIME will match deep slab structure to the geological surface record of subduction – volcanic arcs and other crustal slivers that stayed afloat, survived collisions, and form the world’s largest mountain belts. Integrating these two direct records of subduction, the project will
* Add paleo-trenches to existing plate reconstructions and extend them 2-3 times longer into the past.
* Produce a 3-D atlas of the mantle that matches subducted seafloor with paleo-oceans inferred by land geology.
* Rigorously test the hypothesis of vertical slab sinking, which may yield an absolute mantle reference frame.
Tomographic models and geological land records will be synthesized into quantitative and testable paleogeographic reconstructions that complement and extend existing ones, especially in paleo-oceanic areas. This is likely to transform our understanding of the earth’s physical surface environment and biosphere during Mesozoic times, as well as the formation of natural resources. It also will put observational constraints on elusive mantle rheologies. Nearly every subdiscipline of the earth sciences could benefit.
Max ERC Funding
1 438 846 €
Duration
Start date: 2015-08-01, End date: 2022-01-31
