Search Marsden awards 2008–2017
Search awarded Marsden Fund grants 2008–2017
Fund Type: Marsden Fund
Category: Standard
Year Awarded: 2015
Title: Generating novel biosensors to monitor oxidative stress in the heart
Recipient(s): Dr PP Jones | PI | University of Otago
Dr GR Burgio | AI | Australian National University
Dr JR Erickson | AI | University of Otago
Professor AF Dulhunty | AI | Australian National University
Public Summary: Protein oxidation, a consequence of reactive oxygen species (ROS), is a fundamental form of intracellular signalling. ROS production is differentially regulated in discrete regions of the cell, but despite the importance of these 'ROS microdomains' there are currently no tools with the necessary spatial resolution to examine them. In the heart, oxidation is a key regulator of contraction and excess ROS leads to disease, particularly following ischemia-reperfusion injury. ROS augments contraction by increasing calcium release. The calcium release unit in cells of the heart is located within a unique structure, the cardiac dyad, with highly restricted diffusion and localised ROS production. This creates a discrete ROS microdomain. In this project we propose to create mice expressing genetically encoded ROS sensors targeted to the calcium release unit to pioneer the study of the dyad ROS microdomain. We will use these mice to determine when and how ROS within this microdomain is altered. This will allow us to unravel the interplay between ROS and calcium signalling. Understanding how and when the ROS microdomain surrounding the calcium release unit is perturbed will offer a new perspective on how calcium homeostasis is maintained physiologically and becomes corrupted during disease.
Total Awarded: $805,000
Duration: 3
Host: University of Otago
Contact Person: Dr PP Jones
Panel: BMS
Project ID: 15-UOO-034
Fund Type: Marsden Fund
Category: Fast-Start
Year Awarded: 2009
Title: Genetic diversity and recombination analysis of geminiviruses in Australasia
Recipient(s): Dr A Varsani | PI | University of Canterbury
Dr P Lefeuvre | AI | Centre de Coopération Internationale en Recherche
Dr J-M Lett | AI | Centre de Coopération Internationale en Recherche
Dr D Martin | AI | Univeristy of Cape Town
Dr D Shepherd | AI | University of Cape Town
Public Summary: Geminiviruses are emerging plant viruses that cause severe diseases in a wide variety of plant species including those of agricultural importance. A staggering amount of geminivirus diversity has been catalogued throughout the world, however, very little has been documented for Australasia. Our research on Indian Ocean Islands suggests that islands are potential geminvirus diversity hotspots. Our project aims to: (1) catalogue and characterise the geminivirus diversity in Australasia (including novel viruses) using high through put methods developed by us; (2) analyse inter and intra island recombination patterns; (3) determine the movement of geminiviruses within and between the pacific islands.
Total Awarded: $266,667
Duration: 3
Host: University of Canterbury
Contact Person: Dr A Varsani
Panel: EEB
Project ID: 09-UOC-014
Fund Type: Marsden Fund
Category: Standard
Year Awarded: 2012
Title: Genetic jigsaws with missing pieces: mathematical challenges for piecing together evolution from patchy taxon coverage.
Recipient(s): Prof MA Steel | PI | University of Canterbury
Prof C Semple | PI | University of Canterbury
Prof O Gascuel | AI | Universite Montpellier 2
Prof Mossel | AI | University of California, Berkeley
Prof MJ Sanderson | AI | The University of Arizona
Public Summary: Modern ‘next generation’ DNA technology is providing unprecedented amounts of sequence data, within which the evolutionary story of life on our planet is hidden. Fundamentally new mathematical and algorithmic techniques are needed to deal with the scale of these data, and their fragmented and often patchy structure, in which genes are present in some taxa but not in others. This proposal will address three central theoretical projects that underlie the theory for building evolutionary trees from such incomplete data, combining the international expertise of five researchers across four disciplines – mathematics, statistics, computer science, and biology.
Total Awarded: $482,609
Duration: 3
Host: University of Canterbury
Contact Person: Prof MA Steel
Panel: MIS
Project ID: 12-UOC-016
Fund Type: Marsden Fund
Category: Standard
Year Awarded: 2008
Title: Genetic programming for classification tasks
Recipient(s): Dr M Zhang | PI | Victoria University of Wellington
Public Summary: Classification is an important but difficult problem. Genetic programming (GP) has been used to automatically build classification programs via evolution and has achieved a certain level of success. However, the class imbalance problem, the high dimension of features and the evolution cost still make GP not sufficiently powerful and expressive for complex classification tasks. This project will address these issues by exploring a novel fitness function, a new feature discovery/construction algorithm, and novel redundancy removal operators in GP that will extend the expressive power of GP, reduce the computational cost, and improve the effectiveness and comprehensibility of evolved programs for classification.
Total Awarded: $413,333
Duration: 3
Host: Victoria University of Wellington
Contact Person: Dr M Zhang
Panel: MIS
Project ID: 08-VUW-014
Fund Type: Marsden Fund
Category: Standard
Year Awarded: 2015
Title: Genetic programming for dynamic flexible job shop scheduling
Recipient(s): Professor M Zhang | PI | Victoria University of Wellington
Associate Professor KC Tan | AI | National University of Singapore
Public Summary: Flexible Job Shop Scheduling(FJSS) is an important challenging optimisation problem in the manufacturing industry. Given a set of heterogeneous machines and a set of jobs with different requirements, FJSS aims to allocate the jobs to machines(routing) and sequence the allocated jobs on each machine (sequencing) to achieve good delivery speed and high customer satisfaction. FJSS extends the simpler job shop scheduling(JSS) problem, which only requires sequencing decisions. All jobs and their attributes are completely known in advance in static-FJSS, but are not known in dynamic-FJSS, which better reflects the situations in real-world systems.
Dispatching rules are an attractive solution to dynamic-FJSS because of their compactness and interpretability by human operators. However, manual design of such rules is difficult and time-consuming. Due to its flexible representation and global search ability, genetic programming(GP) has shown promise for automatically generating effective and robust dispatching rules for JSS and static-FJSS.
However, GP program representations, fitness evaluations, and optimisation/search mechanisms are not yet sufficiently powerful and expressive to solve dynamic-FJSS problems effectively. We will explore novel GP program representations, surrogate evaluation models, local-search mechanisms, and optimisation mechanisms for multiple conflicting objectives to evolve robust, effective, generalisable and comprehensible scheduling rules for dynamic-FJSS under uncertainty.
Total Awarded: $550,000
Duration: 3
Host: Victoria University of Wellington
Contact Person: Professor M Zhang
Panel: MIS
Project ID: 15-VUW-044
Fund Type: Marsden Fund
Category: Standard
Year Awarded: 2012
Title: Genetic programming for job shop scheduling
Recipient(s): Prof M Zhang | PI | Victoria University of Wellington
Assoc Prof KC Tan | AI | National University of Singapore
Public Summary: Job Shop Scheduling (JSS) is important in the manufacturing industry. Given sets of machines and jobs, the objective is to determine a schedule that achieves good delivery speed and customer satisfaction. Due to interpretability to human operators and ability to cope with dynamic problems, dispatching rules are considered very attractive to JSS. However, manual design of such rules is difficult, time consuming and often infeasible: human operators are unable to identify all subtle and interrelated conditions between scheduling attributes.
Due to flexible representation and global search ability, Genetic Programming (GP) has been recently used to automatically evolve dispatching rules. Although promising, GP still has issues concerning program representations, attribute selection/construction ability, and optimisation/search mechanisms that are not sufficiently powerful and expressive, which often lead to unsatisfactory performance and non-reusable rules for complex JSS in dynamic environments.
We will explore novel GP program representations, new attribute discovery/selection algorithms, new algorithms for optimising multiple conflicting objectives, and novel dynamic rules with local search mechanisms to evolve reusable and competitive dispatching rules for JSS. These developments are expected to extend the expressive power of GP and improve the effectiveness and comprehensibility of evolved dispatching rules for JSS in dynamic environments.
Total Awarded: $452,174
Duration: 3
Host: Victoria University of Wellington
Contact Person: Prof M Zhang
Panel: MIS
Project ID: 12-VUW-134
Fund Type: Marsden Fund
Category: Standard
Year Awarded: 2016
Title: Genomes, phenotypes and fossils: integrative models of species evolution
Recipient(s): Professor AJ Drummond | PI | The University of Auckland
Dr D Welch | PI | The University of Auckland
Dr NJ Matzke | AI | Australian National University
Professor T Stadler | AI | Eidgenossische Technische Hochschule
Dr TG Vaughan | AI | The University of Auckland
Public Summary: Today, we have vastly more data than at any point in the past; whole genome sequences, ancient DNA, phenotypic traits and fossil data all contain fingerprints of evolution. With this wealth of data, we expect coherent estimates of the pattern and timing of evolutionary events. Yet modern solutions to these questions often consider different information in isolation, leading to conflicting results.
This project aims to create a unified inference framework to simultaneously analyse genomic sequences, phenotypic traits and fossil evidence to give an integrated understanding of both the process and pattern of evolution.
We will develop evolutionary models that combine fossils, morphology, species trees and multi-locus gene trees. We will allow fossilisation and fossil discovery rates to depend on organismal traits, historical biogeography and geological strata. Our models of phenotypic evolution will connect the evolution of continuous traits and discrete morphological characters.
Together, these developments will allow researchers to bring more evidence than ever to bear on the most important questions in evolutionary history. To ensure the broadest reach of our models, we will implement them in free, open-source software tools and prove their utility on outstanding questions in human, animal and pathogen evolution.
Total Awarded: $830,000
Duration: 3
Host: The University of Auckland
Contact Person: Professor AJ Drummond
Panel: EEB
Project ID: 16-UOA-277
Fund Type: Marsden Fund
Category: Standard
Year Awarded: 2009
Title: Geodesics in diffeomorphism groups: geometry and applications
Recipient(s): Dr SR Marsland | PI | Massey University
Professor RI McLachlan | AI | Massey University
Dr M Perlmutter | AI | Massey University
Public Summary: Since Arnold's discovery that the Euler equations for a perfect fluid are geodesic equations on the group of volume-preserving diffeomorphisms there has been much research in this area, driven partly by applications. We will use geometric and analytic approaches to study the Euler equations on the full diffeomorphism group, principally treating the problem as cotangent bundle reduction in infinite dimensions.
Total Awarded: $466,667
Duration: 3
Host: Massey University
Contact Person: Dr SR Marsland
Panel: MIS
Project ID: 09-MAU-044
Fund Type: Marsden Fund
Category: Standard
Year Awarded: 2008
Title: Geographic dynamics of evolving viral populations
Recipient(s): Dr AJ Drummond | PI | The University of Auckland
Dr MA Suchard | PI | University of California, Los Angeles
Public Summary: We propose a unification of geographical population dynamics and phylogenetic analysis of viral populations. Current phylogenetic models allow inferences about genetic variation, however important evolutionary and ecological questions remain inaccessible. A major question asks what the roles of landscape, geography, dispersal and migration in shaping viral genetic history. We propose a novel class of tractable phylogenetic models that explicitly include the spatial distribution of individuals and the geographic features of their habitat. This allows our methods to automatically identify geographic features responsible for structuring populations, and estimate rates, directions and modes of dispersal and migration.
Total Awarded: $616,000
Duration: 3
Host: The University of Auckland
Contact Person: Dr AJ Drummond
Panel: EEB
Project ID: 08-UOA-170
Fund Type: Marsden Fund
Category: Standard
Year Awarded: 2011
Title: Geographies of media convergence: spaces of democracy, connectivity and the reconfiguration of cultural citizenship
Recipient(s): Assoc Prof KT Glynn | PI | University of Canterbury
Dr J Cupples | PI | University of Canterbury
Assoc Prof C Adams | AI | The University of Texas at Austin
Prof L Parks | AI | University of California, Santa Barbara
Public Summary: We are currently living through a period in which centralized forms of media, such as national television and mainstream journalism, are perceived to be in crisis. This crisis is creating new spaces for the development of alternative ways of knowing, watching and making media. Along with them, “media convergence” has emerged as a multidimensional concept that references expanding interconnections and interactivity between media technologies, sites, users and production processes, as well as increasingly interactive relationships between politics and popular media cultures. Technological development, media convergence, and attendant transformations of everyday media production, circulation and consumption practices are giving rise to new forms of political discourse and involvement. The proposed research seeks to delineate the possibilities and limitations for contemporary social transformation within this new media ecology. We will do this by exploring a series of media forms, discourses, practices and technologies (including indigenous people’s media and contemporary developments in entertainment television) whereby new kinds of cultural citizenship are being actively forged. This project is thus designed to advance incipient dialogues between human geography and media studies by asking how practices within popular cultures of media convergence can contribute to the construction or renovation of democratic citizenship.
Total Awarded: $695,652
Duration: 3
Host: University of Canterbury
Contact Person: Assoc Prof KT Glynn
Panel: SOC
Project ID: 11-UOC-008