Skip to main content »

Trinity College Dublin

Ongoing Research Projects supported by TCHPC

TCHPC allocates resources to assist research in many different fields. Below is a list of current projects been undertaken with the help of TCHPC.
  • Mathematics
    Compute
    B
    Dr. Mike Peardon
    Trinity College Dublin
    The aim of the project is to investigate methods for the numerical solution of the classical N-body problem on graphics processing units. In particular celestial mechanics and molecular dynamics simulations will be performed. One goal of the research is to implement a tree code in CUDA for the solution of long range potential problems (gravity). The implementation of a number of sophisticated integrators for the equations of motion in CUDA will also be attempted. Among these are the Force Gradient class of integrators and the Variational integrators.
  • Engineering
    Storage
    Compute
    B
    Prof. John FitzPatrick
    Trinity College Dublin
    Jet noise is the dominant noise source at take-off from aircraft with conventional engine architectures. A significant amount of experimental and numerical modelling effort is still required in order to characterise the physical mechanisms responsible for the production of sound in jet flows. In addition, the capacity of current prediction methods to inform new noise reduction designs is limited, particularly in the subtler features of turbulence responsible for high frequency noise. This research uses a computational fluid dynamics (CFD) methodology to investigate the capacity for current turbulence modelling techniques, specifically the Large Eddy Simulation (LES) turbulence modelling approach, to predict the frequency dependence of the spatio-temporal properties, especially at high frequencies. This is intended to directly facilitate improvements in the modelling of the smallest scales of turbulence for noise prediction, those length scales too small to be resolved by the computational mesh. In addition, the causal relationship between turbulence and noise is investigated in order to advance the understanding of turbulence as a noise source mechanism.
  • Economics
    Compute
    C
    Dr. Micheal Collins
    Trinity College Dublin
    This project will examine the theory of, and empirical evidence for, the economic mechanisms through which the market structure of the banking sector can influence overall societal welfare, with a view to outline the policy implications for banking sector competition regulation.
  • Life Sciences
    Compute
    C
    Dr. Carlos Pinto
    Trinity College Dublin
    University of Cambridge
    We apply a novel artificial neural network technique to the problem of classifying individuals according to their genetic profiles. These methods have applications in population genetics, qualilty control of genetic data and case control studies.
  • Physics
    Compute
    A
    Dr. Clotilde Cucinotta
    Trinity College Dublin
    Understanding the electronic transport properties of nanoconfined systems under wetting conditions isessential for many applications ranging from molecular nanojunctions to nanoelectronics.In this project we focus on the finite bias transport properties of water in different phases confined at the interface with gold and graphene. To this end we will perform a number of combined DFT based static and dynamicalcalculations and quantum transport simulations. These will be at the level of the nonequilibriumGreen function method, as implemented in SMEAGOL code [1]. We will first address the fundamental question of the length scale under which the coherent tunnelling regime is dominant over ionic transport. Wewill calculate the dependence of the conductance in liquid water on the separation between the confiningsurfaces. Electron transport will be evaluated over a representative number of configurations extracted from a50 ps long ab-initio molecular dynamical simulation, performed using the Quantum-Espressosuite of programs. The effect of an electric field on water distribution and, correspondingly, on theconductance will be also considered. Within the tunnelling regime we will study the dependence of transportproperties of the nanoconfined system on the specific phase of water, also comparing it with thetransport in vacuum. We will explore different Auelectrodes/Ice interfaces, obtained by using differentphases and orientations for the Ice. The dependence will be explained further in terms of the complex bandstructure of the Ice moiety, which determines together with the interface structure the decay of theconductance with thickness. This study will allow us to unravel the relationship between conductance andthe microscopic structure of nanoconfined water and will provide an useful tool to detect the phase of awater sample by current measurements.[1] A.R. Rocha et al., Physical Review B. 73, 085414 (2006); www.smeagol.tcd.ie.30.04.2010 17:
  • Life Sciences
    Storage
    C
    Ian Robertson
    Trinity College Dublin
    The work will focus on validating robust clinical end-points that can be used early in drug development to demonstrate pharmacodynamic activity of potential symptomatic and disease-modifying agents for AD in either healthy volunteers or patients, and thereby ultimately accelerate drug progression into clinical practice. These end-points will be based on enhanced understanding of the cognitive and neurophysiological impairments underlying normal ageing, mild cognitive impairment (MCI) and AD and also on novel markers identified from preclinical models of ageing and AD.
  • Life Sciences
    Storage
    C
    Ian Robertson
    Trinity College Dublin
    To look at how different measures of behavioural arousal interact with one another and whether there is an association between such measures and genes already known to influence arousal and attention.The study involves performing a simple computerized test of cognition and attention in which participants will be asked to identify particular target sounds presented through headphones. Participants will perform this task while we measure electrical activity of the brain using an electroencephalography (EEG) cap and also the size of the pupils using a small camera. For the genetic collection we will ask participants to provide a small saliva sample spitting into a small plastic cup.This session will take no longer than two hours to complete in total.
  • Life Sciences
    Storage
    C
    Prof Hugh Garavan
    Trinity College Dublin
    Neurocognitive risk and protective factors for addiction The area of neurocognitive inquiry into drug abuse is relatively new, but it proceeds from a long history of studying addiction as a cognitive and neurochemical process. We intend to study opiate addiction from the perspective of impulsivity, and hope to find areas of the brain that may be protective for addiction, or conversely create a propensity for addictive behaviour. In this study we have chosen to focus mostly on measures of impulsivity. The reasoning behind this is that many studies have shown that impulsivity can predict drug use and addiction. Evidence abounds for this analysis of the role of impulsivity in addiction. This includes a review of many studies that concludes that "impulsivity is a vulnerability marker for drug addiction" (Verdejo-Garcia Et al, 2008). In addition it has been shown specifically that "heroin and cocaine users show less ability to delay gratification than controls or alcoholics" (Kirby
  • Mathematics
    Compute
    C
    Dr. Sinead Ryan
    Trinity College Dublin
    Quantum chromodynamics describes the strong interaction of quarks and gluons in particle physics. Quantum chromodynamics can be simulated on a discrete 4-D spacetime lattice using high performance computers. Various physical observables such as particle masses can be measured using lattice QCD. There are six different flavours of quark: up, down, charm, strange, top and bottom. The up, down and charm quarks have masses lower than the strange, top and bottom quarks. In this project, the mass spectra of baryons containing one bottom quark and two light quarks is calculated. Large sets of operators which create and annihilate baryons with fixed quantum numbers such as isospin, charge, parity and spin are formed. On the lattice it is difficult to form baryon operators with fixed orbital angular momentum. However, the symmetries of the lattice can be used to form operators which couple to an allowed set of orbital angular momenta. Large sets of operators in each symmetry channel are formed. A variational analysis of correlation matrices for each channel is carried out. The variational analysis allows for the ground state mass and higher excited state masses to be extracted from the correlation matrices. This amounts to pinning down states with hitherto ambiguous orbital angular momenta. A new method known as
  • Life Sciences
    Compute
    B
    Matthew Teasdale
    Trinity College Dublin
    The past 10,000 years have seen a lasting interdependence grow between humans and cows. Domestic cattle are now vital to Northern European economies, with Irish beef exports worth in excess of €1.5 billion each year. Population genetic techniques, applied to next generation sequencing data of both modern and ancient material, have the potential to answer important questions across the whole time period of bovid domestication. We will apply these methods to look at the beginnings of farming in both Europe and Africa, an area of research that is subject to recent debate. Understanding the processes of early bovid domestication are also vital to investigations of the newly sequenced bovine genome, as the effect of these early humans interventions live on in the genomes of modern cattle.

Last updated 29 Oct 2014Contact TCHPC: info | support.