Main Content Region

Ongoing Research Projects supported by Research IT

Listing of project codes and abstracts, describing work undertaken which use the resources of the compute clusters hosted by the Research IT team.

New codes can be obtained by filling in this resource application form.

Search Results
Showing 10 of 423 Results
Project Title Finding Metastable Magnets
Project Code HPC_17_01007
Principal Investigator Prof Stefano Sanvito
Start Date 2017-11-29
End Date 2018-05-31
Abstract In Physics it is not always the most energetically favourable structure the one that forms. In fact, there are many situations in which together with the most stable structure many others also exist. These are called metastable. For example, diamond is a metastable form of carbon at standard temperature and pressure, with graphene being the most stable one. Thus, at least in principle, all diamonds will eventually turn into graphite. In practice, however, such metastable structures take extremely long times to be converted into their thermodynamically stable counterparts, so that to all purposes they can be considered as stable. There are even several examples, where only metastable structures are found in nature, simply because the most stable one is not accessible during the growth process. Since the minimum energy criterion does not apply any longer, predicting the existence of metastable structures is a complicate task. In this project we will explore possible high-throughput [1] ways to predict metastable structures for metallic and magnetic alloys. In particular we will try to identify random metallic alloys, which can sustain a magnetic ground state. The tools for the investigation will be advanced electronic structure theory, in the form or density functional theory, and a statistical analysis of the possible alloy configurations. We will focus on a specific class of materials known as Heusler alloys [2], which contains a large number of known members, with many presenting a large degree of disorder. We will evaluate whether known compounds do exist because of their low energy or because of their high entropy, and estimate the probability to find new ones. If new promising compounds are identified, there will be the possibility to grow them in Prof. Coey’s lab. In particular the student will: 1. Learn how to use a density functional theory code to calculate the electronic " " structure of materials 2. Run calculations for random Heusler alloys 3. Perform a statistical analysis of the results and establish stability rules
Project Title Metastablity for magnets
Project Code HPC_17_01006
Principal Investigator Prof Stefano Sanvito
Start Date 2017-11-29
End Date 2018-11-28
Abstract In Physics it is not always the most energetically favourable structure the one that forms. In fact, there are many situations in which together with the most stable structure many others also exist. These are called metastable. For example, diamond is a metastable form of carbon at standard temperature and pressure, with graphene being the most stable one. Thus, at least in principle, all diamonds will eventually turn into graphite. In practice, however, such metastable structures take extremely long times to be converted into their thermodynamically stable counterparts, so that to all purposes they can be considered as stable. There are even several examples, where only metastable structures are found in nature, simply because the most stable one is not accessible during the growth process. Since the minimum energy criterion does not apply any longer, predicting the existence of metastable structures is a complicate task. In this project we will explore possible high-throughput ways to predict metastable structures for metallic and magnetic alloys. In particular we will try to identify random metallic alloys, which can sustain a magnetic ground state. The tools for the investigation will be advanced electronic structure theory, in the form or density functional theory, and a statistical analysis of the possible alloy configurations. We will focus on a specific class of materials known as Heusler alloys, which contains a large number of known members, with many presenting a large degree of disorder. We will evaluate whether known compounds do exist because of their low energy or because of their high entropy, and estimate the probability to find new ones.
Project Title Amidine-like derivatives targeting the Ras/Raf/MEK/ERK pathway
Project Code HPC_17_01005
Principal Investigator Prof Isabel Rozas
Start Date 2017-10-01
End Date 2021-09-30
Abstract About 30% of human tumours are caused by the Ras oncogene and its signalling cascade (MAPK kinase pathway: involving the Ras/Raf/MEK/ERK kinases), which cause uncontrolled cell division and growth, proliferation, and evasion of apoptosis. Particularly, finding mutationally activated BRaf in several cancers has changed the understanding of the role of the Raf kinases in oncogenesis. BRaf mutations are common in melanoma but BRaf point mutations also occur in thyroid (60%), colorectal (10%) and lung (6%) cancers, many of which are insensitive to current BRaf inhibitors. Thus, inhibiting the MAPK pathway is an attractive approach for chemotherapeutic agents. Rozas’ group has previously developed a series of guanidine containing derivatives with similarity to known kinase inhibitors, which inhibit 96-99% the Raf/MEK pathway. Additionally, cell-viability and apoptosis studies showed positive results in several cancer cell lines and further biological assays suggested that these compounds are type- III (allosteric) BRaf inhibitors. In a targeted therapy approach, in this project we aim to develop more selective anticancer therapeutics by targeting the MAPK kinases pathway and will explore the biochemical implications of this pathway disruption by: (i) designing, using computational molecular modelling techniques, the suitability of the compounds proposed as BRaf inhibitors by means of docking experiments with an ATP-containing BRaf model developed by Rozas’ group; (ii) synthesise the most promising compounds indicated by the computational study as non-ATP-competitive BRaf inhibitors, making use of Rozas’ standard synthetic approaches and new optimized methodologies; (iii) biochemically study their cytotoxicity, by means of in vitro cell viability assays, in a number of cancer cell lines (leukemia: HL-60, breast cancer: MCF-7, colorectal cancer: RKO-T29, cervical cancer: HeLa, neuroblastoma: Kelly) and potential to induce apoptosis; and (iv) assess the inhibitory potential of these compounds in the MAPK pathway
Project Title Derivatives of guanidine-based DNA minor groove binders as antiprotozoal agents
Project Code HPC_17_01004
Principal Investigator Prof Isabel Rozas
Start Date 2017-10-01
End Date 2021-10-01
Abstract According to the World-Health-Organization (WHO), in 2015 there were 207-million cases of malaria and 627,000 deaths and 7,216 cases of sleeping-sickness were recorded. Still no definitive cure has been found for parasitic diseases due to the emergence of resistance to the drugs used. Therefore, it is highly important to find new cures for these conditions that affect a large number of people in the world. Considering the good results previously obtained by Rozas' group for a series of aminoalkyl derivatives of guanidine DNA minor groove binders as antimalarial agents, and taking into account that two possible mechanism of action could be behind this activity (i.e. binding to the DNA minor groove or inhibition of the dihydrofolate reductase thymidine synthetase of the parasite –DHFR-TS-), we propose to prepare new analogues to improve and better understand such antiprotozoal activity. This optimization phase will include several steps: i) computational modelling of the interaction of the proposed molecules with the minor groove of DNA and with the active site of DHFR-TS; ii) synthesis of the most promising compounds as informed by the computational study and following synthetic pathways known in Rozas’ laboratory; iii) study of the affinity of the compounds prepared towards DNA by means of biophysical experiments; and iv) biological evaluation of the antiprotozoan effect of these compounds in several parasites such as those causing malaria or sleeping sickness as well as of the potential inhibition of the DHFR-TS of the malaria parasite. In addition, the teratogenic properties of these compounds will be assessed. In the long-term, if the compounds prepared show the activity we are aiming for, this research could benefit people in less developed countries who suffer from parasitic infections such as malaria.
Project Title QoS optimisation for IoT Service Composition
Project Code HPC_17_01003
Principal Investigator Professor Siobhan Clarke
Start Date 2017-11-13
End Date 2018-09-01
Abstract IoT services have a number of QoS attributes that are constantly changing. When used in a composed service, the QoS needs to be continuously optimised to account for the variability in resources.
Project Title Autotune – Determining the Feasibility of using Objective Electrophysiological Audiometry to develop an Integrated Sel
Project Code HPC_17_01002
Principal Investigator Assistant Professor Edmund Lalor
Start Date 2017-11-10
End Date 2018-04-30
Abstract Currently audiologists rely on time consuming behavioural measures to assess a person’s hearing. The aim of project is to explore a rapid, automatic technique. The idea was founded on a closed-loop ‘self-tuning hearing aid’ that uses electrophysiological signals (electroencephalography, EEG) from the auditory nervous system to measure hearing to tune the hearing aid to the user’s audiogram. The project is based on converging scientific, technological and market trends in the hearing aid sector. The aim is to evaluate the feasibility of integrating an objective electrophysiological hearing test into a self-tuning hearing aid using a bench prototype. The first stage aims to replicate and improve upon the results of clinical audiometry by using multi-channel EEG measures to characterise the function of the auditory nervous system. The viability of such an approach has been established by the auditory neuroscience community. However, integrating a full EEG system into a hearable is not feasible. Therefore, the second aim is to evaluate the feasibility of using a more integration-friendly approach, which utilises single- or bi-channel EEG, to replicate clinical audiometry. If such an approach performs as well as clinical audiometry, then we believe that the miniaturisation and integration into a hearing aid is technically feasible.
Project Title FERROVOLT
Project Code HPC_17_00999
Principal Investigator Prof Stefano Sanvito
Start Date 2017-10-26
End Date 2019-10-25
Abstract For a better understanding and design of ferroelectric photovoltaics: First-principles study of optical absorption and charge-carrier transport at ferroelectric domain walls in BiFeO3 The goal of this project is to help find the rules for a domain-wall engineering that optimizes photovoltaic efficiency of potential future-generation ferroelectric solar cells. The material to be studied is BiFeO3 as the most promising photovoltaic ferroelectric material known. Does the photovoltaic effect in BiFeO3 occur at the domain walls or in the bulk? What does it take a domain-wall to conduct electrons? The project aims at establishing the necessary conditions for electric fields and electrical conductivity at ferroelectric domain walls. Since experimental evidence is inconclusive, state-of-the-art ab initio methods will be applied. Electric fields have a long spatial range, so we will go beyond the standard supercell approach to obtain the spatial gradient of the band structure at the domain wall, needed to obtain charge-carrier distributions and electric fields. The Green's-function method for electronic quantum transport will be used for this purpose because it is suitable for extended, non-periodic systems. We will obtain the electrical conductivity as a function of the domain-wall type, structure, and purity. Conclusions for the role of the domain walls in BiFeO3 will be generalized as far as possible in order to apply them to other ferroelectric semiconductors as well. The project is positioned where fundamental condensed-matter physics meets applied solar-cell research. It is expected to advance the frontier of knowledge in basic research and to lay the ground for further research on ferroelectric photovoltaics. It is a contribution to the efforts of the European Union to develop innovative solutions for a sustainable energy supply that help achieve independence of fossil energy. In addition, high-throughput screening and computational characterization for promising photovoltaic absorber materials is planned, including, but maybe not limited to, perovskites and related compounds.
Project Title Finding Metastable magnets
Project Code HPC_17_00998
Principal Investigator Prof Stefano Sanvito
Start Date 0000-00-00
End Date 0000-00-00
Abstract In Physics it is not always the most energetically favourable structure the one that forms. In fact, there are many situations in which together with the most stable structure many others also exist. These are called metastable. For example, diamond is a metastable form of carbon at standard temperature and pressure, with graphene being the most stable one. Thus, at least in principle, all diamonds will eventually turn into graphite. In practice, however, such metastable structures take extremely long times to be converted into their thermodynamically stable counterparts, so that to all purposes they can be considered as stable. There are even several examples, where only metastable structures are found in nature, simply because the most stable one is not accessible during the growth process. Since the minimum energy criterion does not apply any longer, predicting the existence of metastable structures is a complicate task. In this project we will explore possible high-throughput ways to predict metastable structures for metallic and magnetic alloys. In particular we will try to identify random metallic alloys, which can sustain a magnetic ground state. The tools for the investigation will be advanced electronic structure theory, in the form or density functional theory, and a statistical analysis of the possible alloy configurations. We will focus on a specific class of materials known as Heusler alloys, which contains a large number of known members, with many presenting a large degree of disorder. We will evaluate whether known compounds do exist because of their low energy or because of their high entropy, and estimate the probability to find new ones.
Project Title Persistency and population dynamics in perennial ryegrass (Lolium perenne)
Project Code HPC_17_00997
Principal Investigator Dr Trevor Hodkinson
Start Date 2017-10-23
End Date 2020-10-31
Abstract Perennial ryegrass (Lolium perenne) is an important forge crop species in global temperate regions. It underpins the beef and dairy industry in Ireland. As its name suggests, perennial ryegrass (PRG) is perennial in nature and swards can remain in the ground up to decadal periods. Population structure can change dramatically over time, as multiple selective pressures acting on the highly heterogeneous PRG population results in survival of the most persistent individuals. Persistent plants are often not productive, reducing yield and over all sward productivity. Persistency is a complex trait which refers to the ability of PRG individuals to survive in swards and has been identified as a trait of economic performance in Teagasc's Pasture Profit Index. By tracking changes in allele frequencies in monovarietal and mixed swards over time using a genotyping-by-sequencing (GBS) approach, we aim to identify persistent genotypes. GBS is an effective high throughput assay for SNP discovery. However, for some applications the level of coverage which GBS offers is unnecessarily large and in some cases a more cost effective, targeted approach would be more appropriate. Genotyping-in-thousands by sequencing (GT-seq) . For applications such as varietal discrimination, association mapping genotyping thousands of individuals at many thousand this approach may be more suitable. The potential of this assay in these applications in PRG will be investigated as part of this project
Project Title Optimisation via simulation for stochastic set covering problem
Project Code HPC_17_00996
Principal Investigator Prof Stephan Onggo
Start Date 2017-11-01
End Date 2018-11-01
Abstract Set covering problem is an optimisation problem where we want to maximize coverage subject to a set of policy and operational constraints. The problem is often solved using deterministic optimisation algorithms such as integer linear programming and integer non-linear programming. However, many real-world set covering problems are stochastic. In this case, the objective function can only be estimated using a simulation model. In addition, some constraints may need to be estimated using a simulation model too. This research applies optimisation via simulation algorithm to solve stochastic set covering problem