Courses

STEM Summer Research Dublin Courses

You will earn 6 research credits over 8 weeks, conducting faculty-supervised, hands-on, directed study research projects with results that will culminate in the preparation of a research paper. You will complete a minimum of 240 hours on research in and out of the laboratory.

Faculty mentors will work closely with you to direct your continued growth and knowledge development in the chosen research topic discipline.

Please review your project with your academic or study abroad advisor to ensure it will transfer back to your home school and that you are following your home school’s policies.

Choosing Your Research Project

Review Project titles and descriptions below.
List 3 (in order of preference) in your Academic Preferences Form, using DUBL as the course code.
Program is highly individualized, with limited enrollment.
You will need to complete a brief Literature Review in consultation with your research supervisor prior to departure before the start of the program. More details here.
We encourage you to contact Arcadia’s Associate Dean of Academic Access and Curricular Solutions, Rob Hallworth, to discuss your particular research interests further.

Course ID	Title	Credits	Syllabus
DUBI RSLW 392S	International Independent Research in STEM Fields	6	PDF

Summer 2025 Research Projects

Is There Mislabeling of Native Irish Seaweed Products?

Supervisor: Dr Craig Wilding (School of Biology & Environmental Science)

A number of manufacturers sell purported native Irish seaweed products – typically dried seaweed, e.g. dulce, to the health-food market. But are the products what they claim? Fish product mislabelling has been a major issue across Europe - is the same true of seaweed products? Here, multi-locus DNA sequencing of commercial samples will be undertaken to identify the species composition and determine if product labeling is applied correctly.

Relevant majors: Biochemistry, Molecular Biology, Environmental Biology, Genetics, Zoology

The Molecular Characterization of Variation at the Hexokinase and Malate Dehydrogenase loci between Color Morphs of the Beadlet Anemone Actinia Equina

Supervisor: Dr Craig Wilding (School of Biology & Environmental Science)

Desired qualifications: the student should be comfortable with (limited) field-work (collection of Actinia), and learning and applying both molecular techniques and basic bioinformatics.

Relevant majors: Biochemistry, Molecular Biology, Genetics, Zoology

The Molecular Characterization and Expression of Heat Shock Proteins in the Stress Response of the Beadlet Anemone Actinia Equina

Supervisor: Dr Craig Wilding (School of Biology & Environmental Science)

The beadlet anemone Actinia equina (L.) (Cnidaria: Anthozoa: Actiniaria: Actiniidae) is a common intertidal species on European rocky shores and occurs in a variety of color morphs. Previous research has indicated that anemones with red/pink pedal discs (the structure used to attach to the substratum) are genetically differentiated from those with green/gray pedal discs and have different distributions on the intertidal zone (red/pink pedal disc forms are found predominantly higher on the shore and green/gray forms lower on the shore). As intertidal organisms, A. equina are subjected to heat stress on both diurnal and seasonal cycles and are particularly prone to the effects of climate change. This project will study the Heat Shock Protein (HSP) repertoire and response of A. equina. HSPs are an evolutionary conserved family of proteins that act as molecular chaperones, correcting protein folding and/or maintaining protein homeostasis during periods of environmental stress. In this project, the HSP family members will be identified in the A. equina genome. Then gene family members previously identified as involved in the stress response in other organisms (e.g. HSP60 and HSP90) will be characterized in heat-stressed anemones using either qPCR or Western blotting, utilizing assays developed in concert with the student.

Relevant majors: Biochemistry, Molecular Biology, Genetics, Zoology

Unraveling Cold Stress Responses in Brassica napus L.: Physiological and Molecular Approaches to Enhance Crop Resilience

Supervisor: Dr Mortaza Khodaeiaminjan

Cold stress is a significant environmental challenge that affects crop productivity and quality. Brassica napus L., commonly known as oilseed rape, is an economically vital crop cultivated for its oil-rich seeds. However, its growth and yield potential are highly sensitive to low temperatures, particularly during early developmental stages such as germination, seedling establishment, and flowering. This research aims to investigate the physiological and molecular responses of B. napus to cold stress by monitoring plant growth and analyzing gene expression.

The study will be conducted under controlled glasshouse conditions, where B. napus plants will be subjected to cold stress to assess key physiological responses, including changes in overall plant growth, root development, and yield potential. At the molecular level, RNA will be extracted from different plant tissues to analyze the expression profiles of genes associated with cold response. The findings from this research are expected to contribute to breeding strategies for developing cold-tolerant B. napus varieties, ultimately enhancing crop yield and stability in regions prone to low temperatures.

Relevant Majors: Plant Biology, Molecular Biology, Environmental Science

Evaluating Waterlogging Stress Responses in Brassica napus L.: A Physiological and Molecular Approach to Enhancing Flood Tolerance

Supervisor: Dr Mortaza Khodaeiaminjan (School of Biology & Environmental Science)

Waterlogging stress is a major environmental factor that limits crop productivity by restricting oxygen availability to plant roots, which disrupts essential physiological processes.

Brassica napus L., also known as oilseed rape, is an economically important crop grown for its oil-rich seeds, but it is highly susceptible to waterlogging, especially during early developmental stages like germination and seedling establishment. This research aims to investigate the physiological and molecular responses of B. napus to waterlogging stress by assessing plant growth, root function, and yield characteristics and gene expression under simulated waterlogged conditions. The study will be conducted in a controlled glasshouse environment, where B. napus plants will be subjected to waterlogging stress for varying durations to observe the effects on key physiological parameters such as biomass, root and shoot growth, leaf chlorosis, and yield. At the molecular level, RNA will be extracted from different plant tissues to analyze the expression profiles of genes associated with waterlogging response.

Data from this research will contribute to identifying physiological traits that could enhance waterlogging tolerance in B. napus. Such insights are valuable for breeding programs aimed at developing resilient varieties with improved performance under waterlogged conditions, thereby enhancing crop yield stability in flood-prone areas.

Relevant Majors: Plant Biology, Molecular Biology, Environmental Science

Using Camera Trap and Acoustic Surveys to Reveal the Dynamics of Deer “Hotspots” in Ireland

Supervisors: Dr Simone Cuiti and Dr Colin Brock LAB LINK

Camera trap and passive acoustic monitoring surveys are important for gathering information on terrestrial wildlife’s distribution and behavior, providing insights into ecosystem health and functioning. These methods play a crucial role in informing conservation strategies and wildlife management decisions. The UCD Laboratory of Wildlife Ecology and Behaviour has established two major Irish projects monitoring wildlife using non-invasive techniques. The first one -Snapshot Europe - is a coordinated camera trap effort to collect data on mammals across Europe. The second one – bioDEERversity – is a govt-funded biodiversity monitoring program set up in the Wicklow and Dublin mountains. These projects have been gathering data on wildlife and general biodiversity (soil ecology, plant diversity, and vertebrate diversity) in a deer hotspot area of Ireland, where Sika deer have been shown to occur at unsustainable high densities. Camera traps and acoustic recorders have been set up to capture data in areas with no deer (fenced exclusion zones) as well as in those areas spread across a gradient of sika deer relative density. Students involved will learn how to use user-friendly deep learning softwares (e.g., DeepFaune and Kaleidoscope Pro) designed to identify species using large numbers of image and audio files. Furthermore, the students will analyze camera trap data and/or acoustic data, and will have the opportunity to tackle a research question that will be defined with the help of the supervisor (e.g.: What is the effect of deer presence on the occurrence and diversity of the other mammal species?).

Relevant majors: Biology, Ecology, Environmental Science

Investigating a Novel Therapeutic Strategy for Bacterial Meningitis-Related Neuronal Dysfunction

Project Supervisor: Dr Derek Costello Head of Neuroscience

Bacterial meningitis (BM) is a severe condition involving inflammation of the protective membranes surrounding the brain and spinal cord, often leading to irreversible neuronal damage and degeneration. Despite medical advances, BM continues to have a significant mortality rate and long-term consequences for survivors, with approximately 30% experiencing issues like seizures, sensory deficits, and cognitive or neurodevelopmental impairments. These survivors are also at an increased risk of age-related neurodegenerative diseases such as Alzheimer’s disease (AD). The variability in prognosis is exacerbated by regional disparities in vaccine uptake and rising antibiotic resistance, making BM a persistent global health threat.

Gram-positive bacteria, particularly Streptococcus pneumoniae in adults and Streptococcus agalactiae in neonates, are among the leading causes. The complex pathology is characterized by neuroinflammation, largely due to the activation of microglia. Metal ion imbalance further promotes oxidative stress and excitotoxicity, leading to neuronal death. This damage is especially pronounced in the hippocampus, where atrophy contributes to the seizures and cognitive deficits common in BM sequelae.

Our collaborators at TU Dublin have developed a new class of coumarin-derivative chemical agents with metal-chelating and antioxidant properties. Our preliminary evidence shows that these agents can reduce microglial activation and preserve neuronal health under AD related conditions. This study aims to assess this novel set of multi-functional agents as potential therapeutics for BM. We will evaluate their effectiveness in ameliorating bacterial induced inflammation and neuronal dysfunction, in cell-based models of AD in vitro and a zebrafish model of AD in vivo.

Techniques: Cell culture, cell viability, nitric oxide and protein expression analysis, Western immunoblot/qPCR analysis, in-vivo zebrafish behavioral assays.

Relevant majors: Neuroscience, Pharmacology, Pre-Med

The project objectives can be expanded to accommodate two students.

CRISPR Base Editing to Treat Inherited Retinal Disease Models in Zebrafish

Project Supervisors: Professor Breandan Kennedy, Dr. Tess McCann

The Kennedy group uses zebrafish as a model organism to study biological processes relevant to human retinal diseases, many of which arise from single nucleotide mutations in individual genes. Currently, therapeutic options for these conditions are limited. CRISPR base editing offers a promising approach, enabling precise single-base changes in genomic DNA; however, aspects of safety and efficiency still need refinement. At UCD, we have established CRISPR base editing in zebrafish, following protocols by Rosello et al. (2023) and Qin et al. (2024). This project will extend this work to rescue mutations in vision-related genes using zebrafish models.

The student will design guide RNAs targeting zebrafish that carry single nucleotide mutations in genes, such as emc1 and rab28, which are associated with retinal disease phenotypes. Base editing components will be introduced into single-cell zebrafish embryos via microinjections, using established protocols. To assess whether visual function has been restored, the student will conduct optokinetic and visual motor response assays. Genomic DNA from edited models will then be extracted and analyzed through Sanger sequencing to confirm successful base editing. Finally, retinal cryosections will be prepared to evaluate whether retinal morphology has been restored.

This project aims to refine CRISPR base editing applications in zebrafish to rescue retinal disease, potentially paving the way for therapeutic approaches targeting single nucleotide mutation-based vision disorders in humans.

References: Rosello, M., Serafini, M., Concordet, JP. et al. Precise mutagenesis in zebrafish using cytosine base editors. Nat Protoc 18, 2794–2813 (2023). https://doi.org/10.1038/s41596-023-00854-3

Relevant majors: Neuroscience, Genetics, Pharmacology

Association of Lipopolysaccharide Binding Protein (LBP) Concentration as a Biomarker of Gut Barrier Function, and LBP Genetic Variation with Colorectal Cancer

Supervisor: Dr David Hughes - Cancer Biology and Therapeutics Lab, Conway Institute, UCD

Colorectal cancer is a common cancer with currently alarming increases at younger ages, making prevention highly important. Although there are complex disease causes, environmental factors, particularly obesity and lifestyle, are known to play a strong role. Recent evidence suggests that commensal microbial dysregulation and exposures to microbial toxins are involved in its development. We hypothesize that this occurs through inflammatory-induced weakening of the protective gut mucosal barrier by obesity, dietary/lifestyle, and microbiome factors that lead to translation of pathogenic bacteria and their toxins into the epithelial lining of the gastrointestinal tract. To help explore this hypothesis, the student will measure biomarkers of gut-barrier function and bacterial translocation (as biomarkers of microbial dysbiosis) by custom ELISAs, e.g., the human Lipopolysaccharide Binding Protein (LBP) for bacterially derived lipopolysaccharide – integral to an intact barrier (which modulates the permeability of tight junctions between intestinal tract cells). This will be done in a series of blood samples from non-disease controls (n=120) and patients with colorectal cancer (n=120). Secondly, common genetic variants (SNPs; single nucleotide polymorphisms) in human genes within the LPS / LBP pathway (see our related study at doi:10.1093/mutage/geae008) will be genotyped by Taqman genotyping assays from blood DNA samples from these same patients. Together, these findings will illuminate gut-barrier function in colorectal cancer and possible contribution of bacterial toxin exposure from the gut. Note that both Postdoctoral and PhD researchers in the lab will be available to provide support to students who are unfamiliar with these methods. By the end of the project the student will become familiar with several techniques, including ELISA and DNA genotyping assays, and biostatistical packages such as R.

Relevant majors: Genetics, Biochemistry, Epidemiology, Biology

Determining the Host Response to Novel Vaccine Antigens against ESKAPE Pathogens

Supervisor: Assoc Prof Siobhán McClean

Antimicrobial resistance is a massive growing problem in the fight against bacterial infections. The number of antibiotics that are effective at treating many bacterial infections is shrinking. In particular the ESKAPE pathogens are a group of highly virulent pathogens that can escape the majority of antibiotics. Vaccines represent one of the best ways to prevent bacterial infections and have also been shown to reduce antimicrobial resistance 1, We focus on discovering candidates in order to prevent these difficult and challenging infections. We use a proteomic approach to identify highly effective vaccine antigens which prevent infections in mouse models. We have several vaccine projects ongoing in our laboratory against antibiotic resistant infections such as respiratory infections that impact the lives of people with cystic fibrosis 2 ; the tropical infection, melioidosis 3, 4 ; O157 E. coli and three of the ESKAPE pathogens, namely Klebsiella pneumoniae, A.baumannii and P. aeruginosa 5, 6.

We recently demonstrated that three P. aeruginosa antigens were very protective in an acute pneumonia mouse model (7) Immunized mice showed 80-times less bacteria after immunization compared with unimmunized control mice. We are currently examining the protective immunological responses, including antibody responses and cytokine responses in serum or immune cells. This project will focus on investigating the mechanisms of protection of these antigens and investigate how we can maximize the protective response. It will involve using ELISA to determine the levels of antigen specific IgGs in immunized mice. In addition the host response will be further examined by exposing immune cells to antigen and evaluating the profile of cytokines produced using flow cytometry, ELISpot and/ or ELISA. Understanding how the antigens protect against infection is an important stage in progressing the vaccines towards human trials.

Relevant majors: Immunology, Microbiology, Biochemistry

References:

[1] Frost, I., Sati, H., Garcia-Vello, P., Hasso-Agopsowicz, M., Lienhardt, C., Gigante, V., and Beyer, P.(2023) The role of bacterial vaccines in the fight against antimicrobial resistance: an analysis of the preclinical and clinical development pipeline, Lancet Microbe 4, e113-e125.

[2] McClean, S., Healy, M. E., Collins, C., Carberry, S., O'Shaughnessy, L., Dennehy, R., Adams, A.,Kennelly, H., Corbett, J. M., Carty, F., Cahill, L. A., Callaghan, M., English, K., Mahon, B. P.,Doyle, S., and Shinoy, M. (2016) Linocin and OmpW Are Involved in Attachment of the Cystic Fibrosis-Associated Pathogen Burkholderia cepacia Complex to Lung Epithelial Cells and Protect Mice against Infection, Infect Immun 84, 1424-1437. [3] Casey, W. T., and McClean, S. (2015) Exploiting molecular virulence determinants in burkholderia to develop vaccine antigens, Curr Med Chem 22, 1719-1733. [4] Casey, W. T., Spink, N., Cia, F., Collins, C., Romano, M., Berisio, R., Bancroft, G. J., and McClean, S.(2016) Identification of an OmpW homologue in Burkholderia pseudomallei, a protective vaccine antigen against melioidosis, Vaccine 34, 2616-2621. [5] Jurado-Martin, I., Sainz-Mejias, M., and McClean, S. (2021) Pseudomonas aeruginosa: An Audacious Pathogen with an Adaptable Arsenal of Virulence Factors, Int J Mol Sci 22. [6] Ma, C., and McClean, S. (2021) Mapping Global Prevalence of Acinetobacter baumannii and Recent Vaccine Development to Tackle It, Vaccines (Basel) 9.

[7] Jurado-Martin, I., Tomas-Cortazar, J., Hou, Y., Sainz-Mejias, M., Mysior, M. M., Sadones, O.,Huebner, J., Romero-Saavedra, F., Simpson, J. C., Baugh, J. A., and McClean, S. (2024) Proteomic approach to identify host cell attachment proteins provides protective Pseudomonas aeruginosa vaccine antigen FtsZ, NPJ Vaccines 9, 204.

Investigating the Regulation of Cell-Cell Interactions in Chronic Inflammatory Intestinal Diseases

Supervisor: Dr. Mario Manresa

The intestinal mucosa is a dynamic environment full of structural cells such as fibroblasts that interact closely with the immune system to maintain a healthy gut and fight infectious pathogens. Alterations of the normal functioning of this environment can lead to abnormal immune responses that are characteristic of chronic inflammation. In the gut, chronic inflammation leads to ulcerative colitis or Crohn's disease, two debilitating conditions that affect millions of people worldwide and can lead to complications requiring surgical removal of gut segments. In recent years, fibroblasts have been identified as an important component of the abnormal mucosal immune response seen in chronic inflammation. These cells are now thought to establish intricate interactions with macrophages and neutrophils that may exacerbate the inflammatory response and lead to tissue damage. As a result, understanding the signals that activate these disease-contributing functions on fibroblasts and uncovering the ways in which they communicate with the immune system is of great therapeutic value. My research lab recently discovered a mediator that drives this cell-cell communication network. Delving further into this concept, we use primary cells and patient biopsies to understand the mechanisms that coordinate fibroblast-immune interaction. In this project, the student will integrate into our team and contribute to one of our CORE research areas, learning techniques such as in vitro cell culture, flow cytometry and western blot to characterize functional changes in fibroblasts and immune cells during inflammatory responses.

Relevant majors: Biology, Genetics, Biochemistry, Pharmacology

Study of the Cross-Talk between SRF and AR in Prostate Cancer

Supervisor: Dr Maria Prencipe

Current treatments for prostate cancer mainly target the Androgen Receptor (AR), however despite initial response these treatments fail. Serum response factor (SRF) was previously identified as an important transcription factor in in vitro models of castrate resistant prostate cancer (CRPC) and a cross-talk between AR and SRF was demonstrated. To further understand this cross-talk, we used mass spectrometry to identify common interactors between these two proteins. The aim of this project is to manipulate the key common interactors to assess cellular response and to study their signaling pathway in prostate cancer cell lines.

Objectives:

Manipulate common interactors of SRF and AR including HSP70, HSP90, PI3K and Akt using small-molecule inhibitors in a panel of prostate cancer cell lines.
Assess cell proliferation in response to these inhibitors singly and in combination with enzalutamide (current treatment for CRPC).
Assess protein expression and activity of several proteins in the pathway after treatments.

Techniques: Cell culture, MTT assays (cell viability), colony forming assay (cell proliferation), Incucyte (cell proliferation for combination treatments), treatment with small molecule inhibitors, western blotting (protein expression) and luciferase assays (protein activity).

Relevant majors: Pharmacology, Genetics, Biochemistry, Molecular Biology

Two students possible.

Acyl-Dihydropyridines - A New Class of Type-1 Photoinitiators for Holography

Supervisor: Dr Bartosz Bieszczad

Acyl-dihydropyridines (acyl-DHPs, Figure 1) are synthetic analogues of nicotinamide adenine dinucleotides (NADH), important biological co-enzymes. It has been found that upon irradiation with visible light, acyl-DHPs release highly reactive acyl radicals. Acyl radicals can be trapped and used in organic synthesis in order to attach a carbonyl unit to a molecule. Although acyl-DHPs are currently used by many groups around the world to synthesize biologically active molecules, their application to material chemistry is much less developed.

Acyl-DHPs can act as type-1 photoinitiators, i.e. they are able to initiate the polymerization upon irradiation with a visible light, while remaining stable and unreactive towards solvents and other weak nucleophiles. This is a very rare and beneficial property.

Acyl-DHPs can be easily prepared from simple starting materials in a single synthetic step. They are also very tuneable, and many different types are synthetically possible. This project aims to develop new types of acyl-DHPs to act as type-1 photoinitiators for polymerization of methacrylates and other monomers for application in holography. In particular, it is expected that the addition of electron donating groups to the acyl moiety will cause the red shift in absorbance, allowing the use of longer wavelengths: green and red lights. These types of photoinitiators are highly sought after in material science. The student will be involved in the synthesis and characterization of new acyl-DHPs and will apply them to the photopolymerization of methyl methacrylates. The progress of the polymerization will be monitored by Raman spectroscopy.

Relevant major: Chemistry

CORE Olefination: Organocatalytic Extension of the Wittig Chemistry

Supervisor: Dr Kirill Nikitin

The Wittig olefination reaction has gotten a very significant makeover. First, the annoying phosphine oxide by-product can now be used as the starting material. It can be converted directly to quaternary phosphonium salt, QPS, via new fast and high-yielding “Umpolung quaternization.” We have eliminated the waste problem and olefinations can be run using phosphine oxide and avoiding phosphines at the interim stages.

Second, we have developed novel ion-pair carboxylate reagents containing their own (hence Eigenbase) endogenous anionic base. The Eigenbase reagents work in the absence of added bases and this process is hinged on the interplay of structure and function of phosphonium carboxylate ion pairs in different solvents. The olefinations furnish a range of alkenes in high yields, no protecting groups are needed.

Third, most Eigenbase reagents can be prepared directly from alcohols as shown in route 1. This variant termed acidic stoichiometric olefination reaction (SORE) avoids use of halogen derivatives, bases and metal salts altogether.

Fourth, we went a step further and have achieved a shortcut catalytic cycle 2 and developed cycle 3. These circular olefination reactions (CORE) are single-step organocatalytic protocols. The venerable Wittig-type olefination is done without phosphorus waste, protecting groups, organic halides, metal salts and bases. Ironically, it is now acid catalyzed.

Project Aims: identify set of conditions for CORE process and reagents; explore new classes catalyst for example solid acids. You will learn: air-free wet chemistry (O2 and water); detailed NMR characterisation: 1H, 13C, 31P Isolation and purification; Intro-level DFT computations.

Relevant majors: Chemistry, Chemical Engineering

Application of Cinematic Videography for Scientific Communication

Supervisor: Dr. Fun Man Fung

Scientific research is a major global investment and changes people’s lives. In 2024, a projected $2.53 trillion is being allocated worldwide to research and development (R&D). This substantial figure demonstrates the increasing recognition of the importance of scientific innovation in addressing global challenges and driving economic growth. Scientific research is funded by taxpayer’s money. However, science is deemed difficult to comprehend by the public as most of them are not scientists by training. Therefore, there is a need to leverage the technology of videography to bridge the gap between scientists and the public, making scientific research more accessible and engaging.

This study aims to provide empirical evidence for the benefits of using cinematic videography as an accessible tool for scientific communication, potentially contributing to a more informed and scientifically literate society.

Researchers on this project will create and employ cinematic videography as a technique to conduct the research. The person will be assigned to combine visually engaging imagery with clear and concise explanations in order to captivate audiences and facilitate a deeper understanding of complex scientific processes. This work is important and could lead to the next stage of a longitudinal research – Using a mixed-methods approach to investigate the effectiveness of cinematic videography.

Relevant majors: Chemistry, Biology, Physics, Geology

Fundamentals and Applications of Nanopore Chemistry

Supervisor: Dr Robert Johnson

The Johnson group has expertise in the fabrication of nanopore structures and is interested in investigating the confinement of molecules within these structures. The transport of ions, as well as other fundamental chemical properties, can be quite different when confined to the nanoscale as opposed to the bulk. Typical projects will involve the fabrication of nanopores and their characterization, followed by the development of a sensing system for trace analytes of interest. Target analytes range from chemical and biological contaminants in foods and medicine through to toxic ions in the environment. Our research is highly interdisciplinary, with past students in biology, chemistry, physics and chemical engineering all enjoying successful placements within the research group. Projects are tailored to the student’s interests, but typically involve learning/developing some of the following methodologies:

Electrochemical characterization (e.g. cyclic voltammetry, impedance spectroscopy)
Nanoscale imaging (SEM, TEM)
Surface chemistry measurements (e.g., QCM, contact angle)
Surface modification chemistry (“solid-state synthesis”)
Mathematical modeling with Finite Element Analysis
Microbiology (PCR, cell culturing etc.)

Relevant majors: Chemistry, Physics, Chemical Engineering, Biology

Can host up to two students.

Phosphorus-Based Cations as Main Group Catalysts

Supervisor: Dr. Tom Hooper

Catalysis produces over $500 billion worth of products worldwide each year and homogeneous catalysts are used to produce a huge range of compounds, from large quantities of feedstock chemicals to complex drug molecules. Many of these catalysts are based on rare and expensive transition metals, the supplies of which are limited. Developing catalysts based on main-group elements will make catalytic processes less expensive and more sustainable. Phosphorus is an excellent candidate as a catalytic center because it can adopt different coordination numbers and oxidation states and its NMR active nucleus provides an excellent handle to directly interrogate reactions. Phosphenium cations have 2 substituents and phosphorus can interact with substrate molecules in a similar way to metals.

The aim of this project is to synthesize and characterize a range of phosphenium cations with varying steric and electronic properties. This is done by manipulation of the ligand backbone through organic synthetic methods with the phosphenium center introduced using standard air sensitive methods (Schlenk line and glove box techniques). These potential catalysts will be tested for reactivity towards small molecules (H2, NH3, CO2 etc.) and organic substrates (alkenes, alkynes, furans, carbonyls etc.) with emphasis on reversible binding and reactivity. Onward reactivity towards the functionalization of these substrates will be targeted with the aim of regeneration of the catalytic center. In situ analysis of these reactions will be performed by multinuclear NMR spectroscopy, mass spectrometry and reaction kinetics studies.

Relevant majors: Chemistry

WebXR Visualization of the Aqueducts of the Greater Iraklio Area (AGIA)

Supervisor: Dr. Abraham Campbell

The student will collaborate with the School of Archaeology to develop a visualization of the Aqueducts of the Greater Iraklio Area (AGIA). The project’s primary goal is to explore how models can be generated and experienced by users online with varying levels of computer hardware, such as additional notes on the models, to provide users with more historical details. The student will be part of a team working on the project, so they could either come from a Computer Science background to help with that side of the project or come from an Archaeology background to help more with the Archaeology side as one area of research is to see if using the 3D models alone can provide real-world archaeology insights. The models will be generated in multiple ways such as photogrammetry, Neural Radiance Fields (NeRF) and techniques like Gaussian Splatting for presentation.

Relevant majors: Computer Science, Archaeology

Can an LLM Help Shape and Improve an Existing Mobile App?

Supervisor: Dr. Abraham Campbell

The student will join an existing project exploring the use of LLM in app development. The project explores whether Large Language Models (LLMs) can assist non-experts in improving an existing app, specifically focusing on transforming the "Pruritus Severity Scale" (PSS) for burn patients into a mobile application. The goal is to create an accessible, user-friendly app that allows patients to record their experiences daily. Given the unique challenges faced by burn patients, the project will emphasize accessibility and usability while exploring the potential of LLMs in refining the app’s features.

Relevant majors: Computer Science

Large Language Models (e.g., ChatGPT/BERT/LLaMA) for Digital Forensic Investigation

Project Supervisor: Assoc. Prof. Mark Scanlon

A large language model (LLM) is a type of artificial neural network designed for understanding and generating human-like text. It is a subset of models within the broader domain of natural language processing (NLP), which focuses on enabling machines to interpret, generate, and respond to human language. This project aims to develop a novel approach for digital forensic investigation using language models, such as ChatGPT/BERT/LLaMa, to generate forensically sound solutions for discovering and analyzing digital evidence. The proposed method will evaluate the potential for LLMs to enable digital investigators to analyze and interpret digital evidence from a variety of digital data sources and devices. Specifically, the project will look at the prospect of natural language digital forensic query being taken as an input, a step by step process being defined to answer the query, automating the evidence discovery through generated scripting, interpreting the results and presenting this interpretation back using natural language. The solution will be benchmarked using a framework designed in my group.

The work of the student would involve working in a team of researchers in the creation of end-to-end tests to assess the viability of LLM-aided digital forensic investigation. Working with this ongoing project in my research group will expose the research intern to the full research lifecycle from research design, development, experimentation with devices and datasets, and dissemination/discussion of results at UCD Forensics and Security Research Group meetings.

Relevant majors: Computer Science, Computer Engineering, Data Science, Software Engineering

Computer Vision Based Indoor Multimedia Geolocation

Project Supervisor: Assoc. Prof. Mark Scanlon

The task of multimedia geolocation is becoming an increasingly essential component to effectively combat human trafficking and other illegal acts. While text-based metadata can easily provide geolocation information with access to the original media, this metadata is stripped when shared via social media and common chat applications. Geolocating, geotagging, or finding geographical clues in the multimedia content itself is a complex tax. While there are numerous manual/crowdsourcing approaches to this, recent research has shown that computer vision is one viable avenue for research.

The work of the intern would involve working in a team in the creation of novel datasets for multimedia geolocation and developing computer vision-based techniques for indoor multimedia geolocation. The aim is to develop powerful methods for image geolocation that enable more efficient investigations in the field of human trafficking. Color values serve here as a key component to describe specific characteristics of an image and color-based descriptors will be used for Content-Based Image Retrieval. The performance of the developed methods will be evaluated using the Hotels-50K dataset as a foundation. Working with this ongoing project in my research group will expose the research intern to the full research lifecycle from research design, development, experimentation with devices and datasets, international collaboration, and dissemination/discussion of results at UCD Forensics and Security Research Group meetings

Relevant majors: Computer Science, Computer Engineering, Data Science, Cybersecurity

Large Language Model Powered Password Cracking to Overcome Encryption for Law Enforcement

Project Supervisor: Assoc. Prof. Mark Scanlon

Passwords have been and still remain the most common method of authentication in computer systems. From accessing your smartphone, to setting up your online banking account or social identification, there is a plethora of passwords that users are required to set and remember in hundreds of websites. Complex passwords make the job of law enforcement engaged in a digital investigation more difficult, especially since time is often of the essence.

This project aims to provide insights into the password selection process of users and the impact of contextual information in it. Additionally, the ways that this contextual information can be leveraged in order to assist with the lawful password cracking process will be explored. Large Language Models (LLMs) will be trained to generate password candidate dictionaries. To this end, intelligent, community-targeted dictionaries will be assembled. For example, when targeting a community of superhero enthusiasts, we assume and want to prove that a larger proportion of passwords would be contextually relevant to that community than not.

The work of the intern would involve contributing to the benchmarking process through a framework for evaluating different password cracking methods, as well as evaluating existing and newly created “smart” dictionaries against existing password datasets. The tools and dictionaries will be evaluated for different metrics and for varying degrees of contextualization, with the aim of establishing the impact of context in passwords. Additionally, the intern can contribute to the dictionary creating process and be able to test their own dictionaries with this framework.

Relevant majors: Computer Science, Computer Engineering, Data Science, Cybersecurity

Optimization of a Multi-Well Pipeline for Imaging Programmed Cell Death in Arabidopsis Cell Suspension Cultures

Project Supervisors: Assoc. Prof. Carl Ng and Prof. Jeremy Simpson

We have recently developed a multi-well pipeline for imaging the classical morphology of protoplast retraction from the cell wall in Arabidopsis suspension cells undergoing stressed induced programmed cell death. This project will build on the methods we have developed to optimize the pipeline. The project will also involve the possible application of AI-tools for post acquisition analysis (time-permitting).

Anticipated methodologies for the project: High-throughput microscopy, image analysis, AI tools.

Relevant majors: Cellular Biology, Botany

Targeting Oxidative Stress and Prolyl Hydroxylase Domain Inhibition as Neuroprotective Strategies in an Oxygen Glucose Deprivation (OGD) Model in Rat Hippocampal Slices

Project Supervisor: Professor John O’Connor

During hypoxia a number of physiological changes occur within neurons including the stabilization of hypoxia-inducible factors (HIFs). The activity of these proteins is regulated by O2, Fe2+, 2-OG and ascorbate dependant hydroxylases which contain prolyl-4- hydroxylase domains (PHDs). PHD inhibitors have been widely used and have been shown to have a preconditioning and protective effect against a later and more severe hypoxic insult. In addition oxidative stress plays an important role in ischemia-reperfusion injury (IRI). Antioxidants have been shown to have beneficial effects during increased levels of reactive oxygen species in ischemic stroke. Therefore, this study will investigate the effects of antioxidants, SOD mimetics and novel PHD inhibitors on synaptic transmission and neuronal viability in an oxygen-glucose deprivation (OGD) rat stroke model. Field excitatory post-synaptic potentials (fEPSPs) will be elicited by stimulation of the Schaffer collateral pathway in young rat CA1 hippocampal slices. During OGD hippocampal slices will be perfused with glucose-free aCSF bubbled with 95% N2/5% CO2 for 20 min.

Relevant majors: Neuroscience, Pharmacology

High-Throughput Screening for Aerotolerance in the Human Pathogen Campylobacter Jejuni

Project Supervisor: Assoc. Prof. Tadhg Ó Cróinín

Campylobacter jejuni is the world’s leading cause of bacterial gastroenteritis. It is commonly found as a commensal organism in the chicken gut, with human infections typically arising from the consumption of food contaminated with raw or undercooked chicken. Despite its fastidious nature and poor growth under atmospheric oxygen levels, C. jejuni can persist throughout the food chain and is frequently isolated from retail poultry products. Some strains of C. jejuni show increased tolerance to aerobic conditions, which has implications for C. jejuni transmission, and may be associated with other phenotypic changes. This project will aim to develop a high-throughput method for screening aerotolerance in C. jejuni and apply it to identify aerotolerant isolates in an established collection of supermarket chicken isolates. Aerotolerant and susceptible isolates will then be examined for differences in relevant phenotypes, including biofilm formation, catalase activity, and motility.

Relevant majors: Microbiology, Biochemistry, Food Science

Investigating the Role of Disease-Causing Proteins in Motor Neurons

Project Supervisor: Dr. Niamh O’Sullivan

My lab studies inherited forms of motor neuron disease, particularly hereditary spastic paraplegia (HSP). Individuals with HSP develop weakness in their legs leading to difficulties walking which is caused by degeneration of the very longest motor neurons. Extensive work in recent years has successfully identified many of the genetic causes underpinning HSP, but there are currently no treatments to prevent, cure or even to slow the course of these diseases. To address this, my lab uses cutting-edge genetic engineering to generate novel animal models of HSP in the fruit fly, in which we can study the molecular events underpinning this disorder. Recently, researchers in my lab have found that HSP-causing genes play a role in the organization of the endoplasmic reticulum (ER) network within motor neurons. The aim of this project will be to study how this impaired ER network contributes to neurodegeneration in motor neurons. You will learn various techniques associated with molecular genetics, confocal microscopic image analysis and the assay of behavioural readouts.

Lab website: fniamhy.wixsite.com/osullivanlab

Relevant majors: Pre-med, Biology

Billion Agents on my Laptop - High Performance Agent Based Simulation

Project Supervisor: Dr Vivek Nallur

A common problem with agent-based simulations is that the scale required for truly interesting emergent effects is difficult to achieve with consumer-level machines. This project will investigate high-performance and single-file deployment tools such as Redbean to create a high-performance agent-based simulation. Advanced: The simulation can be visualized on a typical-laptop hosted web server and browser [e.g. similar to https://py.cafe/app/tpike3/boltzmann-wealth-model ]

Desired qualifications: this is an exploratory, programming-heavy project that requires a high degree of comfort with installing and experimenting with different languages (PHP, Lua) and tools (Redbean server, Fullmoon framework). It also requires programming skill, and understanding of basic agent-based simulation

Relevant majors: Computer Science

Using TinyLLMs as Decision-Makers for Agent-Based Simulation

Project Supervisor: Dr Vivek Nallur

The use of contextual decision-making is an emerging idea in agent-based simulation. However, since multiple agents need to make decisions, they cannot all use the same LLM. This project will attempt to create an agent-based simulation, where each agent uses a tiny LLM (an LLM that can use consumer hardware) to decide on its next course of action.

Desired qualifications: This is a development-heavy project that requires the student to be comfortable with installing and experimenting with new tools. The final product should look something like: https://py.cafe/app/tpike3/boltzmann-wealth-model. The big difference, of course, is that the agents in the simulation consult different LLMs to decide what to do at every step.

Relevant majors: Computer Science

Online Asynchronous Games for Ethical Dilemmas

Project Supervisor: Dr Vivek Nallur

This is an interdisciplinary development project, i.e., while technical development will require programming skills, the topic is about humanitarian action. Humanitarian action relies on negotiations with powerful actors for access and programming at local, national and international levels. These negotiations are filled with ethical dilemmas like cooperating with war criminals and repressive regimes, favoring certain social and political groups, or putting the lives of local staff at risk in order to save lives and alleviate suffering Project Goal: Develop asynchronously-played games, that allow users (practitioners) to 'play' through real dilemmas they face on the field. The objectives of the game(s) are: a) To introduce the practitioner to a range of ethical dilemmas b) Understand the basic principles of negotiation c) Reflect on what criteria is used for their decision-making process

Desired qualifications: Web-based development (Javascript/Python/ WASM)

Relevant majors: Computer Science

Extended Reality-Based Poisoning Attacks and Development of Defenses

Project Supervisor: Dr. Madhusanka Liyanage

With the increasing adoption of Extended Reality (XR) technologies for real-world applications, facial recognition has emerged as a key use case in VR environments for identity verification, personalized user experiences, and interactive applications. However, training facial recognition models using approaches like distributed Federated Learning (FL) on VR devices introduces unique privacy and security vulnerabilities. Poisoning attacks, where adversaries inject malicious data into the training process, pose a critical threat to the integrity and performance of these models. For instance, attackers can manipulate the VR facial recognition model to misidentify individuals or create backdoors, compromising the user experience and privacy.

In this project, we propose to create a simulated poisoning attack scenario on a VR-based facial recognition use case to showcase the real-world implications of such threats. A VR facial recognition application will be developed to train and deploy models on local VR headsets, and the poisoning attack will be implemented by introducing adversarial data into the system. We will analyze the impact of these attacks on model performance, including misclassification rates and privacy breaches, and assess the feasibility of implementing real time defense mechanisms. This work aims to highlight the importance of secure FL practices and provide practical solutions to mitigate poisoning attacks in VR-based ML model training. Research Lab: https://netslab.ucd.ie/

Poisoning attacks on VR-based application of a target user

Key Learning Outcomes:

Understanding the mechanics and impact of poisoning attacks on FL-based facial recognition in VR environments.
Developing a VR facial recognition application to simulate and analyze poisoning attack scenarios.
Implementing and testing defense mechanisms to mitigate adversarial threats in FL pipelines.
Gaining insights into securing ML models in VR applications while balancing performance and privacy.

References:

[1] Sandeepa, C., Wang, S. and Liyanage, M., 2023, June. Privacy of the Metaverse: Current Issues, AI Attacks, and Possible Solutions. In 2023 IEEE International Conference on Metaverse Computing, Networking and Applications (MetaCom) (pp. 234-241). IEEE.

Relevant majors: Computer Science

Enhancing Privacy and Compliance in AI: An LLM-Driven Recommender Application for Machine Learning Models

Project Supervisor: Dr. Madhusanka Liyanage

Research Lab: https://netslab.ucd.ie/

The rapid advancement of Machine Learning (ML) and its integration into diverse applications have amplified concerns about privacy and compliance with regulatory frameworks. Large Language Models (LLMs) present an opportunity to address these challenges by providing intelligent insights into privacy-preserving practices. ML modesl often face vulnerabilities such as data memorization and unintended pattern retention, which can lead to privacy leaks and non-compliance with frameworks like the EU AI Act [1]. For instance, improperly analyzed models may inadvertently expose sensitive user information, making them susceptible to misuse or regulatory scrutiny. To tackle these challenges, we propose the development of an LLM-based Privacy Recommender System that leverages fine-tuned LLMs trained on regulatory texts and privacy evaluation metrics. This system will provide actionable recommendations for improving ML model privacy and ensuring alignment with the EU AI Act. Additionally, privacy risk assessments will be conducted to identify vulnerabilities, and cross-verification techniques will be employed to validate recommendations against real-world scenarios. This project aims to bridge the gap between technical privacy solutions and regulatory compliance, empowering organizations to deploy ML models responsibly and securely.

Figure 01: LLM training and report generation for the analyzed model

Key Learning Outcomes:

Identification and assessment of privacy vulnerabilities in ML models, particularly those related to data memorization and regulatory non-compliance.
Practical implementation of privacy-preserving techniques such as differential privacy and adversarial testing tailored for LLM-based systems.
Evaluation of trade-offs between privacy protection measures, system performance, and regulatory alignment.
Hands-on experience with fine-tuning LLMs for specific regulatory texts such as the EU AI Act.
Validation and benchmarking of privacy recommendations against real-world privacy risks and compliance standards.

Resources Required:

eGPU for LLM Fine-Tuning: For efficient fine-tuning of large language models on high performance computing tasks as an extension of available computing resources.

References

[1] European Commission (n.d.) Regulatory framework on Artificial Intelligence. Available at: https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai

Relevant majors: Computer Science

Virtual Reality-Based Digital Twin for Real-Time Smart Vehicle Simulation Research

Project Supervisor: Dr. Madhusanka Liyanage

Lab: https://netslab.ucd.ie/

The convergence of Virtual Reality (VR), sensor technologies, and advanced networking like 5G has opened new possibilities for real-time digital twin[1] applications in smart transportation. This project aims to develop a VR-based digital twin for simulating and interacting with smart vehicles in real time. Using an RC car equipped with multiple sensor nodes, we will create a virtual environment that replicates the car’s movements, sensor readings, and interactions with its surroundings. The synchronization between the physical RC car and its VR counterpart will be achieved over a 5G network testbed, ensuring low latency and high-speed communication. The VR application will provide an immersive platform for analyzing vehicle behavior, testing sensor integrations, and exploring real-time decision-making scenarios, with potential applications in autonomous vehicle development, intelligent transportation systems, and 5G IoT ecosystems. This project will bridge the gap between physical and virtual environments, enabling innovative approaches to smart vehicle testing and simulation.

Figure 01:

VR-based digital twin for smart vehicles[1]

Key Learning Outcomes:

Understanding the integration of sensor data and VR applications for real-time digital twin development.
Gaining hands-on experience in using 5G networks for low-latency synchronization between physical devices and virtual environments.
Developing and testing a VR-based simulation platform for smart vehicle behavior analysis and decision-making.
Exploring applications of digital twins in autonomous systems, intelligent transportation, and IoT-enabled environments.

References:

[1] Bhatti, G., Mohan, H. and Singh, R.R., 2021. Towards the future of smart electric vehicles: Digital twin technology. Renewable and Sustainable Energy Reviews, 141, p.110801.

Relevant majors: Computer Science

Optimizing Greenhouse Gas Monitoring and Prediction through Deep Learning

Project Supervisor: Dr Soumyabrata Dev

Greenhouse gas (GHG) emissions are one of the most pressing challenges facing humanity today, contributing significantly to climate change. Advances in machine learning and artificial intelligence provide transformative opportunities [1] for improving the monitoring and prediction of GHG emissions. This project seeks to address a key gap in existing research: evaluating the predictive capabilities of emissions data without dependence on external economic indicators.

You will explore a range of machine learning techniques, from traditional models like Decision Trees, Random Forests, and XGBoost, to advanced deep learning approaches [2] such as LSTMs, BiLSTMs, and ARIMA. Through comparative analysis, you will assess how regional variations in datasets influence model performance. Additionally, the project will emphasize enhancing temporal feature engineering to boost prediction accuracy. Using Bayesian optimization, you will fine tune hyperparameters to achieve optimal model performance and gain insights into how these adjustments amplify predictive capabilities.

This internship offers hands-on experience at the intersection of data science, environmental sustainability, and advanced computational methods. By the end of the project, you will contribute to building more robust models for understanding and mitigating GHG emissions.

Reference:

[1] Y. Zhang, A. Pakrashi, and S. Dev, Assessing Interconnected Factors in CO2 Emissions: A Case Study of India Using Principal Component Analysis, Proc. IEEE Conference on Energy Internet and Energy System Integration, 2023. [2] P. Dey, S. Dev, and B. S. Phelan, BiLSTM-BiGRU: A fusion deep neural network for predicting air pollutant concentration, IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 2023.

Relevant majors: Data Science, Software Engineering, Statistics, Environmental Science

Adapting Facial Expression Recognition Systems to Cultural Variations in Emotion Representation

Project Supervisor: Dr Soumyabrata Dev

Facial Expression Recognition (FER) systems play a critical role in applications ranging from healthcare to human-computer interaction. However, their accuracy often falters when applied across diverse cultural contexts, as emotional display norms and facial feature emphasis can vary significantly between cultures. This project aims to bridge this gap by investigating how FER systems can be improved to better account for these cultural variations, ensuring fairness and inclusivity in emotion detection.

By analyzing facial images from diverse cultural backgrounds, you will identify key differences in emotion representation and their impact on FER model performance. Leveraging these insights, the project will develop adaptive methods to enhance model accuracy for culturally varied datasets. This includes experimenting with advanced deep learning architectures and integrating culturally sensitive training techniques to reduce bias. The ultimate goal is to create a system that not only recognizes emotions with high precision but also respects and adapts to the unique ways emotions are expressed across different regions.

Research Articles:

Karnati, M., Seal, A., Bhattacharjee, D., Yazidi, A., & Krejcar, O. (2023). Understanding deep learning techniques for recognition of human emotions using facial expressions: A comprehensive survey. IEEE Transactions on Instrumentation and Measurement, 72.

Sample Datasets:

AffectNet: A large-scale database for facial expression, valence, and arousal computing in the wild. (Mollahosseini, A., Hasani, B., & Mahoor, M. H., 2017).

JAFFE: Japanese Female Facial Expression Database.

Relevant majors: Computer Science, Data Science, Psychology

Collaborative Multi-Agent SLAM for Large-Scale 3D Mapping

Project Supervisor: Dr Soumyabrata Dev

Simultaneous Localization and Mapping (SLAM) systems [1] have shown remarkable success in mapping small-scale environments with single-agent setups. However, their limitations become apparent as the scale of mapping expands to large, complex areas. This project aims to overcome these challenges by developing a collaborative multi agent SLAM system [2] that enables multiple agents to work together, efficiently and accurately mapping large-scale 3D environments.

In this system, agents will utilize advanced techniques like inter-agent communication, data fusion, map merging, and loop closure detection to enhance performance. A key focus will be on resolving challenges in data fusion [3], ensuring that agents can correctly recognize and integrate overlapping scenes without sacrificing accuracy or computational efficiency. Cutting-edge decentralized frameworks, such as Swarm-SLAM [4], will be explored to improve scalability and reduce data synchronization delays. Participants will actively participate in designing, developing, and testing the SLAM models, with the final system evaluated using large-scale datasets like the Waymo Open Dataset[5] and field tests to validate real world applicability.

This project has significant applications in fields like search and rescue, environmental reconstruction, smart cities, and digital twins, where precise and rapid mapping is essential. Interns will gain hands-on experience with the latest advancements in SLAM, contributing to the development of state-of-the-art solutions in collaborative 3D mapping.

[1] Alsadik, Bashar, and Samer Karam. “The simultaneous localization and mapping (SLAM): An overview.”; Surveying and geospatial engineering journal 1.2 (2021): 1-12.

[2] Zou, Danping, Ping Tan, and Wenxian Yu. “Collaborative visual SLAM for multiple agents: A brief survey”; Virtual Reality & Intelligent Hardware 1.5 (2019): 461-482.

[3] Lowry, Stephanie, et al. “Visual place recognition: A survey.”; ieee transactions on robotics 32.1 (2015): 1-19.

[4] Lajoie, Pierre-Yves, and Giovanni Beltrame. “Swarm-slam: Sparse decentralized collaborative simultaneous localization and mapping framework for multi-robot systems”; IEEE Robotics and Automation Letters 9.1 (2023): 475-482.

[5] Sun, Pei, et al. “Scalability in perception for autonomous driving: Waymo open dataset” Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2020.

Relevant majors: Computer Science, Data Science

Efficient Deep Learning for Solar Forecasting

Project Supervisor: Dr Soumyabrata Dev

Solar energy, the most abundant and sustainable natural resource, is being rapidly adopted worldwide due to concerns about fossil fuel depletion and climate change [1]. However, its intermittent nature poses significant challenges to maintaining stable power grid operations. Accurate and timely solar irradiance forecasting is crucial for optimizing solar power generation and enhancing energy planning efficiency. While deep learning techniques have demonstrated superior performance over traditional numerical weather prediction models, their high computational demands limit practicality for large-scale or real-time deployment, especially on resource constrained devices.

This project aims to develop efficient, lightweight deep learning models [2] for solar irradiance forecasting by employing model compression techniques such as knowledge distillation, pruning, and quantization. These methods will reduce computational requirements, enabling faster inference, lower costs, and scalable deployment on IoT devices and smart energy grids [3]. The outcome will facilitate responsive and privacy-aware energy solutions, ensuring greater adaptability and stability in renewable energy systems.

Key Tasks:

- Develop a lightweight deep learning model or optimize an existing one using model compression techniques.

- Apply methods like knowledge distillation, pruning, and quantization to improve computational efficiency without sacrificing accuracy.

- Evaluate the model's performance against baseline models in terms of accuracy, computational cost, and scalability.

Participants will gain hands-on experience with cutting-edge deep learning optimization techniques and practical AI deployment. The project offers an excellent opportunity to contribute to impactful research in renewable energy while enhancing skills in efficient architecture design, model compression, and edge computing integration.

Key References:

Nijhum, I. R., Kaloni, D., Kenny, P., & Dev, S. (2024, July). Enhancing Intra-Hour Solar Irradiance Estimation through Knowledge Distillation and Infrared Sky Images. IGARSS 2024.
Menghani, G. (2023). Efficient deep learning: A survey on making deep learning models smaller, faster, and better. ACM Computing Surveys.
Zhang, Y., et al. (2024). A new lightweight framework based on knowledge distillation for reducing the complexity of multi-modal solar irradiance prediction model. Journal of Cleaner Production, 475.

Relevant majors: Computer Science, Electrical Engineering, Data Science

Predicting Nearshore Wave Heights Using SAR, Scatterometer, and Buoy Data

Project Supervisor: Dr Soumyabrata Dev

Accurate prediction of nearshore wave heights is vital for coastal activities like shipping, fishing, offshore energy planning, and hazard management, especially along Ireland’s coastline. This project focuses on developing a machine learning model that integrates data from Synthetic Aperture Radar (SAR), scatterometer wind measurements, and buoy observations. By combining these diverse datasets, the project aims to improve wave height prediction accuracy [1], supporting safer and more efficient coastal operations.

The project involves preprocessing SAR data to extract wave-induced surface roughness, scatterometer data for wind speed measurements, and buoy data for ground truth validation. Advanced machine learning techniques, such as temporal transformers and graph neural networks, will be used to develop a predictive model. The project will also evaluate the benefits of integrating these datasets compared to using them individually. The model’s accuracy and reliability will be validated against buoy measurements to ensure its robustness.

Participants will gain hands-on experience with remote sensing and environmental datasets, learning essential skills in data processing, feature extraction, and applying machine learning to real-world challenges. The project provides an opportunity to explore state-of-the-art algorithms, develop expertise in coastal modeling, and contribute to improving wave height prediction models [2]. The findings will support safer navigation, efficient energy planning, and enhanced coastal hazard management.

References:

Wang, H., Yang, J., Zhu, J., Ren, L., Liu, Y., Li, W., & Chen, C. (2021). Estimation of significant wave heights from ASCAT scatterometer data via deep learning network. Remote Sensing, 13(2), 195.
Pramudya, F. S., Pan, J., & Devlin, A. T. (2019). Estimation of significant wave height of near-range traveling ocean waves using Sentinel-1 SAR images. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 12(4), 1067-1075.

Relevant majors: Data Science, Computer Science, Environmental Engineering, Oceanography

Accurate Weather Forecasting Model Based on ERA5 Reanalysis Dataset

Project Supervisor: Dr Soumyabrata Dev

Accurate weather forecasting is critical for addressing societal challenges, including disaster management, agricultural planning, and climate resilience. This project focuses on developing a high-performance weather forecasting system to predict rainfall and temperature with faster training speeds and greater accuracy. By leveraging the ERA5 reanalysis dataset, provided by the European Centre for Medium-Range Weather Forecasts (ECMWF), the project aims to model spatiotemporal meteorological data and effectively capture complex atmospheric dynamics.

The ERA5 dataset represents a state-of-the-art global reanalysis system that integrates historical observational data with numerical weather prediction (NWP) outputs, offering comprehensive records of atmospheric, oceanic, and land conditions from 1950 onward. Using this dataset, the project seeks to address the limitations of traditional NWP methods, including their high computational costs and inefficiencies in modeling extreme weather events. Advanced machine learning techniques will be applied to enhance the forecasting model’s precision and scalability.

Participants will preprocess large-scale meteorological data, design and train innovative predictive models, and analyze meaningful weather patterns. The findings are expected to improve decision-making for disaster management, agricultural planning, and climate adaptation efforts. The intern will gain valuable hands-on experience with large-scale data processing, advanced machine learning, and practical applications in weather forecasting [1].

The participant will explore cutting-edge techniques in weather modeling, including preprocessing reanalysis datasets, implementing scalable machine learning architectures, and interpreting results for real-world applications. This project offers a unique chance to contribute to innovations in weather and climate forecasting, addressing critical global challenges.

References:

Y. Li, N. Akrami, and S. Dev, “Harnessing ERA5 Reanalysis Data for Improved Long- Term Rainfall Forecasting in Southern Iran”; 2023 IEEE 7th Conference on Energy Internet and Energy System Integration (EI2), Hangzhou, China, 2023, pp. 2045-2049.

Relevant majors: Computer Science, Data Science, Meteorology

Using Satellite Images for Rainfall Prediction in Ireland

Project Supervisor: Dr Soumyabrata Dev

Accurate rainfall prediction is crucial for effective water resource management, agriculture, and disaster preparedness. In Ireland, traditional methods relying on ground-based measurements and numerical weather prediction models often struggle to account for the country’s complex weather patterns and varied topography. This project aims to enhance rainfall prediction accuracy by leveraging satellite imagery and advanced machine learning techniques.

Satellite images capture comprehensive atmospheric information over large areas, providing valuable insights into precipitation patterns. By analyzing historical satellite data alongside corresponding rainfall measurements, machine learning models can be trained to identify visual features linked to different rainfall levels [1]. These models will then predict future rainfall based on new satellite imagery, offering a scalable and data-driven approach to weather forecasting.

The project holds significant potential to improve the timeliness and accuracy of rainfall predictions in Ireland, benefiting stakeholders such as farmers, water resource managers, and disaster preparedness agencies. The intern will engage in preprocessing satellite data [2], developing and training predictive models, and validating model performance against ground-based measurements.

The participant will gain hands-on experience in working with satellite data, applying machine learning techniques, and addressing real-world environmental challenges. This project offers a unique opportunity to contribute to the development of innovative rainfall forecasting methods, enhancing practical skills in data science, remote sensing, and environmental modeling.

References:

[1] O. M. Pasaribu, A. Poniman, A. A. Lestari, Y. Prihanto, A. A. Supriyadi and Y. Darmawan, “Exploration of CHIRPS Satellite Data as Rainfall Estimation Data in Medan City and Deli Serdang Regency”; 2022 IEEE Asia-Pacific Conference on Geoscience, Electronics and RemoteSensing Technology (AGERS), Surabaya, Indonesia, 2022, pp. 141-145.

[2] S. Vasavi, P. V. S. Krishna, P. D. L. N. Sri, M. Navena and C. HariKiran, “Rainfall Estimation From Satellite Images Using Cloud Classifications”; 2022 IEEE North Karnataka Subsection Flagship International Conference (NKCon), Vijaypur, India, 2022, pp. 1-5.

Relevant majors: Data Science, Meteorology, Environmental Engineering, Computer Science

Multi-Scale Fusion of Near-Infrared Images for Optimized NDVI Calculation

Project Supervisor: Dr Soumyabrata Dev

Infrared imaging plays a crucial role in environmental monitoring and precision agriculture, particularly for assessing vegetation health through metrics like the Normalized Difference Vegetation Index (NDVI). However, acquiring high-resolution infrared images is often constrained by sensor limitations and high costs. This project aims to address these challenges by developing a methodology to upscale low resolution near-infrared (NIR) images and fuse them with existing data to enhance NDVI calculation accuracy.

The project involves using a hybrid approach to upscale NIR images, combining traditional interpolation techniques with advanced deep learning-based methods. These enhanced high-resolution images will be fused with lower-resolution data to generate multi-scale composite images, improving the reliability of vegetation health assessments. Studies have demonstrated that fusing images of different modalities, such as RGB and NIR, significantly enhances analytical outcomes [1, 2]. This research builds on this principle, integrating multi-scale image data to advance the utility of infrared imaging for precise and reliable environmental applications. The participant will have the opportunity to work on image preprocessing, apply deep learning algorithms for image upscaling, and evaluate the comparative performance of various methodologies. The project’s findings will contribute to advancing the application of infrared imaging in agriculture and environmental monitoring, with significant implications for cost-effective and scalable solutions.

The participant will gain hands-on experience in cutting-edge image processing and deep learning techniques, as well as practical skills in Python programming and data fusion. This project offers a unique opportunity to apply technical skills to impactful environmental and agricultural challenges.

References:

Aslahishahri, Masoomeh, et al. “From RGB to NIR: Predicting near-infrared reflectance from visible spectrum aerial images of crops”; Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021.
Liu, Zhihao, et al. “Improved kiwifruit detection using pre-trained VGG16 with RGB and NIR information fusion”; IEEE Access, 8 (2019): 2327-2336.

Relevant majors: Computer Science, Data Science, Environmental Engineering

Innovative Deep Learning Approaches for Water Body and Quality Classification

Project Supervisor: Dr Soumyabrata Dev

Water is a vital resource, yet its quality and availability are increasingly affected by human activities. This project aims to enhance the classification of water bodies and evaluate their quality by leveraging hyperspectral satellite imagery [1] and UAV (Unmanned Aerial Vehicle) data. Using advanced deep learning models, the study seeks to improve the identification and mapping of water bodies while assessing key quality indicators such as turbidity, algal blooms, and color changes indicative of harmful conditions.

The methodology involves analyzing remote sensing imagery with state-of-the-art deep learning techniques to differentiate water bodies [2] from other land features and detect physical and biological characteristics of water. The proposed models will be benchmarked against existing classification techniques to evaluate performance and identify areas of improvement. Anticipated outcomes include more accurate water body classification, detailed mapping of water quality parameters, and improved tools for environmental monitoring and management.

This research offers valuable contributions to sustainable water management, supporting informed decision-making in environmental policy and resource management. The participant will have the opportunity to work on cutting-edge machine learning models, process satellite and UAV imagery, and tackle real-world challenges in environmental sustainability.

Learning Opportunities:

The participant will gain expertise in satellite image processing, deep learning applications, and the integration of remote sensing technologies. The project provides hands-on experience with Python programming, big data handling, and advanced classification techniques, offering an excellent platform for career development in environmental monitoring [3] and machine learning.

References:

S. Kumari, J. Wu, D. Ayala-Cabrera, and S. Dev, “Effective Water Body Extraction from Hyperspectral Data: A Focus on Unsupervised Band” 2024 IEEE 9th International Conference on Computational Intelligence and Applications (ICCIA), Haikou, China, 2024, pp. 251-255, doi: 10.1109/ICCIA62557.2024.10719176.
Cao, H., Tian, Y., Liu, Y. et al. “Water body extraction from high spatial resolution remote sensing images based on enhanced U-Net and multi-scale information” Scientific Reports 14, 16132 (2024). https://doi.org/10.1038/s41598-024-67113-7 3. Rahul, T.S., Brema, J. “Assessment of water quality parameters in Muthupet estuary using hyperspectral PRISMA satellite and multispectral images” Environmental Monitoring and Assessment 195, 880 (2023). https://doi.org/10.1007/s10661-023-11497-y

Relevant majors: Computer Science,Environmental Science, Data Science

Spatiotemporal Ground Movement Estimation System Using Machine Learning Models and Sentinel-1 Data Classification

Project Supervisor: Dr Soumyabrata Dev

Ground displacement poses significant risks to infrastructure, especially in the context of rapid urbanization. Understanding and predicting ground movement is crucial for mitigating these risks and ensuring infrastructure stability. This project focuses on developing a ground displacement estimation system by leveraging Interferometric Synthetic Aperture Radar (InSAR) datasets and Machine Learning (ML) models.

InSAR technology provides highly accurate time-series data of ground movement [1], while machine learning has proven effective for prediction and pattern recognition tasks [2].

The project involves processing geographic imagery captured by Sentinel-1 satellites over Ireland and other selected European regions. The intern will learn the fundamentals of remote sensing, including how geographic images are acquired and processed using InSAR techniques. Once the time-series datasets are generated, advanced machine learning models, such as CNN-LSTM, LightGBM, and Graph Convolutional Networks (GCNs), will be applied to train the data and evaluate model performance. The ultimate goal is to create an accurate estimation system capable of predicting ground displacement trends in the targeted regions.

Learning Opportunities:

The participant will gain hands-on experience in:

- Processing remote sensing data using InSAR.

- Training and evaluating machine learning models for spatiotemporal data. - Applying Sentinel-1 satellite data to real-world geospatial challenges. The project provides an excellent platform to develop interdisciplinary skills at the intersection of geographic information systems (GIS), remote sensing, and machine learning.

References:

A. Hrysiewicz, A. Trafford, and S. Azadnejad, “SAR supported by geophysical and geotechnical information constrains two-dimensional motion of a railway embankment constructed on peat” Engineering Geology, 2024.
B. Li, K. Liu, M. Wang, and Y. Wang, “InSAR time-series deformation forecasting surrounding Salt Lake using deep transformer models”; Science of The Total Environment, 2022.

Relevant majors: Computer Science, Data Science

Sustainable Continuous Chemical Synthesis of Bioactives Inspired by Nature

Project Supervisor: Assoc. Prof. Marcus Baumann

We wish to offer this unique summer research project that will provide a talented student with the opportunity to acquire important research skills in a project at the interface of organic chemistry, sustainability, biology and chemical engineering. The synthesis and spectroscopic characterization of a small collection of drug-like building blocks will be targeted with a specific emphasis on exploiting light as a tuneable and sustainable energy source. This will involve bespoke continuous flow reactors equipped with efficient light-emitting diodes (LEDs) available in our lab. Light in the UV-A and visible range of the spectrum will be used to functionalise the molecules studied which will contribute to greener chemical reactions. Continuous flow chemistry (A field guide to flow chemistry for synthetic organic chemists - Chemical Science (RSC Publishing)) is a novel and exciting addition to the chemist’s toolbox that allows to perform chemistry in a safer, more effective and sustainable manner yielding readily scaled and automated processes that are highly desirable in academia and industry alike. Flow chemistry has been identified as one of the 10 emerging technologies that will change the world: IUPAC announces the top ten emerging technologies in chemistry - IUPAC | International Union of Pure and Applied Chemistry. This approach will avoid the isolation of unstable molecules and provides a powerful and streamlined route into important bioactive structures that will be studied by biologists and medicinal chemists. The ability to automate such processes and couple them with AI and machine learning is highly advantageous as it circumvents tedious downstream processing to directly give clean products. The successful student will be embedded in our international research group (currently 12 PhDs, 2 postdocs) and gain new skills in chemical synthesis, purification, spectroscopic characterisation as well as the use of modern flow reactor technology.

For an example of a past project that was subsequently published, please see here.

Relevant majors: Chemistry, Chemical Engineering, Biology

Glycoconjugate Metal Complexes for Sensing Bacterial Proteins

Project Supervisor: Dr Joseph Byrne

Many pathogens, such as bacteria P. aeruginosa and E. coli or fungus C. albicans, produce characteristic carbohydrate-binding proteins. These pathogens are designated as ‘Critical Priority’ by the WHO, requiring new targeting strategies. We have recently shown that luminescent glycoconjugate–lanthanide complexes can detect this class of proteins by characteristic changes in luminescence. You will expand the family of sensors by synthesizing novel molecules, learning skills in carbohydrate synthesis, coordination chemistry and spectroscopy. This project aims to find more efficient sensors. Protein sensing and/or antimicrobial behavior of these materials will be probed and characterized either in the School of Chemistry, or with collaborators.

Further reading: K Wojtczak, E Zahorska, IJ Murphy, F Koppel, G Cooke, A Titz, JP Byrne*, “Switch-onluminescent sensing of unlabelled bacterial lectin by terbium(III) glycoconjugate systems”, Chem.Commun., 2023, 59, 8384; doi:10.1039/D3CC02300A ; K Wojtczak et al. “Antiadhesive glycoconjugate metal complexes targeting pathogens Pseudomonas aeruginosa and Candida albicans”. ChemRxiv. 2024; doi:10.26434/chemrxiv 2024-wkdpf [pre-print])

Desired qualifications: Some experience with organic synthesis and purification required.

Relevant majors: Chemistry

Studies Towards the Synthesis of an Epoxy-Cinnamate Amino Acid

Project Supervisor: Dr Paul Evans

The skyllamycins are a family of complex cyclic peptide natural products that possess a cinnamate sidechain (see for example skyllamycin A in the figure below).1 A previously unreported version of skyllamycin (compound 1) has recently been isolated from the bacterium Streptomyces nodosus. 2 We have determined that this natural product contains an epoxide moiety in the cinnamate side-chain and are currently working on a project aimed at unpicking whether this compound possesses any useful biological activity – members of the skyllamycin family have been reported to possess antimicrobial properties.

Additionally, and in specific relation to this project, we have been working to determine the stereochemistry around the epoxide. To do this, we plan to prepare both the Z- and E alkenes 2 shown in Scheme 1 on the following page, and then separately convert them to their corresponding epoxides 3 and 5. For the Z-3 preliminary studies have demonstrated that Jacobsen’s catalyst 4 generates 3 with good enantioselectivity and that in NMR spectroscopy the methyl signals for the di0erent epoxide isomers, 3 and 5, are very different. From here we will couple the epoxy-cinnamate to L-threonine, the first amino acid in the cyclic central unit of the skyllamycins (Scheme 2). We believe that when we do this, we will be able to compare the NMR spectroscopic data for 6 with that recorded for epoxy-skyllamycin (1) and that this will help determine the stereochemistry of the natural product. In addition, we will compare our data with that reported for compound 7 (isolated recently from Streptomyces sp. HS-NF-1222A).3 Notably, in 7, the epoxide stereochemistry is also not determined, and our hypothesis is also that this work has incorrectly assigned the structure of this natural product. It is likely that an amide functional group attaches the cinnamate, as in 1, to the amino acid and not, as claimed, an ester.

References: (1) V. Schubert, F. Di Meo, P. -L. Saaidi, S. Bartoschek, H. -P. Fiedler, P. Trouillas, R. D. Süssmuth, Chem. Eur. J. 2014, 20, 4948-4955.

(2) Y. Song, J. A. Amaya, V. C. Murarka, H. Mendez, M. Hogan, J. Muldoon, P. Evans, Y. Ortin, S. L. Kelly, D. C. Lamb, T. L. Poulos, P. Ca0rey, Org. Biomol. Chem. 2024, 10.1039/d4ob00178h.

(3) H. Qi, Z. Ma, Z.-L. Xue, Z. Yu, Q.-Y. Xu, H. Zhang, X.-P. Yu, J.-D. Wang, Nat. Prod. Res. 2020, 34, 2080- 2085.

Relevant majors: Chemistry

Study of Molecular Probes of Non-Canonical DNA towards New Therapeutics

Project Supervisor: Prof. Susan Quinn

As the body’s information repository, DNA provides the essential role of programming all biological functions. This information is translated through numerous molecular interactions. Proteins and small molecules bind to DNA through a variety of modes that include groove binding, electrostatic interactions and intercalation. The nature of these interactions is influenced by structure of the molecules and the secondary structure of DNA, which in addition to the common B-DNA form can also adopt other non-canonical forms. One important structure is guanine rich quadruplex DNA. Quadruplex DNA comprise stacked tetrads of four guanine bases to form a unique DNA structure with loops at the edges, see below.

The presence of quadruplex structures in the cell has been confirmed by fluorescence microscopy. The binding interactions of small molecules to these structures is of interest due to their potential as therapeutic targets. The quadruplex structure is also of interest as it is formed from human telomere DNA. In this project we aim to investigate the binding of metal complexes to distinguish the different structures, which is essential to resolving their biological roles. The project allows for spectroscopic measurements and DNA binding studies which will provide experience across a range of diverse techniques including circular dichroism, UV visible and emission spectroscopy. This will build on our extensive experience in this field [Nature Chemistry 2015, 7, 961, Chem. Sci., 2017,8, 4705-4723, Chem. Commun., 2020,56, 9703-9706, Chem. Eur J. 2020, DOI: 10.1002/chem.202002165, J. Am. Chem. Soc. 2023, 145, 39, 21344–21360, doi.org/10.1021/jacs.3c06099]. If interested, students will also be given an opportunity to prepare new complexes.

Relevant majors: Chemistry

Development of Nanoparticle-Based Therapies

Project Supervisor: Prof. Susan Quinn

We are interested in preparing gold nanoparticle systems to complement radiotherapy treatment for resistant cancers [ACS Appl. Nano Mater. 2020, 3, 3157]. This relies on the fact that X-ray irradiation of AuNPs releases electrons that form reactive oxygen species that can kill cells [Sicard-Roselli, C. et al. Small 2014, 10, 3338–46] see (1) below. Nanoparticles have a high surface area which allows them to carry molecules that can bind to their surface. We have developed methods to prepare composite particles using polystyrene nanoparticles with high loadings of Au NPs and demonstrated uptake in cells [Chem. Commun., 2016,52, 14388- 14391] (see 2 below) and to enhance radiotherapy towards radiation resistant breast cancer cells [ACS Applied Nanomaterials 2020, 3, 4, 3157-3162]. In this project, the aim is to replace polystyrene spheres with tiny glass beads in the form of biocompatible silica (SiO2) nanoparticles to prepare gold nanoparticle composites. Silica particles also have the advantage that we can load their transparent structure with luminescent molecules to act like tiny light bulbs and allow them to be tracked in cells. Furthermore, over time silica are degraded by enzymes in the body. The low toxicity, high surface area and ease of functionalization of silica nanoparticles makes them attractive systems for cellular imaging, diagnostics and therapeutics. The project will involve metal nanoparticle synthesis, spectroscopic characterisation, composite preparation and characterisation as well as assaying the stability of the composites formed to biological media.

Relevant majors: Chemistry

Using Analog Experiments to Explore Volcanic Dome Growth

Project Supervisors: Dr Claire Harnett and Dr Eoghan Holohan

Volcanic domes form when viscous lava extrudes from a volcanic event and piles up rather than flowing away. These domes grow incrementally through time, with varied extrusion rates, material properties, and underlying pre-eruptive topography. Volcanic domes can therefore be highly unstable, and their collapse can lead to rock and debris avalanches, and pyroclastic density currents, that can devastate surrounding communities.

The participant in this project will work in the Analog Modeling Laboratory at UCD, where we use analogs of viscous lava (syrup-sugar suspension) and solid rock (sand-plaster mix) to simulate lava dome growth. Through high temporal resolution imaging of the experiments, the student will perform photogrammetric analysis and produce digital elevation models. This allows exploration of various factors controlling volcanic dome growth and hazard, such as fracturing, landsliding, and impact of material properties. During their time at UCD, the student will sit within an active Volcanology Group that meets weekly. This student project will contribute to active research on dome growth and collapse at UCD. The student will gain skills in data analysis and spatial data collection, photogrammetry, and experiment design.

Relevant majors: Earth Science, Environmental Science, Geography

An Atomic Force Microscopy Study of the Effect of a Novel Ionic Drug on Cancer Cell Morphology, Adhesion and Visco-Elasticity

Project Supervisor: Dr Antonio Benedetto

Live cells are the building blocks of life. Their viscosity and elasticity govern several vital processes, including cell division, adhesion, signaling, and migration.

Cancer cells' elastic profile has been shown to differ from that of their healthy counterparts. Restoring the cells’ healthy elasticity can lead to novel therapeutics and diagnostic approaches.

The project will focus on the effect of two selected drugs made of ionic liquids—a vast family of organic electrolytes with unique and tuneable properties—on the morphology, adhesion, elasticity, and viscosity of the model cancer cell line MCF-10A. The drug's effect on the morphology, adhesion, elasticity, and viscosity of the model normal/healthy cell line MCF-7 will also be measured for benchmarking.

This project will be part of an active research line in Benedetto’s Lab supported by Science Foundations Ireland (Fig. 1).

As part of the project, the student will be trained in basic cell biology, sample preparation of live cells, and the use of the cutting-edge bio-atomic force microscopy recently installed in Benedetto’s Lab.

Fig. 1 – Representative (a) topography, (b) elastic modulus, and (c) adhesion maps of a cancer cell obtained using the Lab bio-AFM.

Relevant majors: Physics, Biology, Chemistry

Novel Ionic Drugs Targeting the Morphology & Elasticity of Amyloid Fibrils

Project Supervisor: Dr Antonio Benedetto

Proteins are the molecular machines of life. They carry out various biochemical functions in cells and are exploited in many nano-technological applications, including the new protein based COVID-19 vaccines. Under specific conditions, proteins undergo structural transformations leading to the formation of aggregated structures known as amyloid fibrils. Amyloid fibrils have been observed in several pathological conditions, including Alzheimer’s and Parkinson’s diseases, and have been exploited as advanced nano-materials in biomedicine, tissue engineering, etc.

This project will focus on measuring, using atomic force microscopy (AFM), the effect of two selected drugs made by ionic liquids—a vast family of organic electrolytes with unique and tuneable properties—on the structure, morphology, and viscoelasticity of amyloid fibrils. A set of first results presented at the European Biophysical Conference EBSA 2021 in Vienna and published in 2022 in the Journal of Physical Chemistry Letters provide the starting point and feasibility proof of the proposed project (Fig. 1).

As part of the project, the student will be trained in sample preparation of amyloid fibrils and in the use of a cutting-edge bio-AFM recently installed in Benedetto’s Lab.

Fig. 1 – Representative height AFM (a) 3D images, (b) zoom 3D images, and (c) histogram distributions of lysozyme amyloid fibrils incubated in water (black, for control) and water solutions of two drugs, i.e. EAN (green) and TMGA (red), at a molar ratio of 3.5 drug molecules per protein.

Relevant majors: Physics, Biology, Chemistry

What Can Collagen’s Piezoelectricity Reveal About Health and Aging?

Project Supervisor: Prof. Brian Rodriguez

Piezoelectric materials are all around you, in microphones, lighters, watches, and ultrasound transducers, but they are also in you! Collagen is the most abundant protein in the human body, and it is piezoelectric - the protein generates an electric charge under mechanical stress. Studying collagen piezoelectricity can lead to biomedical applications and provide a deeper insight into biological processes, explaining for instance why astronauts lose bone density.

Structural differences in collagen, such as those observed in conditions like osteogenesis imperfecta (where collagen contains three alpha-1 chains instead of the typical two alpha-1 and one alpha-2 chains found in healthy collagen), significantly influence its mechanical and electromechanical properties. This project involves characterizing the nanoscale piezoresponse of collagen and comparing it across samples sourced from healthy, unhealthy, and aging tissues.

As part of the Nanoscale Function Group at University College Dublin, the participant will collaborate with a multidisciplinary team driving innovations in biology and nanotechnology. You will learn a key nanotechnology-enabling tool called atomic force microscopy and specialized electrical modes such as piezoresponse force microscopy. You will be involved in sample preparation, data collection, and analysis – you will quantify collagen’s piezoelectric properties and evaluate how these responses vary with health conditions, offering insights into the relationship between structural differences and electromechanical behavior. This opportunity provides hands-on experience with advanced techniques and builds invaluable skills in nanotechnology and biomedicine. The project can be tailored to your interests, across a wide range, for example, from science communication/outreach to machine learning.

Relevant majors: Physics, Biology, Pre-Med, Engineering

Investigating the Phase-Properties Relationship of Ni(OH)2 Nanoparticles on Ni Foam for Wastewater Electrolysis

Project Supervisors: Prof. Brian Rodriguez, (School of Physics) and Dr Veronica Sofianos (School of Chemical and Bioprocess Engineering)

Nanomaterials are increasingly being used in environmental and energy sectors for a range of applications including wastewater electrolysis, catalysis, and green hydrogen production. Their high surface area and catalytic efficiency and their tunable physical and chemical properties allow them to be tailored for specific applications.

To optimize the performance of these materials, they need to be characterized at the scale of the film or electrode they are deposited as down to the length scale of the nanomaterials themselves. Atomic force microscopy (AFM) is a particularly attractive method for this characterization, as it provides direct, spatially resolved measurements of local electrical properties such as conductivity and surface potential, along with roughness and mechanical properties. This detailed characterization is essential for tailoring nanomaterials for renewable energy, environmental remediation, and sustainable industrial processes. In this project you will characterise Ni(OH)2 on Ni foams by AFM to visualize the materials and to measure their roughness and conductivity. You will learn a key nanotechnology enabling tool called atomic force microscopy and specialized electrical modes. You will be involved in sample preparation, data collection, and analysis. This opportunity provides hands-on experience with advanced techniques and builds invaluable skills in nanotechnology and materials science.

Relevant majors: Physics, Engineering

Molecular Modeling of Conformational Dynamics of Pigments from the Light Harvesting Complex LHC2

Project Supervisor: Assoc. Prof. Nicolae-Viorel (Vio) Buchete

Photosystem II (PS-II) is a key complex found in Photosynthetic organisms and is the first link in the chain of the photosynthetic machinery. It is located in the thylakoid membrane within the chloroplasts. LHC2 is a major molecular antenna found in PSII and the most abundant integral membrane protein in the chloroplasts. LHC2 not only performs essential light harvesting functions but also exhibits photoprotective abilities under high-light conditions. LHC2 is comprised of a protein core bonded with 18 pigments, including 6 Chlorophyll As, 8 Chlorophyll Bs and 4 Xanthophylls: 2 Luteins, Violaxanthin and Neoxanthin.

In this project, we analyze existing data and perform new Molecular Dynamic (MD) simulations of the LHC2 system in a POPC lipid membrane bilayer. By understanding the dynamics of the protein-bilayer system, and especially of the Violaxanthin and Neoxanthin pigments, we can advance the understanding of LCH2 molecular mechanisms, including its complex energy dissipation pathways, and shed new light on the rather poor understood but crucial photoprotective abilities (i.e., their response under increased light conditions (over-excitation). The student will learn molecular dynamics (MD) modeling techniques (NAMD and VMD packages) and use them on Linux-based high performance computational resources to generate and/or analyze MD data. If interested, the student can apply Markov State Modelling techniques to characterize the system’s dynamics.

Desired qualifications: previous familiarity with proteins, molecular modeling (especially molecular dynamics) and data analysis in a Linux environment would be very useful but not required.

Relevant majors : Physics, Chemical Engineering

Molecular Modeling of Conformational Dynamics of the Polycystin-1 (PC1) Protein in Kidney Disease

Project Supervisor: Assoc. Prof. Nicolae-Viorel (Vio) Buchete

Autosomal dominant polycystic kidney disease (ADPKD) is a common genetic disorder characterized by numerous cysts in the kidneys, leading to progressive renal dysfunction and eventual failure. The disease primarily results from mutations in the PKD1 and PKD2 genes, with the PKD1 gene encoding polycystin-1 (PC1), a protein crucial for renal epithelial cell function and mechanosensation. Understanding PC1's structure and function is essential for elucidating the molecular mechanisms of ADPKD and developing targeted therapies. We have already employed homology modeling (HM) and molecular dynamics (MD) simulations to investigate the structural dynamics and interactions of PC1. The student will perform new MD simulations and analyse new data as needed to shed new light on the PC1 conformational dynamics in the membrane and its underlying molecular mechanisms. The student will learn how to use packages such as NAMDand VMD and have access to Linux-based high-performance computational resources to generate and/or analyze MD data on PC1. Overall, this study offers valuable insights into the structural dynamics of PC1 and its role in ADPKD, laying the groundwork for future research and the development of targeted therapies.

Desired qualifications; previous familiarity with proteins, molecular modeling (especially molecular dynamics) and data analysis in a Linux environment would be very useful but not required.

Relevant majors : Physics Chemical Engineering

A 3D Cell Model for Toxicity Testing Using Advanced Fluorescence Imaging

Project Supervisor: Professor Jeremy Simpson

The development of new therapeutics requires that they are first evaluated using a series of in vitro cell-based assays to assess their efficacy and toxicity. Imaging approaches to measure toxicity, including metabolic activity and ultimately cell death, are powerful, as they provide visual and quantitative data with respect to how cells respond to a drug. Traditionally these cell-based assays have been carried out using cells growing as monolayers, as they are relatively simple to grow and their analysis is straightforward. However, it is increasingly being realized that more complex in vitro models that better represent the forms that cells take in vivo are needed. This project will focus on the development of a cell-based toxicity assay in which human cancer cell lines are grown as 3D assemblies, termed spheroids. Various fluorescent tools will be evaluated, which have the potential to provide a readout for the state of these spheroids as they grow. Automated fluorescence microscopy will be used to collect data from the spheroids, and an assay will be developed in which their response to a drug can be quantitatively measured. Results from this assay carried out in spheroids will be compared to results obtained from monolayer-grown cells, allowing us to conclude whether a spheroid model is suitable for such toxicity studies.

Desired qualifications: Students should have completed two years of undergraduate study in biology.

Relevant majors: Biology

The following information is vetted and provided by the American Association of Collegiate Registrars and Admissions Officers (AACRAO) on the Electronic Database for Global Education (EDGE).

Letter Grade	Percentage	Ranking	U.S. Equivalent
A+/A/A-	70 - 100%	First Class Honours	A
B+/B/B-	60 - 69%	Second Class Honours Upper	B+
C+/C/C-	50 - 59%	Second Class Honours Lower	B
D+	45 - 49%	Third Class Honours	C+
D/D-	40 - 44%	Pass	C
F	0 - 39%	Fail	F