Science Foundation Ireland

Poly-γ-glutamic acid (γ-PGA) is a valuable biopolymer with diverse industrial applications, produced naturally by several Bacillus species. The dairy environment is an under-explored niche for identifying efficient, food-grade γ-PGA producers. In this study, four legacy dairy-derived Bacillus licheniformis strains: DPC3803, DPC6338, DPC6339, and DPC6340, producing high γ-PGA titres were examined using whole genome sequencing (WGS) and comparative genomic analysis to evaluate their suitability for future industrial applications. The genomes ranged from 4.19 to 4.29 Mb with an average GC content of 45.8–46.2%. Pangenome analysis of the four strains, together with 51 publicly available B. licheniformis genomes, identified 12,415 gene clusters, of which 18.9% and 81.1% were core and accessory genes respectively. Average nucleotide identity (ANI) analysis demonstrated >99% sequence identity among all 55 B. licheniformis genomes, despite their isolation from diverse environments, indicating strong genomic conservation within the species. Experimental validation confirmed γ-PGA production by all four strains, with maximum titres (g/L) of 43.27 ± 1.49, 59.54 ± 4.33, 27.93 ± 1.87, and 47.74 ± 0.19 for DPC3803, DPC6338, DPC6339, and DPC6340, respectively. Genomic screening revealed multiple γ-PGA metabolism and CAZyme-encoding genes, as well as unique secondary metabolite clusters with potential antimicrobial activity. Although no plasmids or virulence factors were detected, twenty-one prophages were identified, sharing no significant homology with known cultivated phages, and a single β-lactamase gene suggested intrinsic resistance to β-lactams. These findings highlight the genomic and functional potential of these dairy-derived B. licheniformis as efficient, food-grade candidates for industrial γ-PGA production.

Seasonal energy storage is key to the future of fully decarbonised energy systems. Geological features capable of storing hydrogen at low costs are not universally available, and solutions to store renewable energy above ground will be required. In this study, an isolated energy hub investment optimisation model is developed with a constant demand centre of 100 MW, and different aboveground seasonal energy storage solutions are evaluated with different minimum load operation levels. In addition, the effects of different temporal optimisation resolutions and optimisation windows are evaluated. The results show that the current utility-scale lithium-ion battery system performs poorly in techno-economic terms at the seasonal storage level. Similarly, compressed hydrogen tanks are among the least competitive options. In contrast, ammonia and methanol are the most cost-effective solutions for new development sites reaching break-even electricity selling prices of 116€/MWh. However, for power generation sites with existing infrastructure like gas turbines, storage and pipelines, Fischer-Tropsch fuels become the most competitive option. Process capacity flexibility is also studied, and it is shown that it significantly decreases the costs. For these evaluated processes to reach commerciality, developing flexible production processes is crucial and should be prioritised. Finally, it is shown that seasonal storage investment is lower when lower temporal optimisation resolutions are used and increased when longer optimisation windows are considered. • Ammonia and methanol are the most cost-effective solutions for new developments. • Fischer-Tropsch fuels are preferred for existing power generation sites. • Lithium-ion batteries are the worst-performing option in all scenarios. • Capacity flexibility is key to decreasing energy storage costs. • Low temporal resolution and short optimisation periods decrease storage investment.

Bioimpedance and absolute blood volume measurement are objective assessments of volume status. However, uncertainties remain in their application at extreme body habitus. We aimed to compare blood volume and bioimpedance-based volume status at extreme body habitus and explore normalisation of blood volume in this context to support incorporation of these tools into fluid management decisions. We defined two common extreme volume-status phenotypes in haemodialysis: obese and bioimpedance-dry (n = 29) and lean and bioimpedance-wet (n = 29). Control subjects were haemodialysis patients who were haemodynamically stable with normal volume status (n = 20). We compared post-dialysis blood volume at clinical "target weight", measured by dialysate dilution, in the obese/dry and lean/wet groups with controls. We explored normalisation of blood volume (n = 98) by body mass, lean tissue mass and an anthropometric-based estimate of blood volume determined by Lemmens' formula. Post-dialysis blood volume normalized for body mass (specific blood volume) was lower in the obese/dry (57.9 mL/kg; P = 0.02) and higher in the lean/wet group (84.5 mL/kg; P = 0.001) than in controls (68.0 mL/kg). These differences were not present for lean tissue mass-normalised or Lemmens' formula-normalised blood volume. Normalising blood volume to body mass results in differences by sex and body mass, whereas normalising to Lemmens blood volume, which accounts for fat mass variation, does not. Setting target weights to achieve fixed reference bioimpedance or blood volume values may not be appropriate at extreme body habitus, reinforcing the need for individualised approaches. Normalising blood volume to body mass may exacerbate established measurement bias observed with bioimpedance.

The west-central Keewatin region of northern Canada preserves a rich record of former ice sheets and their impact on the landscape. The diverse assemblage of glacial landforms in the region reflects spatial variations in glacial modification driven by changes in basal thermal regime, subglacial hydrology, and ice-flow dynamics, all of which are key to reconstructing the history and behaviour of palaeo-ice sheets. Resolving how these landforms relate to changes in basal regime and glacial modification requires integrated datasets such as geomorphological mapping and to provide the robust reconstructions needed for climate and ice-sheet models.Our work aims to provide both a qualitative and quantitative assessment of glacial landscape modification in the west-central Keewatin region of the Northwest Territories and mainland Nunavut using two independent proxies: geomorphic mapping and cosmogenic nuclide concentrations. We used the ArcticDEM and Landsat 8 imagery to remotely map This new inventory of landforms, integrated with existing mapping, provides a qualitative assessment of glacial landsystems and landscape modification across the Keewatin region. As a quantitative proxy for glacial modification from the last glacial period(s), we collected 10 bedrock and 7 boulder samples for cosmogenic 10Be measurement along a north-south transect east of Dubawnt Lake, Nunavut. The sample transect was informed by glacial geomorphic mapping and was selected to test field-based predictions of the degree of landscape modification from the previous glaciation(s). The northern portion of the transect contains high concentrations of streamlined landforms associated with the onset zone of the Dubawnt Ice stream, inferred to be a region of high glacial modification. In contrast, the southern end of the transect contains lower concentrations of streamlined features, flow sets with contrasting orientations, and cross-cutting striations, suggesting a higher degree of landscape preservation during the last glaciation(s). Preliminary 10Be results broadly confirm our initial assessment of glacial modification along our transect, revealing slightly higher concentrations of 10Be in the south (less modification) and decreasing 10Be concentrations towards the north (more modification). Both bedrock and boulder samples follow this trend, however, our results show that in both low- and high-modification landsystems, some samples retain significant nuclide inheritance, including some boulders which suggests transport from a less modified landscape. Combining qualitative and quantitative approaches to evaluate glacial modification associated with specific landform assemblages informs our understanding of the basal thermal conditions of palaeo-ice sheets, the distribution of which informs our understanding of ice sheet evolution through space and time. Furthermore, the identification of streamlined features into palaeo-flow sets supports potential mineral exploration by helping to determine glacial transport directions and dispersal patterns. However, our results show the concentrations of cosmogenic nuclides vary within individual landsystems, suggesting that glacial modification varied through time and can be influenced by multiple factors (e.g., subglacial thermal conditions, topography, glacio-isostatic adjustment, lake development) that remain to be quantified.

Read the winning submission for the ASiT Foundation Doctor Essay Prize 2025.

Field-based contact sports, including American football and rugby are among those with the highest sport-related concussion (SRC) rates. Regarding SRC treatment, gradual return-to-play (RTP) protocols are commonly used, however variability in RTP protocols between sporting bodies and scholastic institutions makes accurately interpreting published SRC severity data difficult. Compare SRC recovery times, RTP protocols, and mandated sit-out period duration among adult amateur contact field sports. MEDLINE, CINAHL Complete, SPORTDiscus, APA PsycInfo, and APA PsycArticles were searched on May 13th, 2024. Eligibility criteria included (1) published in English in a peer-reviewed journal from 2015 to 2024, (2) adult population, (3) amateur, field-based contact sport athlete population, (4) diagnosed and/or suspected SRC, and (5) recovery time (days/weeks) data available. 19/6085 (< 1%) studies comprising 58 datasets met all inclusion criteria. The studies comprised American football (10/19, 52.6%), field-hockey (5/19, 26.3%), lacrosse (9/19, 47.4%), ladies gaelic football (1/19, 5.3%), rugby (8/19, 42.1%), soccer (10/19, 52.6%), and sprint football (1/19, 5.2%). Mandated sit-out periods ranged from 0 to 21 days across sports, and RTP protocol durations ranged from 5 to 21 days. The most common RTP protocol duration was 7 days (n = 31/58), and mandated sit-out period was no same-day return (n = 25/58). Field hockey had the longest reported SRC recovery duration (34 days), and rugby had the longest mandated sit-out period (21 days). Recovery time following concussion in contact sports should be interpreted in light of sport type, mandated sit-out policies, and RTP protocols. Empirical evidence should drive decisions in RTP protocols and sit-out periods, which are currently inconsistent between sports.

The growing number of Internet of Things (IoT) devices requires decentralized Edge Intelligence solutions. As the current FL-based IDS systems are decentralized solutions for privacy protection, face two major problems: (1) network traffic manipulation through adversarial evasion attacks (2) privacy threats from gradient-based inference attacks and (3) server-side Robustness issue. The current methods which use Differential Privacy (DP) or adversarial training result in 5-15% accuracy reduction which makes them unsuitable for deployment. The key novelty of our work is the integration of a novel dual-defense framework that uniquely reconciles the conflict between differential privacy noise and adversarial gradient requirements, effectively eliminating the conventional ”accuracy tax” along with server-side Robust Aggregation. Our research develops an enhanced two-stage federated system which is robust and protects privacy while delivering secure IoT edge intelligence solutions. The core system FedEdge-ID provides 99.73% detection performance across different edge devices. Spectre-Fed enhances the FedEdge-ID framework via three key defenses: (1) Hybrid Loss Adversarial Training (α=0.5) to fortify decision boundaries against evasion, (2) Gradient-Guided Adaptive Privacy with decreasing noise injection (σ0=0.0005, γ =0.95) for secure gradient updates, and (3) Robust Trimmed Mean Aggregation to counter Byzantine poisoning. Experiments demonstrate that Spectre-Fed’s client-side (Layer 1) defense achieves 99.72% clean accuracy with only a 0.01% utility loss versus the non-private baseline. It shows strong adversarial resilience, retaining 99.34% accuracy against FGSM attacks (ϵ =0.01), a mere 0.38% degradation from the clean state. When integrated with server-side Robust Aggregation (Layer 2), the system sustains 99.59% accuracy even under active label-flipping attacks from 20% of clients, while preserving high utility compared to the baseline. The system achieves optimal privacy-utility balance through its formal privacy protection and its ability to resist adversarial attacks which makes it suitable for zero-trust IoT systems.