http://hdl.handle.net/10453/35220 2026-10-15T05:28:25Z http://hdl.handle.net/10453/197097 Title: The immune-modulatory potential of helminth-derived proteins in cellular models of inflammation: a systematic review with cross-study quantitative data analysis. Authors: Stucke, S; Feeney, A; Lalor, R; Donnelly, S; Dalton, JP; McKernan, D; Dowd, E Abstract: Helminths are parasitic worms that secrete a plethora of immune regulatory molecules which allow them to moderate inflammatory responses by their host's immune system to ensure their survival within the host. This ability to drive a compatible existence with their host has led to research into the potential therapeutic effects of helminth-derived molecules the regulation of overactive immune and inflammatory responses in a wide variety of diseases. This systematic review with cross-study quantitative data analysis aims to synthesize the published data on helminth-derived immunomodulatory peptides/polypeptides/proteins (HDIPs) with a focus on determining the extent to which they modulate the inflammatory response in in vitro cellular models of inflammation. In accordance with PRISMA 2020 guidelines, a predefined systematic search of the PubMed, Web of Science and Medline databases identified relevant studies published up to September 2025, and 79 articles were included after screening. We found that most published studies used LPS or Concanavalin A stimulated macrophages, peripheral blood mononuclear cells or dendritic cells as the cellular model of inflammation. Twenty helminth species from which > 60 isolated HDIPs were derived were tested in these models, with the nematodes, Haemonchus contortus and Acanthocheilonema viteae, and the trematode, Fasciola hepatica, the most explored species. A common property of these molecules was the ability to significantly reduce the expression or production of pro-inflammatory cytokines such as IL-12, IL-1β, IL-6 and TNF, and significantly increase the expression or production of anti-inflammatory cytokines such as IL-10, TGFβ and IL-4. The effects on other cytokines, including IFNγ which is known to have both pro- and anti-inflammatory effects, were less consistent, with HDIPs either decreasing or increasing the levels of this cytokine. This systematic review with cross-study quantitative data analysis synthesizes the existing literature in this field and shows that the HDIPs secreted by several helminth species have consistently demonstrated effects though modification of cytokine levels and, as such, have therapeutic potential in conditions in which overactive immune and inflammatory responses play a pathogenic role. 2026-03-04T00:00:00Z http://hdl.handle.net/10453/197071 Title: Exfoliated kidney cells from urine for non-invasive kidney transplant monitoring: A potential opportunity? Authors: Wu, HHL; Parthiban, NK; Goldys, EM; Pollock, CA; Saad, S Abstract: Kidney transplantation is usually the optimal treatment option for patients living with kidney failure given its associations with improved survival, quality of life outcomes and a reduction in the personal, economic, and societal burden of long-term dialysis. While advantages of kidney transplantation are recognized, post-transplant complications, such as graft rejection, ischemia-reperfusion injury, surgical-related complications, and long-term consequences of immunosuppressive therapies, are commonly observed. There has been increased research on developing non-invasive biomarkers for the monitoring of transplanted kidneys over recent decades. The potential of urinary biomarkers to identify graft rejection, post-transplant acute tubular necrosis, detect progression of epithelial-to-mesenchymal transition toward tubulointerstitial fibrosis, and to differentiate between causes of graft dysfunction is an attractive alternative to invasive transplant biopsy. Innovative urinary biomarkers, such as those derived from omics technologies allow for a more holistic assessment of graft status through multi-parametric molecular analysis, although there remain questions on the consistency, reliability, and practicality of utilizing omics-based urinary biomarkers. The international nephrology community has continued to make concerted efforts to improve the procedures and cost-effectiveness of kidney transplant monitoring. In this article, we review the evidence and limitations of currently available urinary biomarkers and propose the application of urine-derived exfoliated kidney cells such as urinary exfoliated proximal tubule cells to prognosticate kidney transplant outcomes and monitor for post-transplant complications. Artificial intelligence and the incorporation of machine learning analysis of proximal tubular cell characteristics may optimize the process of differentiating graft rejection from other forms of kidney dysfunction non-invasively following kidney transplantation. 2026-06-01T00:00:00Z http://hdl.handle.net/10453/197037 Title: Buscogeny: A BUSCO leveraged phylogenomic tree builder. Authors: Webster, J; Chapman, TA Abstract: Accurately resolving evolutionary relationships among organisms is a cornerstone of evolutionary biology, with phylogenetic trees serving as critical tools for this purpose. While single-gene phylogenies are widely used, they often fail to capture the full complexity of evolutionary processes, particularly in organisms with complex or variable genomes. Concatenated gene alignments, built from multiple orthologous genes, offer a more robust framework but are challenging to generate, especially in taxa with incomplete or poorly annotated genomes such as fungi. To address this challenge, we present Buscogeny (Busco Phylogeny), an open-source command-line tool that automates the construction of concatenated gene alignments and phylogenetic trees using BUSCO-derived single-copy orthologs. Buscogeny integrates key features including genome quality assessment, ortholog extraction, multiple sequence alignment, gap filtering, and recombination filtering, followed by automated phylogenetic inference. By allowing users to select ortholog datasets of varying resolutions and tailor filtering thresholds, Buscogeny is adaptable to both closely related and more distantly related taxa. We demonstrate the utility of Buscogeny in phylogenetic analyses of both bacterial and fungal genomes, including complex groups such as Alternaria, where annotation and genome completeness vary significantly. The streamlined, reproducible workflow provided by Buscogeny makes BUSCO-based phylogenetic inference accessible to researchers across disciplines, enabling robust evolutionary insights from whole-genome data. 2026-02-01T00:00:00Z http://hdl.handle.net/10453/196998 Title: MammoMix: Leveraging Mixture of Experts for Robust Mammogram Breast Detection Authors: Nguyen, DT; Dang, HQ; Zhang, C; Ling, SH Editors: Liu, M; Yu, X; Xu, C; Song, Y Abstract: Breast lesion detection in mammography remains a challenging task due to variations in image quality, lesion appearance, and population demographics across datasets. While current object detectors such as YOLO and DETR achieve strong results on individual datasets, their performance often degrades when trained on or applied across heterogeneous sources. To address this, we propose MammoMix, a novel framework based on Mixture-of-Experts (MoE) paradigm for robust and generalizable lesion detection. In MammoMix, each expert model is trained on a specific domain, allowing it to specialize in distinct characteristics of its source data. A gating mechanism adaptively weighs contributions from each expert based on input image, combining their outputs to enable domain-adaptive inference. To improve reliability, we further incorporate a calibration module, MoCAE, which adjusts confidence scores to reflect true predictive uncertainty. We evaluate MammoMix on 3 public mammography datasets: CSAW, DDSM, and DMID, covering diverse clinical settings. Results show that MammoMix outperforms baseline detectors in both average precision and reliability, particularly on datasets with greater variability. Our findings demonstrate that expert specialization and calibrated ensemble fusion significantly enhance model generalization and robustness. MammoMix offers a promising step toward dependable AI-assisted breast cancer screening across real-world clinical domains. Code is available at: https://github.com/tommyngx/MammoMix 2026-01-01T00:00:00Z