Search Results (40)

Background: Ovarian cancer is characterized by high mortality rates, primarily due to diagnosis at late stages. Current biomarkers, such as CA125, have demonstrated limited efficacy for early detection. While high-dimensional proteomics offers a more comprehensive view of systemic biology, the analysis of such data, where the number of features far exceeds the number of samples, presents a significant computational challenge. Methods: This study utilized a nested case–control cohort of longitudinal pre-diagnostic serum samples from the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) profiled for eight candidate ovarian cancer biomarkers (CA125, HE4, PEBP4, CHI3L1, FSTL1, AGR2, SLPI, DNAH17) and 92 additional cancer-associated proteins from the Olink Oncology II panel. We employed a Synolitic Graph Neural Network framework that transforms high-dimensional multi-protein data into sample-specific, interconnected graphs using a synolitic network approach. These graphs, which encode the relational patterns between proteins, were then used to train Graph Neural Network (GNN) models for classification. Performance of the network approach was evaluated together with conventional machine learning approaches via 5-fold cross-validation on samples collected within one year of diagnosis and a separate holdout set of samples collected one to two years prior to diagnosis. Results: In samples collected within one year of ovarian cancer diagnosis, conventional machine learning models—including XGBoost, random forests, and logistic regression—achieved the highest discriminative performance, with XGBoost reaching an ROC-AUC of 92%. Graph Convolutional Networks (GCNs) achieved moderate performance in this interval (ROC-AUC ~71%), with balanced sensitivity and specificity comparable to mid-performing conventional models. In the 1–2 year early-detection window, conventional model performance declined sharply (XGBoost ROC-AUC 46%), whereas the GCN maintained robust discriminative ability (ROC-AUC ~74%) with relatively balanced sensitivity and specificity. These findings indicate that while conventional approaches excel at detecting late pre-diagnostic signals, GNNs are more stable and effective at capturing subtle early molecular changes. Conclusions: The synolitic GNN framework demonstrates robust performance in early pre-diagnostic detection of ovarian cancer, maintaining accuracy where conventional methods decline. These results highlight the potential of network-informed machine learning to identify subtle proteomic patterns and pathway-level dysregulation prior to clinical diagnosis. This proof-of-concept study supports further development of GNN approaches for early ovarian cancer detection and warrants validation in larger, independent cohorts. Full article

In recent years, Meteorin-like protein (Metrnl/IL-41) and Follistatin-like 1 (FSTL1) have emerged as multifunctional molecules that play roles in immunity, metabolism and tissue remodeling. Although they demonstrate pleiotropic effects, they are promising candidates for biomarkers and possible therapeutic targets. The development of new, disease-specific biomarkers will enable clinicians to more effectively monitor inflammatory activity, more accurately assess disease severity, better predict survival, and select appropriate medical treatments. In this review, we present the role of Meteorin-Like Protein (Metrnl) and Follistatin-like 1 (FSTL1) in inflammation in autoimmune rheumatic diseases, as well as in other autoimmune pathologies, cardiovascular diseases, and metabolic diseases. Metrnl, widely expressed in different tissues and organs, is very important for inflammation, immune responses and metabolic disorders. FSTL1 also shows dynamic changes in its expression through various diseases, including cardiovascular conditions, cancer, asthma, and arthritis. Both proteins participate in multiple important signaling pathways, and understanding their diagnostic and therapeutic potential holds great scientific interest. Their complex nature requires careful evaluation of safety concerns and translation to clinical practice. Full article

The successful integration of Terrestrial and Non-Terrestrial Networks (T/NTNs) in 6G is poised to revolutionize demanding domains like Earth Observation (EO) and Intelligent Transportation Systems (ITSs). Still, it requires Distributed Machine Learning (DML) frameworks that are scalable, private, and efficient. Existing methods, such as Federated Learning (FL) and Split Learning (SL), face critical limitations in terms of client computation burden and latency. To address these challenges, this paper proposes a novel hierarchical DML paradigm. We first introduce Federated Split Transfer Learning (FSTL), a foundational framework that synergizes FL, SL, and Transfer Learning (TL) to enable efficient, privacy-preserving learning within a single client group. We then extend this concept to the Generalized FSTL (GFSTL) framework, a scalable, multi-group architecture designed for complex and large-scale networks. GFSTL orchestrates parallel training across multiple client groups managed by intermediate servers (RSUs/HAPs) and aggregates them at a higher-level central server, significantly enhancing performance. We apply this framework to a unified T/NTN architecture that seamlessly integrates vehicular, aerial, and satellite assets, enabling advanced applications in 6G ITS and EO. Comprehensive simulations using the YOLOv5 model on the Cityscapes dataset validate our approach. The results show that GFSTL not only achieves faster convergence and higher detection accuracy but also substantially reduces communication overhead compared to baseline FL, and critically, both detection accuracy and end-to-end latency remain essentially invariant as the number of participating users grows, making GFSTL especially well suited for large-scale heterogeneous 6G ITS deployments. We also provide a formal latency decomposition and analysis that explains this scaling behavior. This work establishes GFSTL as a robust and practical solution for enabling the intelligent, connected, and resilient ecosystems required for next-generation transportation and environmental monitoring. Full article

Background: Dyslipidemia, a significant modifiable risk factor for cardiovascular disease (CVD), represents a major global health challenge, particularly influenced by complex genetic and environmental interactions, mainly in indigenous populations. Methods: In this study, DNA samples from 80 individuals belonging to various indigenous ethnic groups from northern and southern Mexico were analyzed to evaluate DNA methylation profiles and its correlation to lipid levels and other clinical parameters. Ten genes associated with metabolic changes were investigated using targeted bisulfite sequencing. Results: Our results revealed significant correlations between methylation in genes such as ABCA1, ADIPOQ, APOE, FSTL4, and KCNQ1 and clinical parameters including body mass index (BMI), lipid profiles, and body fat. Of the 151 CpG sites analyzed, 16 showed statistically significant correlations. Specifically, two ABCA1 CpGs sites correlated with BMI (p = 0.015) and triglycerides (p = 0.03); three ADIPOQ sites correlated with low-density lipoprotein cholesterol (LDLc) (p = 0.03, p = 0.005, p = 0.04, respectively); one APOE site correlated with BMI (p = 0.04), another with total cholesterol (p = 0.004) and triglycerides (p = 0.03) and two more with high-density lipoprotein cholesterol (HDLc) (p = 0.02 and p = 0.005, respectively); one FSTL4 CpG site with body fat (p = 0.02), another with total cholesterol (p = 0.02), one more with HDLc (p = 0.01), and another one with triglycerides (p = 0.01); and two KCNQ1 CpG sites correlated with body fat (p = 0.01 and p = 0.04, respectively). Conclusions: These findings show potential novel biomarkers for dyslipidemia risk. This research highlights the importance of understanding methylation changes in indigenous populations for developing personalized interventions and prevention strategies that could improve healthcare by linking epigenetic factors to CVD risk. Full article

This study investigated the role of molecular mimicry in the context of autoimmunity associated with viral infection, using SARS-CoV-2 as a model system. A bioinformatic analysis was performed to identify sequence homologies between the SARS-CoV-2 Spike (S) protein and the human proteome, with a specific focus on immunogenic regions to assess potential cross-reactivity. The analysis revealed homologous regions between the viral S protein and several human proteins, including DAAM2, CHL1, HAVR2/TIM3, FSTL1, FHOD3, MYO18A, EMILIN3, LAMP1, and αENaC, which are predicted to be recognizable by B-cell receptors. Such recognition could potentially lead to the production of autoreactive antibodies, which can contribute to the development of autoimmune diseases. Furthermore, the study examined potential autoreactive CD4+ T-cell responses to human protein autoepitopes that could be presented by HLA class II molecules. Several HLA class II genetic variants were computationally associated with a higher likelihood of cross-reactive immune reactions following COVID-19, including HLA-DPA1*01:03/DPB1*02:01, HLA-DPA1*02:01/DPB1*01:01, HLA-DPA1*02:01/DPB1*05:01, HLA-DPA1*02:01/DPB1*14:01, HLA-DQA1*01:02/DQB1*06:02, HLA-DRB1*04:01, HLA-DRB1*04:05, HLA-DRB1*07:01, and HLA-DRB1*15:01. Additionally, seven T helper cell autoepitopes (YSEILDKYFKNFDNG, ERTRFQTLLNELDRS, AERTRFQTLLNELDR, RERKVEAEVQAIQEQ, NAINIGLTVLPPPRT, PQSAVYSTGSNGILL, TIRIGIYIGAGICAG) were identified that could be implicated in autoimmune T-cell responses through presentation by class II HLA molecules. These findings highlight the utility of viral B- and T-cell epitope prediction for investigating molecular mimicry as a possible mechanism in virus-associated autoimmunity. Full article

Adipose-derived mesenchymal stromal/stem cells (AT-MSCs) can be typically isolated from adipose tissue using a minimally invasive procedure. However, since AT-MSCs are usually obtained from subcutaneous tissue, there is a risk of contamination with fibroblasts (FBs), which can reduce the differentiation potential of AT-MSCs. To avoid this contamination, it is crucial to identify specific markers to effectively distinguish AT-MSCs from FBs. Analysis of microarray data obtained from three studies (GSE9451, GSE66084, GSE94667, and GSE38947) revealed 123 genes expressed at levels more than 1.5-fold higher in AT-MSCs compared to FBs. Using STRING, a protein-protein interaction (PPI) network consisting of 80 nodes and 197 edges was identified within the 123 genes. Further investigation using Molecular Complex Detection in Cytoscape identified a module of 12 genes: COL3A1, FBN1, COL4A1, COL5A2, POSTN, CTGF, SPARC, HSPG2, FSTL1, LAMA2, LAMC1, COL16A1. Gene Ontology analysis revealed that these genes were enriched in extracellular region (GO: 0005576). Additionally, these 12 genes corresponded to the top 12 of the 15 hub genes calculated using the Maximal Clique Centrality algorithm. The results of this study suggest that these 12 genes may serve as markers for distinguishing AT-MSCs from FBs, offering potential applications in regenerative medicine. Full article

Background: Cardiomyocyte death is a key factor in myocardial ischemia–reperfusion injury (MI/RI), and the expression patterns and molecular mechanisms of pyroptosis-related genes (PRGs) in ischemia–reperfusion injury are poorly understood. Methods: The mouse MI/RI injury-related datasets GSE61592 and GSE160516 were obtained from the Gene Expression Omnibus database, and differential expression analysis was performed on each to identify differentially expressed genes (DEGs). The DEGs were intersected with the PRGs obtained from GeneCards to identify differentially expressed PRGs in MI/RI. Enrichment analysis identified key pathways, while PPI network analysis revealed hub genes. The expression patterns and immune cell infiltration of hub genes were also investigated. The molecular docking prediction of key genes was performed using MOE software in conjunction with the ZINC small molecular compounds database. Key gene expression was validated in an external dataset (GSE4105), a mouse MI/RI model, and an HL-1 cell hypoxia/reoxygenation model via RT-qPCR. Results: A total of 29 differentially expressed PRGs were identified, which are primarily associated with pathways such as “immune system process”, “response to stress”, “identical protein binding”, and “extracellular region”. Seven key genes (Fkbp10, Apoe, Col1a2, Ppic, Tlr2, Fstl1, Serpinh1) were screened, all strongly correlated with immune infiltration. Seven FDA-approved small molecule compounds exhibiting the highest docking potential with each key gene were selected based on a comprehensive evaluation of S-scores and hydrogen bond binding energies. Apoe, Tlr2, and Serpinh1 were successfully validated across external datasets, the mouse MI/RI model, and the cardiomyocyte H/R model. Conclusions: Apoe, Tlr2, and Serpinh1 may be key genes involved in MI/RI-related pyroptosis. Targeting these genes may provide new insights into the treatment of MI/RI. Full article

Cardiovascular diseases, which are among the most common diseases of the population and among the leading causes of death, are a constant topic of many research centers. A deeper understanding of their pathogenesis may contribute to the development of innovative diagnostic and therapeutic techniques. Recently, the role of myokines—a group of cytokines secreted mainly by muscle cells—has been increasingly emphasized in the development of these diseases. Both their excess and deficiency can cause undesirable effects that are involved in the pathomechanism of these diseases. In this review, we focus on the latest studies on the role of myonectin, irisin, musclin, follistatin-like1 (FSTL1), dermcidin, apelin, and myostatin in the pathogenesis of coronary artery disease, heart attack, heart failure, and hypertension. In particular, we look at myostatin and irisin in the context of the development of heart failure and decreased levels of apelin with higher cardiovascular risk in a group of patients with rheumatoid arthritis. Full article

Arteriovenous malformations (AVMs) are congenital vascular anomalies defined by abnormal direct connections between arteries and veins due to their complex structure or endovascular approaches. Pharmacological strategies targeting the underlying molecular mechanisms are thus gaining increasing attention in an effort to determine the mechanism involved in AVM regulation. In this study, we examined 30 human tissue samples, comprising 10 vascular samples, 10 human fibroblasts derived from AVM tissue, and 10 vascular samples derived from healthy individuals. The pharmacological agents thalidomide, U0126, and rapamycin were applied to the isolated endothelial cells (ECs). The pharmacological treatments reduced the proliferation of AVM ECs and downregulated miR-135b-5p, a biomarker associated with AVMs. The expression levels of angiogenesis-related genes, including VEGF, ANG2, FSTL1, and MARCKS, decreased; in comparison, CSPG4, a gene related to capillary networks, was upregulated. Following analysis of these findings, skin samples from 10 AVM patients were reprogrammed into induced pluripotent stem cells (iPSCs) to generate AVM blood vessel organoids. Treatment of these AVM blood vessel organoids with thalidomide, U0126, and rapamycin resulted in a reduction in the expression of the EC markers CD31 and α-SMA. The establishment of AVM blood vessel organoids offers a physiologically relevant in vitro model for disease characterization and drug screening. The authors of future studies should aim to refine this model using advanced techniques, such as microfluidic systems, to more efficiently replicate AVMs’ pathology and support the development of personalized therapies. Full article

The evaluation of external pelvimetry measurements and the genetic factors influencing them is essential for improving morphological characteristics and reproductive performance in cattle. This study represents the first comprehensive analysis of the association between single nucleotide polymorphisms (SNPs) and external pelvimetry traits in Romanian Simmental cattle, a breed recognized for its distinctive pelvic morphology. The relationship between single-nucleotide polymorphisms (SNPs) and external pelvimetry traits—including croup height (CH), buttock height (BH), croup width (CW), rump angle (RA), and croup length (CL)—was examined in Simmental cows. From an initial set of 110 SNPs, 33 markers were retained after applying quality control filters, including a minor allele frequency (MAF) greater than 0.05 and Hardy–Weinberg equilibrium. These SNPs, located on multiple chromosomes, were identified within intronic, exonic, or regulatory regions of relevant genes such as CLSTN2, DPYD, FBXL7, FBXL13, SEMA6A, RUNX2, FSTL4, DST, DCBLD2, FRMD6, CAV2.3, ABL2, SH3BP4, RSBN1L,and SAMD12, suggesting that these genetic variants may influence the development and morphology of the pelvic bones. Statistical analysis revealed significant relationships between certain allele variants and croup measurements, highlighting that the presence of alternative alleles can modify their morphological traits. Notably, the G allele in CLSTN2 reduced croup height by 5.74 cm (p = 0.0227), while the T allele in RUNX2 decreased rump angle by 4.49° (p = 0.0119). Full article

Background: The global prevalence of type 2 diabetes mellitus (T2DM) with liver fibrosis is rising, with T2DM identified as an independent risk factor and key prognostic factor for liver fibrosis. However, the underlying mechanisms remain unclear. Methods: To explore the shared pathogenesis of liver fibrosis and T2DM, we analyzed gene expression profiles from the GEO database. The co-differentially expressed genes (co-DEGs) were identified and subsequently analyzed through functional enrichment, protein–protein interaction (PPI) network construction, transcription factor prediction, and drug prediction. Machine learning algorithms were then applied to identify key genes. Results: A total of 175 co-DEGs were identified. Functional enrichment analysis indicated their involvement in extracellular matrix (ECM) remodeling, inflammation, and the PI3K/Akt signaling pathway. Through PPI network analysis and four algorithms, eight hub genes were identified, including SPARC, COL4A2, THBS1, LUM, TIMP3, COL3A1, IGFBP7, and FSTL1, with THBS1 being recognized as a key gene by machine learning. The upregulation of THBS1 was observed in both diseases, and it is closely related to the progression of liver fibrosis and T2DM. Transcription factor analysis detected 29 regulators of these hub genes. Drug prediction analysis suggested that retinoic acid may serve as a potential therapeutic agent. Conclusions: This study provides novel insights into the shared pathogenesis of liver fibrosis and T2DM and offer potential targets for clinical intervention. Full article

Background/Objectives: The quality and composition of dietary proteins are crucial during growth, particularly in children who follow vegetarian diets. Branched-chain amino acids (BCAAs: leucine, isoleucine, and valine) and lysine play essential roles in muscle growth, repair, and metabolism and are involved in the regulation of muscle-derived proteins known as myokines. This study aimed to compare the dietary intake and circulating levels of BCAAs, lysine, and myokines—follistatin-like protein 1 (FSTL-1), myostatin, and myonectin—between vegetarian and omnivorous prepubertal children and to explore the impact of diet on muscle metabolism. Methods: Sixty-four healthy Caucasian children aged 4–9 years (forty-two vegetarians and twenty-two omnivores) were assessed for dietary intake using the Dieta 5^® (extended version Dieta 5.0) software. Circulating BCAAs and lysine were measured using high-performance liquid chromatography, while myokine concentrations were determined using enzyme-linked immunosorbent assays. Results: Vegetarian children showed significantly lower intakes of total protein, animal protein, BCAAs, and lysine than omnivores. Correspondingly, the circulating levels of isoleucine, valine, lysine, and albumin were significantly reduced in vegetarians. Among myokines, serum myostatin and myonectin levels were comparable between the groups, but vegetarians had significantly lower median FSTL-1 levels 7.7 (6.5–9.4) ng/mL than omnivores 9.7 (7.5–13.9) ng/mL (p = 0.012). In the entire group of children, positive correlations were observed between dietary total and animal protein intake and circulating valine and lysine levels. Dietary animal protein intake was also positively associated with the serum levels of all myokines, whereas plant protein intake was negatively correlated with myonectin concentration. Conclusions: In conclusion, vegetarian diets in prepubertal children are associated with reduced dietary protein quality and lower circulating BCAAs, lysine, and FSTL-1 levels, which may impact muscle metabolism. Optimizing vegetarian diets using high-quality plant proteins with proper essential amino acids could mitigate their deficiencies and support muscle development during critical growth periods. Full article

Fibrosis represents a terminal pathological manifestation encountered in numerous chronic diseases. The process involves the persistent infiltration of inflammatory cells, the transdifferentiation of fibroblasts into myofibroblasts, and the excessive deposition of extracellular matrix (ECM) within damaged tissues, all of which are characteristic features of organ fibrosis. Extensive documentation exists on fibrosis occurrence in vital organs such as the liver, heart, lungs, kidneys, and skeletal muscles, elucidating its underlying pathological mechanisms. Regular exercise is known to confer health benefits through its anti-inflammatory, antioxidant, and anti-aging effects. Notably, exercise exerts anti-fibrotic effects by modulating multiple pathways, including transforming growth factor-β1/small mother decapentaplegic protein (TGF-β1/Samd), Wnt/β-catenin, nuclear factor kappa-B (NF-kB), reactive oxygen species (ROS), microRNAs (miR-126, miR-29a, miR-101a), and exerkine (FGF21, irisin, FSTL1, and CHI3L1). Therefore, this paper aims to review the specific role and molecular mechanisms of exercise as a potential intervention to ameliorate organ fibrosis. Full article

Arteriovenous malformations (AVMs) are characterized by direct connections between arteries and veins without intervening capillaries, with the concomitant formation of abnormal vascular networks associated with angiogenesis. However, the current understanding of the diagnosis and treatment of AVMs is limited, and no in vitro disease models exist at present for studying this condition. In this study, we produced endothelial cells (ECs) in two-dimensional cultures and three-dimensional (3D) blood vessel organoids (BVOs), comparing gene expression profiles between normal and AVM organoids. The normal and AVM organoids were examined via immunofluorescence staining using CD31 and phalloidin. The AVM organoids showed significantly higher expression levels of CD31 and phalloidin than the normal organoids. Genes such as FSTL1, associated with angiogenesis, showed significantly higher expression in the AVM organoids than in the normal organoids. In contrast, the MARCKS gene exhibited no significant difference in expression between the two types of organoids. The capillaries and related CSPG4 genes exhibited the lowest expression in the 3D AVM organoids. Furthermore, hsa-mir-135b-5p, a small RNA related to AVMs, showed elevated expression in AVM tissues and significantly higher levels in 3D AVM organoids. In our study, we were able to successfully establish AVM organoids (hBVOs) containing ECs and mural cells through advancements in stem cell and tissue engineering. These organoids serve as valuable models for investigating disease mechanisms, drug development, and screening potential therapeutic interventions in drug discovery. These findings contribute essential insights for the development of treatment strategies targeting AVMs. Full article

Background: The COVID-19 pandemic, caused by the novel coronavirus SARS-CoV-2, has posed unprecedented challenges to healthcare systems worldwide. Here, we have identified proteomic and genetic signatures for improved prognosis which is vital for COVID-19 research. Methods: We investigated the proteomic and genomic profile of COVID-19-positive patients (n = 400 for proteomics, n = 483 for genomics), focusing on differential regulation between hospitalised and non-hospitalised COVID-19 patients. Signatures had their predictive capabilities tested using independent machine learning models such as Support Vector Machine (SVM), Random Forest (RF) and Logistic Regression (LR). Results: This study has identified 224 differentially expressed proteins involved in various inflammatory and immunological pathways in hospitalised COVID-19 patients compared to non-hospitalised COVID-19 patients. LGALS9 (p-value < 0.001), LAMP3 (p-value < 0.001), PRSS8 (p-value < 0.001) and AGRN (p-value < 0.001) were identified as the most statistically significant proteins. Several hundred rsIDs were queried across the top 10 significant signatures, identifying three significant SNPs on the FSTL3 gene showing a correlation with hospitalisation status. Conclusions: Our study has not only identified key signatures of COVID-19 patients with worsened health but has also demonstrated their predictive capabilities as potential biomarkers, which suggests a staple role in the worsened health effects caused by COVID-19. Full article