Blog

  • Machine learning based on pangenome-wide association studies reveals the impact of host source on the zoonotic potential of closely related bacterial pathogens

    Machine learning based on pangenome-wide association studies reveals the impact of host source on the zoonotic potential of closely related bacterial pathogens

    Genomic analysis indicates open pangenome of Brucella

    To examine the genomic features of Brucella, we used 11 different Brucella species of publicly available genomes to characterize and define its pangenome. Core genes were defined as a set of genes shared with all 991 strains, accessory genes presented in at least two strain genomes or up to n-1 strain genomes (n: 991, total number of genomes), and the genes presenting only in one strain genome were defined as unique genes. The pangenome was constructed with the dataset consisting of 582 core genes, 4,121 accessory genes, and 2,462 unique genes from 991 Brucella spp. strains. Functional annotation of genes in the pangenome was conducted utilizing the Clusters of Orthologous Genes database (COGs), with the results revealing a distinctive distribution of functional categories across the three different pangenome subsets (Fig. 1A). Unique genes in the pangenome exhibited a higher ratio of genes annotated to the L category (Replication, recombination, and repair) than core and accessory genes (13.3% vs. 2.3% and 5.6%), with the majority of unique genes that belonged to the L category originating from B. inopinata (73.6%) (Supplementary Fig. 1). Furthermore, we found that the partial L category genes of B. inopinata were related to DNA modification, such as DNA adenine methylation, ISP type restriction/modification enzyme and DNA (cytosine-5-)-methyltransferase, indicating that B. inopinata may possess substantial epigenetic plasticity to aid in niche shift.

    Fig. 1: Brucella spp. pangenome statistics.

    A The Brucella spp. pangenome can be divided into three gene subsets: (i) core gene (the gene is present in n genomes, where n represents the total number of genomes), (ii) accessory gene (the gene is present in a number of genomes ranging between 2 and n-1), and (iii) unique gene (the gene is exclusively present in a single genome). The annotation and categorization of genes were performed by the COGs. COG categories encompass the following: Information storage and processing, which encompasses J (Translation, ribosomal structure and biogenesis), A (RNA processing and modification), K (Transcription), L (Replication, recombination and repair), and B (Chromatin structure and dynamics); Cellular processes and signaling, including D (Cell cycle control, cell division, chromosome partitioning), V (Defense mechanisms), T (Signal transduction mechanisms), M (Cell wall/membrane/envelope biogenesis), N (Cell motility), Z (Cytoskeleton), W (Extracellular structures), U (Intracellular trafficking, secretion, and vesicular transport), and O (Posttranslational modification, protein turnover, chaperones); Metabolism, including C (Energy production and conversion), G (Carbohydrate transport and metabolism), E (Amino acid transport and metabolism), F (Nucleotide transport and metabolism), H (Coenzyme transport and metabolism), I (Lipid transport and metabolism), P (Inorganic ion transport and metabolism), and Q (Secondary metabolites biosynthesis, transport and catabolism); Poorly characterized, including S (Function unknown). B The pangenome accumulation plot showed the change in pan genes (purple) and core genes (cyan) with the increasing size of the number of genomes. C The scatter plot displayed the number of accessory genes (X-axis) and unique genes (Y-axis) possessed by 991 Brucella strains, with the color of the points representing the species of each strain.

    Pangenome can be classified into two different types based on openness: open pangenome and closed pangenome. A highly open pangenome indicates that species inhabiting diverse environments possess the ability to exchange genetic material through various mechanisms, while species occupying the strained ecological niches find it challenging to acquire exogenous genes, resulting in a closed pangenome28,29. Heap’s law was used to characterize the pangenome openness of Brucella, and γ > 0 indicated an open pangenome for Brucella. For Brucella, we calculated that the relationship between pangenome size (P) and the number of genomes (N) was P = 1260.58*991^0.25, γ equals 0.25, and the pangenome size increased with the addition of new genomes (Fig. 1B), indicating that Brucella had an open pangenome and the pangenome repertoire of these 11 Brucella species was unable to be completely characterized using only the 991 genomes. Subsequently, we counted the number of accessory and unique genes within each strain of Brucella, and the result revealed that the majority of Brucella strains had a low number of unique genes, ranging from 0 to 100. Only five strains, including both two B. inopinata strains, had more than 100 unique genes. The number of accessory genes in most Brucella strains was around 2300. It is worth noting that both B. neotomae strains possessed a relatively low number of accessory genes, and the B. neotomae NCTC10070 strain had 1891 accessory genes, which is significantly lower than in other Brucella strains (Fig. 1C). This lower count is probably ascribed to the comparatively small size of the genome of this particular strain, which has a size of 2.9 Mbp.

    Quantifying the zoonotic potential of Brucella spp. via phylogenetic analysis presents substantial complexities

    For analyzing the genetic structure among the 991 Brucella strains, we used the core gene alignment result to construct an ML tree of all Brucella strains. In particular, we focused on whether phylogenetic analysis could separate classical zoonotic Brucella species from B. ovis, the only non-zoonotic species identified, and predict the zoonotic potential of other Brucella species accordingly. While Brucella species exhibited a close genetic relationship, they display notable variations in virulence, host preferences, and zoonotic capacity17. B. melitensis, B. abortus, and B. suis biovar 1 and 3 are the primary causative agents of brucellosis in both domesticated animals and humans30,31. Nevertheless, B. ovis infects animals exclusively and is the only known non-zoonotic Brucella species32,33,34,35. As shown in Fig. 2, B. abortus and B. melitensis each formed a highly differentiated clade, while the other clade consisted of a mixture of various Brucella species. The two atypical Brucella species, B. inopinata and B. vulpis, were distant from classical Brucella strains, and the phylogenetic relationship between the two species was also found to be quite distant. Zoonotic Brucella strains were clearly divided into three distinct groups, while the non-zoonotic species B. ovis were located in a separate clade. However, with the exception of these three zoonotic groups and the isolated clade of non-zoonotic species, it was difficult to determine the zoonotic potential of other species such as B. ceti and B. neotomae, based solely on phylogenetic analysis (Fig. 2).

    Fig. 2: Phylogenetic tree of Brucella species.
    figure 2

    This was a phylogenetic tree of 991 Brucella strains, reconstructed using a maximum-likelihood algorithm (raxml-ng) based on core gene alignment. Different species strains were labeled according to different colored dots. The red (zoonotic) and blue (non-zoonotic) background colors designated the zoonotic potential of Brucella strains.

    Unsupervised ML algorithms are incapable of distinguishing between zoonotic and non-zoonotic Brucella strains

    To acquire suitable input data for ML, we classified B. ovis, B. suis biovar 5, B. abortus 104 M strain, and B. suis S2 strains as non-zoonotic Brucella strains. Conversely, the zoonotic Brucella dataset encompassed all strains of B. melitensis, B. abortus, and part of the B. suis strains (B. suis biovar 1 and B. suis biovar 3), while excluding the aforementioned non-zoonotic strains. In total, 883 Brucella strains were manually annotated with zoonotic labels, among which 861 were classified as zoonotic pathogens and 22 as non-zoonotic pathogens. 268 genes with statistically significant associations to zoonotic or non-zoonotic phenotypes (P.adjust<0.05, Benjamini-Hochberg method) were identified as features through pan-GWAS analysis (Supplementary Data 1).

    We initially employed the PCA algorithm to analyze data from all 991 strains, and PCA plots revealed the presence of non-zoonotic strains that were intermingled with zoonotic strains (Fig. 3A). To ascertain the optimal parameter for the K-Means model, the Elbow method and silhouette coefficient were performed to determine the optimal value for the number of clusters in the unsupervised ML algorithm K-Means, and the results revealed that the optimal number of clusters was three (Supplementary Fig. 2). The K-Means algorithm classified all strains into three clusters, although most non-zoonotic strains were split into cluster 3, while clusters 1 and 2 contained a mixture of both zoonotic and non-zoonotic strains (Fig. 3B). The performance of the DBSCAN clustering algorithm was inadequate in distinguishing between zoonotic strains and non-zoonotic strains (Supplementary Fig. 3). These findings implied that unsupervised ML algorithms encountered challenges in discerning the distinction between zoonotic and non-zoonotic strains with precision.

    Fig. 3: Analysis of Brucella strains used the input binarized dataset by unsupervised ML algorithms PCA and K-Means.
    figure 3

    The red point and blue point represented zoonotic Brucella strains and non-zoonotic Brucella strains, respectively. A The PCA plot of the Brucella strains. PC1 and PC2 represented principal component one and principal component two, respectively. The green dashed line signified the admixture of zoonotic Brucella strains with non-zoonotic Brucella strains. B The K-Means plot of the Brucella strains. The border lines display the ranges of the three clusters.

    The supervised ML models based on SVM algorithm can accurately distinguish between zoonotic and non-zoonotic strains

    Due to the limited performance of unsupervised ML algorithms in distinguishing between zoonotic and non-zoonotic strains, five different supervised ML algorithms (RF, DTC, SVM, KNN, and MLP) were used to develop models for predicting the zoonotic potential of Brucella strains. Briefly, we divided 883 Brucella strains manually annotated with zoonotic labels into training and test datasets using random stratified sampling to ensure that both datasets maintained a proportional representation of zoonotic and non-zoonotic strains. The training dataset was then utilized to train each ML model and assessed the effectiveness of the models by stratified 3-fold cross validation. This entire process was repeated 100 times, and the results were averaged to obtain a more factual quality score for the models. The results showed that the models constructed with supervised ML algorithms achieved exceptional performance, with accuracy, recall, F1, and AUPRC mean scores all over 0.90, indicating that the supervised ML models were able to distinguish between zoonotic strains and non-zoonotic strains. Further analysis showed that the SVM ML model obtained the highest mean scores for all metrics except for the AUC and AUPRC score. Matthews correlation coefficient (MCC) score is a reliable metric for evaluating the quality of ML model in an imbalanced dataset. Among the five different ML models, the SVM model obtained the highest mean MCC score (0.78), and its MCC score distribution was more concentrated than that of the other ML models. These results suggested that SVM was the most appropriate algorithm for supervised ML model construction for Brucella zoonotic and non-zoonotic strain classification (Fig. 4A-F).

    Fig. 4: Violin and box plots showed the performance of ML models using different algorithms (RF, DTC, SVM, KNN, and MLP) in the validation dataset.
    figure 4

    The performance of five algorithms was demonstrated by accuracy score (A), recall score (B), F1 score (C), AUC score (D), AUPRC score (E), and MCC score (F). The box plots marked the median, upper and lower quartiles, and 1.5× inter-quartile range (whiskers); outliers were shown as points (n = 100 for each violin plot).

    The SVM prediction models accurately quantify zoonotic capacity of Brucella strains in the test dataset

    As the most suitable ML algorithm, we retrained the models using the hyperparameters of SVM models with the top 10% MCC score previously. To predict the zoonotic potential of Brucella strain, the output results of these retrained SVM models were averaged to obtain decision value for quantifying the zoonotic potential of the strain.

    The SVM prediction models were designed to generate decision values representing the zoonotic potential of Brucella strains on a continuous scale, with zero established as the threshold. This predictive framework allowed for the classification of zoonotic potential, where positive values indicated zoonotic potential and negative values denoted non-zoonotic potential. Moreover, the magnitude of the decision value reflected the strength of the corresponding classification, providing a quantitative measure of zoonotic potential. As shown in Fig. 5A, most of the Brucella strains from the zoonotic subset of the test dataset obtained decision values above +0.3, while certain B. suis strains demonstrated substantially lower decision values than others within the zoonotic subset. One strain of B. suis bv. 1 from the zoonotic test subset was erroneously assigned a decision value of -0.37 (below the threshold of zero), resulting in its misclassification as a non-zoonotic pathogen. (Fig. 5A).

    Fig. 5: Kernel density plot showed the quantitative zoonotic potential predictions of zoonotic and non-zoonotic Brucella strains in the test dataset.
    figure 5

    The red and blue points represented zoonotic Brucella strains (A) and non-zoonotic Brucella strains (B), respectively. The position of the point on the X-axis represented the zoonotic potential of the strain, and the points to the right of zero were predicted to be zoonotic Brucella strains, while those positioned to the left were predicted to be non-zoonotic Brucella strains. The red and blue background colors indicated the zoonotic probability density and the non-zoonotic probability density, respectively. The intensity of the background color correlated with the magnitude of zoonotic potential.

    In addition, the SVM prediction models accurately predicted the zoonotic potential of the non-zoonotic subset in the test dataset. The decision values for all non-zoonotic Brucella strains were below the threshold of zero. Notably, the SVM predictive models accurately predicted that the B. ovis had minimal zoonotic potential. B. ovis are generally recognized as non-zoonotic bacteria, exhibited decision values lower than -0.70, which further supported their classification as non-zoonotic bacteria. Two strains of B. suis bv. 5 were assigned decision values of -0.26 and -0.25, respectively, which were higher than those of B. ovis strains, indicating that their non-zoonotic potential was comparatively weaker than that of B. ovis strains (Fig. 5B).

    The SVM prediction models exhibit strong generalization ability

    Lipopolysaccharide (LPS) serves as a crucial virulence factor in Brucella, playing a pivotal role in the intracellular survival of Brucella within host cells. Brucella spp. are classified into smooth or rough phenotypes depending on whether O-polysaccharide (O-PS) is present on the LPS, and the rough phenotype Brucella strains frequently exhibit significantly attenuated virulence compared to the smooth phenotype Brucella strains, such as B. melitensis and B. abortus36. The natural rough phenotype of Brucella includes B. ovis and B. canis37. Although B. canis is currently the only species within Brucella spp. sharing the same LPS as B. ovis, there have been several reports of human infected by B. canis in recent years, indicating its zoonotic potential15,38,39.

    While the SVM prediction models demonstrated excellent predictive performance in the test dataset, it is essential to further evaluate their generalization ability. B. canis, which is the only zoonotic species sharing the absence of O-PS with the non-zoonotic B. ovis species in Brucella spp., provided appropriate data for assessing the generalization ability of SVM prediction models. Therefore, we utilized the SVM prediction models to predict the zoonotic potential of 26 strains of B. canis. The mean decision value for B. canis was +0.49. Among the strains of B. canis, the highest positive decision value was +0.70. Notably, B. canis was not included in the training dataset of the SVM prediction models. The prediction results suggest that B. canis exhibits an intermediate zoonotic potential, which is in correspondence with the current understanding of this pathogen. The result indicated that the SVM prediction models demonstrated a strong generalization ability and can be utilized to estimate the zoonotic potential of other species strains within the genus Brucella (Fig. 6).

    Fig. 6: Kernel density plot showed the quantitative zoonotic capacity predictions of B. canis, B. ceti, B. pinnipedialis, B. neotomae, B. microti, B. inopinata, B. suis bv. 2, and B. suis bv. 4 strains.
    figure 6

    The color and position of the points indicated the species and predicted zoonotic potential of the strains, respectively. The red and blue background colors indicated the zoonotic probability density and the non-zoonotic probability density, respectively. The intensity of the background color correlated with the magnitude of the zoonotic potential. Animal silhouettes were obtained from PhyloPic (http://phylopic.org) under CC0 1.0 Universal Public Domain Dedication license.

    Multiple various Brucella species display a wide range of zoonotic potential

    The high similarity (~97%) at the genomic level within the genus Brucella allows for the utilization of our pan-GWAS-based SVM prediction models among different species of Brucella strains and various host-derived isolates of the same species. The marine mammal-derived Brucellae, including B. ceti and B. pinnipedialis, have been demonstrated to have potential zoonotic capacity40. Here, nine strains of B. ceti were evaluated for zoonotic potential using the SVM prediction models with an average decision value of +0.36. The highest decision value of +0.57 was obtained from a B. ceti strain isolated from a common bottlenose dolphin (Tursiops truncatus), while the lowest decision value of +0.10 was obtained from a B. ceti strain isolated from common dolphin (Delphinus delphis). Similarly, for B. pinnipedialis, the strains of this species exhibited an average decision value of +0.32. The highest decision value of +0.53 was observed in a B. pinnipedialis strain isolated from hooded seal (Cystophora cristata), while the lowest decision value of +0.11 was observed in a B. pinnipedialis strain isolated from harbor seal (Phoca vitulina). These findings highlight the necessity of surveilling zoonotic potential of marine mammal Brucellae. (Fig. 6).

    B. neotomae, first isolated from desert woodrats (Neotoma lepida) in the United States in 1957, was previously considered a non-zoonotic bacterium. However, two cases of B. neotomae infected human were reported in 201714, indicating that humans may be part of its host range. The SVM prediction models predicted an average decision value of +0.17 for five B. neotomae strains. Among these strains, the highest decision value was at +0.26, and the lowest decision value was at +0.03 (Fig. 6).

    B. microti is the only known species capable of surviving in soil within the Brucella spp.41. In this study, two strains of B. microti were predicted to have zoonotic potential, one strain was isolated from a vole (Microtus arvalis), while the other was found in a pool frog (Pelophylax lessonae). These strains had an average decision value of +0.37. Notably, the strain of B. microti derived from the vole exhibited a higher decision value of +0.54 compared to the strain isolated from the pool frog, which obtained a decision value of +0.21 (Fig. 6).

    Apart from B. ceti, B. pinnipedialis, and B. microti, the atypical Brucella species include B. vulpis and B. inopinata10. Two strains of B. vulpis were assessed for their zoonotic potential by the SVM prediction models. The two highly similar strains of B. vulpis were isolated from red foxes (Vulpes vulpes) in 200842. As a result of their similarity, the decision values for both strains were identical, with a value of +0.15. This lack of variation in decision value made it impossible to construct a kernel density plot for B. vulpis.

    A strain of B. inopinata BO1 was isolated from the breast implant wound fluid of a 71-year-old female patient in 200843, and its zoonotic potential was predicted to be +0.076 based on the SVM prediction models. Despite the B. inopinata 141012304 strain being isolated from a bluespotted ribbontail ray (Taeniura lymma), its predicted zoonotic potential was slightly higher at +0.080 compared to B. inopinata BO1 ( + 0.076) isolated from the human patient (Fig. 6).

    B. suis is divided into 5 various biovars, with B. suis bv 1, 3, and 5 being used to construct the SVM prediction models. The zoonotic potential of B. suis bv. 2 was subsequently predicted using the SVM prediction models, with an average decision value of +0.26. Additionally, B. suis bv. 4 strains obtained an average decision value of +0.42. Two distinct strains of B. suis bv. 4 strains isolated from a reindeer (Rangifer tarandus) and a wild boar were assessed for their zoonotic potential, yielding decision values of +0.29 and +0.54, respectively (Fig. 6).

    The zoonotic potential of Brucella strain is affected by both species and isolated host

    Statistical analysis further corroborated that the SVM prediction models provided species-specific and host-specific zoonotic potential for Brucella strains. The results revealed that the zoonotic potential of B. abortus and B. melitensis is significantly higher than that of most other Brucella species. Notably, the SVM prediction models provided species-specific decision values for B. melitensis strains in the test dataset based solely on the gene features of the training samples, even in the absence of the input species information for all the Brucella strains, which is the most virulent Brucella species for humans32. Conversely, the zoonotic potential of B. ovis was significantly lower compared to B. abortus, B. melitensis, B. suis, B. canis, B. neotomae, and marine mammal Brucella spp. (Fig. 7A, B).

    Fig. 7: Impact of species and host factors on the predicted zoonotic potential of Brucella strains.
    figure 7

    A Scatter plot showed the zoonotic potential of Brucella strains in the test dataset and various Brucella species strains derived from different hosts, the color of each point indicating the decision value for each Brucella strain. B Statistical significance of the decision value across 8 Brucella species. The single asterisk (*, yellow), double asterisk (**, blue), triple asterisk (***, red), and “ns” indicated statistically significant, very significant, extremely significant, and nonsignificant differences, respectively, from the permutation test. C Box plot showed the predicted zoonotic potential of B. melitensis strains derived from various hosts. Statistical significance of the decision values was determined by the Kruskal-Wallis test (**, P.adjust<0.01; ***, P.adjust<0.001). D Box plot showed the predicted zoonotic potential of the B. suis bv. 2 strains derived from swine or wild boar. Statistical significance of the decision values was determined by the independent samples t-test (**, P < 0.01).

    Additionally, B. melitensis strains isolated from Homo sapiens exhibited significantly higher zoonotic potential than those isolated from Bos taurus, Capra hircus, and Ovis aries (Fig. 7C). Among the B. suis bv. 2 strains, the highest decision value was +0.65, isolated from a swine (Sus scrofa domesticus), while the lowest decision value was -0.04, isolated from a wild boar (Sus scrofa) (Fig. 6). The considerable difference between the highest decision value and the lowest decision value indicated a need to compare the zoonotic potential of B. suis bv. 2 strains derived from different hosts. We compared the decision values of Brucella strains isolated from different hosts and found that those isolated from swine exhibited significantly higher zoonotic potential than those isolated from wild boars, despite both belonging to the Sus scrofa species (Fig. 7D).

    Model explanations reveal the genes potentially used to estimate the zoonotic potential of Brucella strains

    Model explanations were conducted using the SHAP method on a set of 268 genes, which served as the features for constructing the SVM prediction models. The SHAP method offers two distinct perspectives for explaining the impact of features on ML model outputs: a global explanation and a local explanation. As shown in Fig. 8A, all 268 features were ranked according to the mean absolute SHAP values in the global model explanation, identifying the ten most influential features on the predicted results (Fig. 8A). Furthermore, the SHAP dependence plot showed how variations in feature values (gene present or gene absent) affected the predictions for Brucella strains. The feature group_2874 was the most significant contributor to the output of SVM prediction models, resulting in predictions towards zoonotic direction when absent and towards non-zoonotic direction when present, with a greater distance of the shift than that observed in the absence of group_2874. It has been preliminarily identified as an uncharacterized protein, and currently, there is no definitive functional characterization available. This indicated that group_2874 was more likely to be associated with a diminished zoonotic potential of Brucella strains (Fig. 8B).

    Fig. 8: Global and local explanations provided by the SHAP method.
    figure 8

    A SHAP summary bar plot showed the top 10 features in descending order of the mean SHAP absolute values. B SHAP dependence plot showed the impact of 10 features on the decision value output of the SVM prediction models. Each dot represented a single Brucella strain in the test dataset, with red or blue indicating that the feature was present or absent in the genome of the Brucella strain, respectively. C SHAP force plot showed the impact of all features on the predicted results for each Brucella strain in the test dataset. The x-axis represented each sample, and the y-axis represented the predicted zoonotic potential of Brucella strains, with the larger part in red indicating greater zoonotic potential of the Brucella strains. D, E SHAP waterfall plot showed the local explanations for the typical zoonotic pathogen B. melitensis 16 M strain and the non-zoonotic pathogen B. ovis ATCC 25840 strain.

    The predicted results for each Brucella strain in the test dataset were explained in Fig. 8C. The upper x-axis represented each sample, while the y-axis represented the contributions of features to the predicted results. As the blue part of the sample increased, the likelihood of the Brucella strain being pathogenic to humans was decreased (Fig. 8C). In addition, the local explanation of two Brucella strains, including the typical zoonotic Brucella strain (B. melitensis 16 M strain, GCA_000740415.1) and non-zoonotic Brucella strain (B. ovis ATCC 25840 strain, GCA_000016845.1), provided a personalized analysis of the predicted results. For instance, the absence of group_2974 resulted in an increase in the predicted zoonotic potential of the B. melitensis 16 M strain. In contrast, the presence of group_2974 in the B. ovis ATCC 25840 strain decreased its predicted zoonotic potential. The local explanation clearly demonstrated the roles that various features play in the zoonotic potential prediction process for individual Brucella strains (Fig. 8D, E).

    Continue Reading

  • Arkema employees honored by acs for Kynar® PVDF innovation in batteries Arkema employees honored by acs for Kynar® pvdf innovation in batteries

    Arkema employees honored by acs for Kynar® PVDF innovation in batteries Arkema employees honored by acs for Kynar® pvdf innovation in batteries

    Kynar® HSV 900 PVDF has demonstrated exceptional versatility, gaining widespread commercial adoption alongside a broad range of cathode active materials, especially lithium iron phosphate (LFP), and has since become a legacy market reference in the battery industry. To date, Kynar® HSV 900 PVDF has already enabled the production of batteries powering over 10 million electric vehicles.

    The development of advanced materials for efficient, long-lasting batteries for electric vehicles is essential to mitigate the impacts of climate change through the diminution of the carbon footprint of transportation. Kynar® HSV 900 plays a critical role in improving the performance, safety and longevity of lithium-ion batteries, key components in EV and energy systems.

    The adoption underscores the critical role of Kynar® HSV 900 has played in enabling the rise of LFP technology, an increasingly preferred solution for EV batteries due to its safety, cost-efficiency and long cycle life. In 2024, the rapid growth of LFP-powered electric vehicles continued, with over 850 GWh of batteries produced.

    Continue Reading

  • Acute and chronic stress produce distinct behavioral effects in male and female rats

    Acute and chronic stress produce distinct behavioral effects in male and female rats

    A study analyzes the negative effects of stress on the brains of male and female rats and concludes that acute stress induces anxiety-like behaviors, especially in males, while chronic stress is more associated with depressive symptoms. Understanding these differences may help develop more effective approaches to preventing and treating mental disorders such as anxiety and depression.

    It is well established that stress can increase susceptibility to various neuropsychiatric disorders, such as depression and anxiety, which are highly prevalent worldwide and represent a significant economic burden and public health issue in our society. The World Health Organization estimated that in 2019, around 970 million people globally – one in eight – suffered from a mental disorder.

    Evidence also points to sex differences in the prevalence of and responses to stress. In fact, while women account for two-thirds of patients with stress-related disorders, another indicator shows that more than two-thirds of suicide victims are men.

    Although stress is a part of life and, in small doses, can even be beneficial, excessive exposure, whether acute or chronic, can have profound negative effects, especially on the brain, potentially leading to cerebrovascular diseases. One of the most sensitive targets of excessive stress is the blood-brain barrier, a structure that protects the brain from potentially harmful substances.

    However, despite evidence suggesting that different types of stress can compromise its integrity and trigger neuroinflammatory responses associated with various neurological conditions, the cellular and molecular mechanisms underlying these effects remain poorly understood.

    In the article “Distinct behavioral and neurovascular signatures induced by acute and chronic stress in rats”, published in September in the scientific journal Behavioural Brain Research, a research team from the University of Coimbra, Portugal, led by Ana Paula Silva, sought to clarify this issue by analysing the effects of acute and chronic stress in rodents.

    With support from the BIAL Foundation, the researchers used open field and forced swimming tests to assess locomotor activity and anxiety- and depression-like behaviors in male and female Wistar rats.

    The results showed that acute stress induces anxiety-like behaviors, especially in males, while chronic stress is more associated with depressive symptoms. Additionally, changes were observed in key proteins of the blood-brain barrier, with significant sex differences.

    The research confirmed that acute stress and so-called chronic mild unpredictable stress induce distinct behavioral and biochemical profiles, highlighting the importance of differentiating stress types and considering biological variables, such as sex, in neuroscience research.

    Our study shows how important it is to understand the differences between types of stress to better grasp the causes of mental disorders like anxiety and depression, and to find more effective ways to prevent and treat these issues.”


    Ana Paula Silva, University of Coimbra, Portugal

     

    Source:

    Journal reference:

    Simões, D. M., et al. (2025). Distinct behavioural and neurovascular signatures induced by acute and chronic stress in rats. Behavioural Brain Research. doi.org/10.1016/j.bbr.2025.115706.

    Continue Reading

  • Kirkland Advises Thoma Bravo-Backed Flexera on Recapitalization | News

    Kirkland & Ellis advised Thoma Bravo portfolio company Flexera Software, a global leader in technology spend and risk intelligence, on its debt financing recapitalization. KKR, through its managed credit funds and accounts, served as the lead investor in the transaction, with KKR Capital Markets acting as Lead Arranger and Bookrunner.

    Read the transaction press release

    The Kirkland team included debt finance lawyers Fred Lim, Omar Raddawi, Matt Park and Jay Gao; and corporate lawyers Corey Fox, Brad Reed, Peter Stach, Steven Page and Ian Hesterly.

    Continue Reading

  • 48-Year-Old’s Wrist Pain Reveals Bone Necrosis

    48-Year-Old’s Wrist Pain Reveals Bone Necrosis

    Key Takeaways

    A 48-year-old man presented with a 2-year history of oedema and pain in his right hand. Initially diagnosed with tendonitis, he underwent physical therapy which provided limited relief. MRI demonstrated avascular necrosis of the lunate.

    The patient underwent surgery and follow-up treatment with home exercises and a structured physiotherapy program; he regained minimal wrist flexion restriction and normal grip strength.

    The Patient and His History

    In 2018, the patient had a vertebral fracture that resolved without complications. He received physical therapy and anaesthetic injections which provided limited relief. Due to the persistent symptoms, further evaluation was conducted.

    Findings and Diagnosis 

    Physical examination revealed palm ecchymosis, reduced range of motion in both flexion and extension of the right wrist, and diminished grip strength. Imaging studies, including x-rays and MRI, indicated osteosynthesis of the carpal bones and fractures of the scaphoid and lunate bones. Coronal MRI showed diffuse hypointensity and collapse of the lunate.

    The patient was diagnosed with avascular necrosis of the semilunar bone of the carpus, Kienböck disease stage IV (according to the Lichtman classification), and carpal arthrofibrosis.

    The patient underwent surgical intervention, including scaphoid-lunate fusion with bone grafting, semilunar tendon arthroplasty, and radiocarpal capsulectomy.

    Postoperatively, the patient experienced significant improvement, although the chronic pain initially persisted. He underwent a 12-session physical therapy program. No medications for pain were prescribed. After 1-year of home exercises, he achieved almost full range of motion, with minimal wrist flexion restriction and normal grip strength.

    Discussion

    Kienböck disease, or lunate osteonecrosis, is a debilitating condition that primarily affects the lunate bone in the wrist. It is characterised by avascular necrosis due to an interrupted blood supply, leading to bone death. This rare condition affects approximately 0.0066% of the population, predominantly men aged 20-40 years.

    The aetiology of Kienböck disease is largely unknown; however, several contributing factors have been identified. These include repetitive microtrauma, acute wrist injuries, and anatomical variations in the lunate blood supply, which are limited to a few vessels, making it susceptible to ischaemic damage. Systemic conditions, such as lupus or sickle cell disease, that impair blood flow may also play a role.

    The pathophysiology begins with an ischaemic event leading to lunate necrosis, followed by changes such as bone fragmentation, collapse, and carpal instability.

    Management varies by stage, focusing on pain relief, preservation of wrist function, and prevention of disease progression. Early stages may respond to conservative treatments such as immobilisation, nonsteroidal anti-inflammatory drugs, and physical therapy, but these are often ineffective in advanced stages.

    Management requires a tailored approach based on disease stage and individual factors. Early diagnosis is crucial for effective treatment to preserve wrist function and slow disease progression. The authors noted that continued research and advances in imaging are expected to improve both the understanding and management of this condition.

    This article was translated from Univadis Germany.

    Continue Reading

  • Where Is the Medical Home for Postinfectious Illness?

    Where Is the Medical Home for Postinfectious Illness?

    Brittany L. Adler, MD, was a rheumatology fellow in 2017 when she evaluated a young woman with severe fatigue, dizziness, gastrointestinal symptoms, and a bluish discoloration in her feet.

    Systemic sclerosis was suspected, but the patient didn’t meet all the diagnostic criteria. Adler was stumped until she happened to hear a preceptor mention postural orthostatic tachycardia syndrome (POTS), a condition she hadn’t recalled ever learning about in any of her medical training. “We ordered a tilt table test, which was positive. Her treatment path shifted dramatically. And I was left wondering: How many of these patients had I already missed?”

    Brittany L. Adler, MD

    Probably quite a few, she now knows. “After seeing this patient, it became impossible to unsee it. I began recognizing POTS more frequently, especially in hypermobile young women. Their clinical histories followed a distinct, consistent pattern marked by orthostatic intolerance, widespread pain, fatigue, and gastrointestinal symptoms. Over time, this spectrum of illness became as real and recognizable to me as systemic lupus erythematosus or myositis,” Adler wrote in an essay published on July 23, 2025, in The Lancet Rheumatology.

    Adler now works at the Johns Hopkins POTS Clinic in Baltimore, exclusively caring for these patients. The COVID-19 pandemic brought a surge of more patients with POTS and infection-associated chronic illness. “What had once felt like a rare or niche diagnosis now seemed to be everywhere, impossible to ignore,” she wrote.

    In her essay, Adler urged her fellow rheumatologists to take on these patients, at least to become more familiar with the entirety of their illness burden and treat their symptoms as much as possible. She also called for rheumatology training programs to broaden their scope to include associated conditions such as POTS and myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS).

    “Rheumatologists are uniquely trained to manage complex, multisystem illnesses. Yet the field has largely remained at the margins of infection-associated chronic illness and autonomic dysfunction, despite their clear overlap with autoimmune disease. Many other specialties have also declined ownership, leaving patients to fend for themselves,” Adler wrote.

    But Adler told Medscape Medical News that she also doesn’t see rheumatology as the only specialty in this arena. “There is an enormous need for multidisciplinary clinics to manage these patients, as they experience symptoms across multiple organ systems. Right now, there is no model for coordinated care, so many patients end up seeing a different doctor each day and are in near-constant contact with the medical system.”

    Many Diagnoses, Many Doctors, Few Answers Yet 

    Indeed, people with postinfectious chronic illness often receive multiple diagnoses from different clinicians, but there is no dedicated space for them in the current healthcare system, and as a result, their care is often suboptimal. Their diagnoses vary in number and degree and may include ME/CFS, dysautonomia/POTS, Ehlers-Danlos syndrome, fibromyalgia, mast cell activation syndrome, and long COVID, among many others.

    Those with ME/CFS specifically experience extreme fatigue, postexertional malaise (PEM), orthostatic intolerance, and cognitive problems (aka “brain fog”).

    Many can trace these symptoms to a specific infection such as SARS-CoV-2 or Epstein-Barr virus or to a bacterial infection such as Lyme disease. But for others, the infectious trigger may not have been confirmed or recognized at the time. They were healthy and active, then they weren’t. Some are severely disabled and can’t work, go to school, or even do simple tasks without feeling depleted and even sicker afterward.

    The terminology is still being worked out. The umbrella terms “infection-associated chronic conditions (IACCs)” and “infection-associated chronic illnesses (IACIs)” have been used, along with the more specific “postacute infection syndromes (PAISs).” The terms PAIS and IACI are “increasingly used by authoritative organizations and very senior scientists,” Anthony L. Komaroff, MD, Simcox-Clifford-Higby Distinguished Professor of Medicine at Harvard Medical School and senior physician at Brigham and Women’s Hospital, both in Boston, told Medscape Medical News.

    photo of Anthony Komaroff
    Anthony L. Komaroff, MD

    Komaroff, who has been researching and publishing about ME/CFS since the 1980s, also told Medscape Medical News, “As for the issue of which medical specialties will ‘own’ PAIS, I think that’s less important than having enough doctors knowledgeable about these illnesses to meet the need in every community, regardless of what subspecialty training they have. I think that will happen as the science becomes increasingly robust.”

    He recently wrote a commentary on the topic for the Proceedings of the National Academy of Sciences, accompanying a paper identifying patient-reported treatment outcomes in ME/CFS and long COVID.

    What All Doctors Can and Should Do, at a Minimum

    Rheumatologist Brayden Yellman, MD, medical director of the Bateman Horne Center, Salt Lake City, told Medscape Medical News that he “agrees wholeheartedly” with Adler’s essay. “I do, ultimately, think that rheumatologists would be as adept as any clinicians at helping manage the more complicated presentations of those with IACCs and their related comorbid conditions.”

    photo of Brayden Yellman
    Brayden Yellman, MD

    But, Yellman noted, several barriers are keeping rheumatologists from stepping up to the plate, including lack of a serologic biomarker or distinct imaging findings, lack of familiarity with the entire range of possible treatment approaches, and a shortage of rheumatologists to manage even patients with well-defined rheumatologic conditions.

    In addition, Yellman noted, “We do not practice in a system that values or allows providers the time and resources necessary for good care of complex multisystem illness to begin with nor can this already taxed and flawed system take on more.”

    Nonetheless, Yellman said, “With the extensive prevalence of these illnesses and the need for immediate action for the millions suffering from them, I concurrently believe that we need to demand better clinical care and support from all providers within the healthcare system, and particularly, those in primary care. At a very minimum, we need to be making the correct diagnoses in those with IACCs instead of telling them they ‘don’t have anything’ or that they have ‘functional neurological disorder.’ This type of dismissal can no longer be tolerated.”

    Furthermore, “we need to be identifying PEM and helping to teach patients how to pace to avoid PEM. We need to be supporting patients’ needs for work and school accommodations or medical leaves of absence to help promote symptomatic improvement and to allow people to emerge from a cyclical push-crash cycle of PEM. We need to be, at a minimum, diagnosing dysautonomia and providing at least some basic support to promote improved vascular regulation. The entire healthcare system, and its providers, need to step up to this challenge.”

    Are Long COVID Clinics a Model?

    The multidisciplinary long COVID clinics that were established soon after that phenomenon emerged from the COVID-19 pandemic could serve as a model for treating all patients with chronic postinfectious illness, if there were sufficient support for them. However, many have either scaled back or closed entirely.

    In Connecticut, for example, there had been at least 10 long COVID clinics, but now there is just one, the Yale New Haven Health Systems Long COVID Consultation Clinic. Medical director Lisa Sanders, MD, is the only MD provider, and she only works there part-time. “And we’re booked out until March,” she told Medscape Medical News.

    photo of Lisa Sanders
    Lisa Sanders, MD

    Sanders does refer some patients with long COVID to specialists within the Yale system, most commonly cardiology or neurology, because “the most distressing symptom for most of these patients is not the shortness of breath, which gets better, and not the tachycardia, which can be managed, but the brain fog and the changes in cognition…which is probably the most common reason that people can’t go back to work.”

    When she has more time, Sanders said she’d like to change the identity of the clinic from addressing just long COVID to encompassing other postinfectious conditions, given that “there are other illnesses…‘long COVID’ existed before COVID.”

    Sanders, perhaps best known for her The New York Times columns, said she hopes that more clinicians take an interest in these patients, “but what I really hope is that we get some better answers than what we have so far…. We need better research. I mean, we can’t even agree on what defines long COVID or a postinfectious syndrome…really basic stuff.”

    Adler, too, called for more research. “One development that I think is going to be transformative, and I don’t think we’re far off from this, is discovering a biomarker. As soon as there’s a blood test that can diagnose this syndrome, it will become much more tangible and accessible to doctors. I’m hoping the field will receive more research funding to make this possible.”

    Adler and Yellman reported having no disclosures. Sanders reported receiving payments from The New York Times. Komaroff’s work was funded by a subcontract to Brigham and Women’s Hospital from the US Public Health Service.

    Miriam E. Tucker is a freelance journalist based in the Washington, DC, area. She is a regular contributor to Medscape Medical News, with other work appearing in The Washington Post, NPR’s Shots blog, and Diatribe. She is on X at @MiriamETucker and on BlueSky at @miriametucker.bsky.social.

    Continue Reading

  • ROG Xbox Ally and ROG Xbox Ally X to Hit Canadian Stores on October 16

    ROG Xbox Ally and ROG Xbox Ally X to Hit Canadian Stores on October 16

    ASUS Computer International

    Gaming handhelds will be on show for the very first time at Gamescom 2025

    KEY POINTS

    • Ready for launch: On-shelf availability of ROG Xbox Ally and ROG Xbox Ally X slated for October 16, 2025

    • Optimized for handheld gaming: Xbox team has developed a new experience designed to make more games ready to play on supported handhelds

    • See ROG Ally in person: Both models are on show at the ROG booth at Gamescom 2025

    • Meet-and-Greet at the ROG booth: Actors Ned Luke and Shawn Fonteno will be in person at the booth to meet with fans

    ROG Xbox Ally and ROG Xbox Ally X to Hit Canadian Stores on October 16
    ROG Xbox Ally and ROG Xbox Ally X to Hit Canadian Stores on October 16

    TORONTO, Aug. 20, 2025 (GLOBE NEWSWIRE) — ASUS Republic of Gamers (ROG) today announced that ROG Xbox Ally and ROG Xbox Ally X — featuring cutting-edge AMD Ryzen™ Z2 Series processors — will be available on shelves on October 16, 2025. Both new gaming handhelds will be available to try out for the first time on the show floor of Gamescom 2025. ROG will have a massive presence at Gamescom 2025, boasting a full lineup of 2025 laptops to give gamers a taste of what next-gen hardware is capable of. Additionally, the stars of the latest ROG Travel campaign film, actors Ned Luke and Shawn Fonteno of Grand Theft Auto V fame, will be at the ROG booth to meet and greet fans.

    Ready for launch

    On October 16, ROG Xbox Ally and ROG Xbox Ally X will be available in Canada, Australia, Belgium, the Czech Republic, China (Xbox Ally X only), Denmark, Finland, France, Germany, Hong Kong, Italy, Ireland, Japan, Malaysia, Mexico, the Netherlands, New Zealand, Norway, Philippines, Poland, Portugal, Romania, Saudi Arabia, Singapore, South Korea, Spain, Sweden, Switzerland, Taiwan, Turkey, United Arab Emirates, United Kingdom, United States and Vietnam.

    Availability will follow for other markets where ROG Ally series products are sold today, including Brazil, India, Indonesia, and Thailand. The Xbox Ally will launch in China early next year. Additional pricing and pre-order details will follow in the coming weeks.

    ROG and Xbox co-developed the ROG Xbox Ally and ROG Xbox Ally X to give gamers the freedom to play their way, anytime and anywhere.

    The ROG Xbox Ally is powered by an AMD Ryzen™ Z2 A processor featuring four Zen 2 cores with eight threads and eight AMD RDNA 2 GPU cores. This offers ultra-efficient performance, paired with 16GB of LPDDR5X‑6400 RAM and a 512GB M.2 SSD, and all backed by a 60Wh battery for extended play.

    The premium, high-performance ROG Xbox Ally X ups the ante with an AMD Ryzen AI Z2 Extreme, a new eight‑core/16‑thread Zen 5 APU with 16 RDNA 3.5 GPU cores, and an integrated NPU. The ROG Xbox Ally X also features 24GB of LPDDR5X‑8000 memory, a 1TB M.2 SSD, and boasts a larger 80Wh battery for longer playtime. Click here to learn more.

    Optimized for handheld

    The team at Xbox has been hard at work behind the scenes partnering with game studios to test and optimize thousands of PC titles for handheld compatibility. This new Handheld Compatibility Program ensures day-one users have the best experience possible on their ROG Xbox Ally handhelds. At launch, compatible games in the game library will sport Handheld Optimized or Mostly Compatible badges.

    Handheld Optimized means that the game is ready to go — with default controller inputs, an intuitive text input method, accurate iconography, clear text legibility, and appropriate resolution in full-screen mode. Mostly Compatible means that the game may require minor in-game setting changes for an optimal experience on handheld.

    The Xbox team is also bringing advanced shader delivery to the ROG Xbox Ally. This allows the Xbox app to preload a game’s shaders during download, so supported games will launch up to 10 x faster, run more smoothly, and use less battery on first play. Xbox is working on adding this feature to even more games over time.

    The ROG Xbox Ally X also features AMD’s cutting-edge Ryzen AI Z2 Extreme processor with a built-in NPU, that unlocks upcoming AI powered features starting early next year—with more to come. These features include:

    • Automatic Super Resolution (Auto SR): a system-level feature that uses the power of the NPU to upscale games running at lower resolutions. This delivers high-resolution visuals and smooth framerates across a wide range of games, with no additional changes required from game developers.

    • Highlight reels: AI captures standout gameplay moments—like epic boss battles or victories—and generates short replay clips to share with friends or on social channels.

    See ROG Xbox Ally in person

    Both handhelds, along with the entire 2025 ROG lineup that includes the latest Strix, Zephyrus, and Flow laptops, will be on show at the Gamescom 2025 ROG booth. The ROG Xbox Ally and ROG Xbox Ally X will be fully playable with a selection of games including Balatro, Clair Obscur: Expedition 33, DOOM: The Dark Ages, Final Fantasy VII Remake Intergrade, Gears of War: Reloaded, Hogwarts Legacy, Lies of P, and Roblox — featuring experiences including 99 Nights in the Forest, Grow a Garden, and Rivals.

    Meet-and-greet at the ROG booth

    To celebrate the upcoming launch of the ROG Xbox Ally and ROG Xbox Ally X, ROG is reuniting Ned Luke and Shawn Fonteno, one of gaming’s most iconic duos, for the latest instalment of ROG Travel now live on the ROG YouTube channel. Ned Luke reprises his role of the supervisor at ROG Travel, while Shawn Fonteno joins the cast as a new hire with strong suggestions for the ROG Xbox Ally as a ticket to send their customer on a perfect gaming getaway to Allywood, a place of dreams and infinite possibilities. Both Ned Luke and Shawn Fonteno will be live at the ROG booth, giving fans the exciting opportunity to meet them in person.

    Gamescom 2025 ROG booth information

    Hall 8 (North Entrance), Koelnmesse, Cologne

    Booth #C-010-B-011

    Thursday, August 21, 10am–8pm

    Friday, August 22, 10am–8pm

    Saturday, August 23, 9am–8pm

    Sunday, August 24, 9am–8pm

    All times are in Central European Summer Time (CEST)

    AVAILABILITY & PRICING

    The ROG Xbox Ally and ROG Xbox Ally X will be available in Canada starting October 16, 2025, across multiple national retailers. Pricing and pre-order details will be announced closer to launch.

    Stay tuned for more information at: https://ca.rog.gg/ROG_Xbox_Ally_CA

    NOTES TO EDITORS

    ASUS Homepage: https://ca.asus.click/ASUS_CA_Homepage

    ROG Xbox Ally Series: https://ca.rog.gg/ROG_Xbox_Ally_CA

    ROG Travel Video: https://youtu.be/9vLV1BjtlXI?si=XpphynmxPhMafbS6

    ROG YouTube Channel: https://www.youtube.com/@asusrog

    ROG 2025 Gaming Laptops: https://rog.asus.com/ca-en/content/2025-rog-gaming-laptops/

    ASUS Pressroom: http://press.asus.com

    ASUS Canada Facebook: https://www.facebook.com/asuscanada/

    ASUS Canada Instagram: https://www.instagram.com/asus_ca

    ASUS Canada YouTube: https://ca.asus.click/youtube

    ASUS Global X (Twitter): https://www.x.com/asus

    SPECIFICATIONS1

    ROG Xbox Ally X

    Operating System

    Windows 11 Home

    Ergonomics & input

    Contoured grips inspired by Xbox Wireless Controllers deliver all-day comfort, complete with impulse triggers for enhanced control.

    ABXY buttons / D-pad / L & R impulse triggers / L & R bumpers / Xbox button / View button / Menu button / Command Center button / Library button / 2 x assignable back buttons / 2 x full-size analog sticks / HD haptics / 6-Axis IMU

    Processor

    AMD Ryzen™ Z2 Extreme

    Display

    7” FHD (1080p) IPS, 500 nits, 16:9

    120Hz refresh rate

    AMD FreeSync™ Premium (Variable Refresh Rate)

    Corning® Gorilla® Glass Victus®

    Corning DXC Anti-Reflection

    Memory

    24GB LPDDR5X-8000

    Storage

    1TB M.2 2280 SSD

    Wireless

    Wi-Fi 6E (2 x 2) + Bluetooth® 5.4

    I/O ports

    1 x USB4® with DisplayPort™ 2.1 / Power Delivery 3.0, Thunderbolt™ 4 compatible

    1 x USB 3.2 Gen 2 Type-C® with DisplayPort™ 2.1 / Power Delivery 3.0

    1 x UHS-II microSD card reader (supports SD, SDXC and SDHC; UHS-I with DDR200 mode)

    1 x 3.5mm Combo Audio Jack

    Battery

    80Wh

    Colors

    Red, blue or white

    Size

    290.8 x 121.5 x 50.7mm (W x D x H) (11.45” × 4.78” × 2.00”)

    Weight

    715g (1.58 lbs)

    Includes

    ROG Xbox Ally X

    65W charger

    Stand

    ROG Xbox Ally (2025)

    Operating System

    Windows 11 Home

    Ergonomics & input

    Contoured grips inspired by Xbox Wireless Controllers deliver all-day comfort.

    ABXY buttons / D-pad / L & R Hall Effect analog triggers / L & R bumpers / Xbox button / View button / Menu button / Command Center button / Library button / 2 x assignable back buttons / 2 x full-size analog sticks / HD haptics / 6-Axis IMU

    Processor

    AMD Ryzen™ Z2 A

    Display

    7” FHD (1080p) IPS, 500 nits, 16:9

    120Hz refresh rate

    AMD FreeSync™ Premium (Variable Refresh Rate)

    Corning® Gorilla® Glass Victus®

    Corning DXC Anti-Reflection

    Memory

    16GB LPDDR5X-6400

    Storage

    512GB M.2 2280 SSD

    Wireless

    Wi-Fi 6E (2 x 2) + Bluetooth® 5.4

    I/O ports

    2 x USB 3.2 Gen 2 Type-C® with DisplayPort™ 1.4 / Power Delivery 3.0

    1 x UHS-II microSD card reader (supports SD, SDXC and SDHC)

    1 x 3.5mm Combo Audio Jack

    Battery

    60Wh

    Size

    290.8 x 121.5 x 50.7mm (W x D x H) (11.45” × 4.78” × 2.00”)

    Weight

    670g (1.48 lbs)

    Includes

    ROG Xbox Ally

    65W charger

    Stand

    About ROG

    Republic of Gamers (ROG) is an ASUS sub-brand dedicated to creating the world’s best gaming hardware and software. Formed in 2006, ROG offers a complete line of innovative products known for performance and quality, including motherboards, graphics cards, system components, laptops, desktops, monitors, smartphones, audio equipment, routers, peripherals and accessories. ROG participates in and sponsors major international gaming events. ROG gear has been used to set hundreds of overclocking records and it continues to be the preferred choice of gamers and enthusiasts around the world. To become one of those who dare, learn more about ROG at http://rog.asus.com.

    1 Specifications, content and product availability are all subject to change without notice and may differ from country to country. Actual performance may vary depending on applications, usage, environment and other factors. Full specifications are available at http://www.asus.com

    A photo accompanying this announcement is available at https://www.globenewswire.com/NewsRoom/AttachmentNg/30f177d8-f331-45a9-8960-42b8b231cf7b

    CONTACT: PRESS CONTACTS Media Relations ASUS Canada media.ca@asus.com Redoine Taoussi Public Relations Manager Redoine_Taoussi@asus.com


    Continue Reading

  • COVID-19 accelerates vascular aging in women

    COVID-19 accelerates vascular aging in women

    The world’s largest study of COVID-19 survivors shows the virus accelerates vascular aging, especially in women, while vaccination and recovery may help lessen the long-term damage.

    Study: Accelerated vascular ageing after COVID-19 infection: the CARTESIAN study. Image credit: Anatoliy Cherkas/Shutterstock.com

    A study published in European Heart Journal revealed that coronavirus disease 2019 (COVID-19) can increase arterial stiffness and accelerate vascular aging, especially in women.

    Background

    Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative pathogen of the COVID-19 pandemic, is persistently associated with significant morbidity and mortality worldwide, even after more than four years of its emergence. Besides acute illness, a large proportion of COVID-19 survivors are still experiencing long-term health complications, which is clinically defined as long-COVID.

    Cardiovascular events are among the most commonly reported long-COVID consequences, which have been observed for up to 12 months after infection. There is a gradient of risk according to the severity of acute COVID-19 infection. This is not surprising, as SARS-CoV-2 is known to directly or indirectly affect the vascular system.

    Identifying COVID-19 survivors who are at higher risk of developing long-term cardiovascular complications is, therefore, essential to protect them through pharmacological or non-pharmacological measures.

    Measurement of arterial stiffness is an effective method to assess vascular aging, a strong parameter for accurately classifying at-risk individuals. In contrast to chronological aging, vascular aging reflects individual variability in vascular disease onset and mortality.

    The CARTESIAN study is the first international multi-center study to explore whether COVID-19 survivors experience accelerated vascular ageing proportional to the severity of the infection.

    The CARTESIAN study

    The study recruited 2390 individuals from 38 centers in 18 countries. Analyses were performed on ~2,094 participants with vascular measurements available. Based on their COVID-19 status, the participants were categorized into four groups.

    The first group included participants with SARS-CoV-2-negative results (control group); the second group included non-hospitalized participants with confirmed SARS-CoV-2 infection; the third group included hospitalized participants with confirmed infection; and the fourth group included participants with confirmed infection who required intensive care unit (ICU) admission. All COVID-19 patients were assessed 6 ± 3 months after SARS-CoV-2 infection.

    All participants were evaluated for carotid-femoral pulse wave velocity, an established biomarker for large artery stiffness and vascular aging.

    Key findings

    The study reported that all participants with confirmed SARS-CoV-2 infection have a significantly higher large artery stiffness than SARS-CoV-2-negative participants. The gender-specific analysis revealed that women with confirmed infection have significantly higher large artery stiffness than those without infection, irrespective of COVID-19 severity. However, no significant difference was observed between men with and without confirmed infection.

    Among infected women, the increase in arterial stiffness compared with controls was ≈ +0.55-0.60 m/s in non-hospitalized and hospitalized cases, and ≈ +1.09 m/s in those admitted to the ICU. Furthermore, women with persistent COVID-19 symptoms had significantly higher arterial stiffness than fully recovered women, regardless of disease severity and cardiovascular confounders.

    The study included another round of vascular measurements taken from the participants at the second follow-up visit, approximately 12 months from the first follow-up visit. These measurements indicated a stable or improved large artery stiffness over time in participants with confirmed infection. In contrast, non-infected participants exhibited increased large artery stiffness, which may be due to chronological aging.

    Study significance

    The study reveals that COVID-19 can significantly accelerate vascular aging regardless of disease severity, particularly in women. Among various cardiovascular risk factors, the study finds that the association between COVID-19 and vascular aging is only partly mediated by elevated blood pressure. The 12-month follow-up findings indicate that the increased arterial stiffness partially attenuates in the long term.

    The study identifies factors positively or negatively associated with accelerated vascular aging in women with COVID-19. These factors are vaccination, which was associated with lower arterial stiffness in women at six months and remained associated with lower stiffness at ~ 12 months, especially in hospitalized groups, and persistent COVID-19 symptoms, which increase the risk of arterial stiffness. However, causality cannot be inferred.

    Evidence regarding COVID-19-related vascular damage suggests that SARS-CoV-2 can alter the functionality of vascular endothelial cells, that viral RNA can persist in these cells, and subsequently induce chronic inflammatory responses, leading to vascular damage.

    An increased vascular inflammation has been observed in the early post-infection phase in patients with severe COVID-19, which may trigger fibrotic changes and initiate the long-lasting process of arterial stiffening.

    Some small-scale studies have previously reported endothelial dysfunction and arterial stiffness up to six months after an acute COVID-19 infection. However, the current study is the first large-scale study to accurately demonstrate COVID-19-induced vascular ageing and its relationship with disease severity, independent of cardiovascular risk factor burden.  

    The increased susceptibility to vascular aging observed in women could be due to the differences in immune system function between females and males. Females exhibit more rapid and robust innate and adaptive immune responses than males, which might accelerate their recovery from initial infection and protect them against severe disease. However, the same difference can increase their susceptibility to prolonged autoimmune-related diseases.

    The study reports that Asians and Latin Americans have lower arterial stiffness than Caucasians in the COVID-19 negative group, but not in the COVID-19 positive group. This finding suggests that the ethnic benefits of cardiovascular fitness can be offset by SARS-CoV-2 infection.

    The study links COVID-19 with mid-term and long-term accelerated vascular aging, especially in women. Further studies are needed to determine whether these preclinical changes are associated with clinical cardiovascular events, and whether newer SARS-CoV-2 variants or SARS-CoV-2 reinfections are associated with accelerated vascular ageing to the same extent.

    Download your PDF copy now!

    Continue Reading

  • DXC Named a Leader in Everest Group’s Custom Application Development Services PEAK Matrix® Assessment 2025 Report

    DXC Named a Leader in Everest Group’s Custom Application Development Services PEAK Matrix® Assessment 2025 Report

    ASHBURN, Va., August 20, 2025 – DXC Technology (NYSE: DXC), a leading Fortune 500 global technology services provider, has been named a Leader by global research firm Everest Group in its new report, “Custom Application Development Services PEAK Matrix® Assessment 2025.” This recognition underscores DXC’s leadership in building differentiated, scalable and secure custom applications tailored to business needs. 

    Everest Group highlighted DXC’s strengths in leveraging GenAI platforms, as well as in efficiency and productivity, identifying and recruiting top-tier talent with critical expertise, and maintaining high delivery standards. DXC’s GenAI platform equips software engineers with a catalogue of powerful, governed and secure AI-powered assets and services, including coding assistants to accelerate software development, simplify complexities and reduce operational costs.  

    “At DXC, we manage over 20,000 applications for clients globally and have transformed over 2 billion lines of code. We help enterprises streamline, modernize and accelerate their most critical applications, enabling greater agility and growth,” said Anand Srivastava, Global Service Line and Capability Lead for Custom Application Services at DXC. “Our deep industry expertise and AI-driven innovation delivers measurable outcomes for our customers, and we’re honored to have our work recognized by Everest Group.” 

    Serving 2,000 customers in over 70 countries across industries including transportation, retail, telecommunications, and energy, DXC’s intelligent automation and data-driven systems have resulted in a 25% acceleration of application development and 40% faster application testing for customers. 

    “DXC Technology has established itself as a Leader in the custom application development space, backed by its strong in-house capabilities and consistent delivery performance,” says Ankit Gupta, Vice President at Everest Group. “With its AI platforms, the company accelerates software development by leveraging GenAI driven automation, significantly boosting productivity and efficiency. Clients value DXC’s ability to identify and onboard top-tier talent, as well as its steadfast commitment to delivery excellence. These strengths have contributed to DXC’s positioning as a Leader in Everest Group’s Custom Application Development Services PEAK Matrix® Assessment 2025.” 

    The PEAK Matrix® is a framework to assess leading application services providers’ relative market success and overall capability. The assessment is based on a comprehensive evaluation of 31 leading technology providers inclusive of case studies, interactions with service providers, client reference checks, and an ongoing analysis of the application services market. Leaders are placed based on their vision and strategy, ecosystem investments, ability to stay ahead of market trends, and maintenance of growth momentum. 

    As a trusted Custom Applications Services partner to enterprises across the globe, DXC empowers customers to take advantage of the latest digital platforms with custom applications, enabling increased resiliency, new product launches, and entrance into additional markets with minimal disruption. 

    An excerpt of Everest Group’s report is available to view here. To learn more about DXC’s Custom Application Services, click here.  

    About DXC Technology  

    DXC Technology (NYSE: DXC) is a leading global provider of information technology services. We’re a trusted operating partner to many of the world’s most innovative organizations, building solutions that move industries and companies forward. Our engineering, consulting and technology experts help clients simplify, optimize and modernize their systems and processes, manage their most critical workloads, integrate AI-powered intelligence into their operations, and put security and trust at the forefront. Learn more on dxc.com. 

    Disclaimer  

    Licensed extracts taken from Everest Group’s PEAK Matrix® Reports, may be used by licensed third parties for use in their own marketing and promotional activities and collateral. Selected extracts from Everest Group’s PEAK Matrix® reports do not necessarily provide the full context of our research and analysis.  All research and analysis conducted by Everest Group’s analysts and included in Everest Group’s PEAK Matrix® reports is independent and no organization has paid a fee to be featured or to influence their ranking.  To access the complete research and to learn more about our methodology, please visit Everest Group PEAK Matrix® Reports.  

    About Everest Group  

    Everest Group is a leading global research firm helping business leaders make confident decisions. Everest Group’s PEAK Matrix® assessments provide the analysis and insights enterprises need to make critical selection decisions about global services providers, locations, and products and solutions within various market segments. Likewise, providers of these services, products, and solutions, look to the PEAK Matrix® to gauge and calibrate their offerings against others in the industry or market. Find further details and in-depth content at www.everestgrp.com. 

    Continue Reading

  • A Case Report Exploring Early-Onset Alzheimer’s Disease With No Known Family History

    A Case Report Exploring Early-Onset Alzheimer’s Disease With No Known Family History


    Continue Reading