Microbiome Data Research Articles

TimeNorm: a novel normalization method for time course microbiome data.

Metagenomic time-course studies provide valuable insights into the dynamics of microbial systems and have become increasingly popular alongside the reduction in costs of next-generation sequencing technologies. Normalization is a common but critical preprocessing step before proceeding with downstream analysis. To the best of our knowledge, currently there is no reported method to appropriately normalize microbial time-series data. We propose TimeNorm, a novel normalization method that considers the compositional property and time dependency in time-course microbiome data. It is the first method designed for normalizing time-series data within the same time point (intra-time normalization) and across time points (bridge normalization), separately. Intra-time normalization normalizes microbial samples under the same condition based on common dominant features. Bridge normalization detects and utilizes a group of most stable features across two adjacent time points for normalization. Through comprehensive simulation studies and application to a real study, we demonstrate that TimeNorm outperforms existing normalization methods and boosts the power of downstream differential abundance analysis.

Relationship between the rumen microbiome and liver transcriptome in beef cattle divergent for feed efficiency

BackgroundFeed costs account for a high proportion of the variable cost of beef production, ultimately impacting overall profitability. Thus, improving feed efficiency of beef cattle, by way of determining the underlying genomic control and selecting for feed efficient cattle provides a method through which feed input costs may be reduced whilst also contributing to the environmental sustainability of beef production. The rumen microbiome dictates the feed degradation capacity and consequent nutrient supply in ruminants, thus potentially impacted by feed efficiency phenotype. Equally, liver tissue has been shown to be responsive to feed efficiency phenotype as well as dietary intake. However, although both the rumen microbiome and liver transcriptome have been shown to be impacted by host feed efficiency phenotype, knowledge of the interaction between the rumen microbiome and other peripheral tissues within the body, including the liver is lacking. Thus, the objective of this study was to compare two contrasting breed types (Charolais and Holstein-Friesian) divergent for residual feed intake (RFI) over contrasting dietary phases (zero-grazed grass and high-concentrate), based on gene co-expression network analysis of liver transcriptome data and microbe co-abundance network of rumen microbiome data. Traits including RFI, dry matter intake (DMI) and growth rate (ADG), as well as rumen concentrations of volatile fatty acids were also included within the network analysis.ResultsOverall, DMI had the greatest number of connections followed by RFI, with ADG displaying the fewest number of significant connections. Hepatic genes related to lipid metabolism were correlated to both RFI and DMI phenotypes, whilst genes related to immune response were correlated to DMI. Despite the known relationship between RFI and DMI, the same microbes were not directly connected to these phenotypes, the Succiniclasticum genus was however, negatively connected to both RFI and ADG. Additionally, a stepwise regression analysis revealed significant roles for both Succiniclasticum genus and Roseburia.faecis sp. in predicting RFI, DMI and ADG.ConclusionsResults from this study highlight the interactive relationships between rumen microbiome and hepatic transcriptome data of cattle divergent for RFI, whilst also increasing our understanding of the underlying biology of both DMI and ADG in beef cattle.

Taxanorm: a novel taxa-specific normalization approach for microbiome data

BackgroundIn high-throughput sequencing studies, sequencing depth, which quantifies the total number of reads, varies across samples. Unequal sequencing depth can obscure true biological signals of interest and prevent direct comparisons between samples. To remove variability due to differential sequencing depth, taxa counts are usually normalized before downstream analysis. However, most existing normalization methods scale counts using size factors that are sample specific but not taxa specific, which can result in over- or under-correction for some taxa.ResultsWe developed TaxaNorm, a novel normalization method based on a zero-inflated negative binomial model. This method assumes the effects of sequencing depth on mean and dispersion vary across taxa. Incorporating the zero-inflation part can better capture the nature of microbiome data. We also propose two corresponding diagnosis tests on the varying sequencing depth effect for validation. We find that TaxaNorm achieves comparable performance to existing methods in most simulation scenarios in downstream analysis and reaches a higher power for some cases. Specifically, it balances power and false discovery control well. When applying the method in a real dataset, TaxaNorm has improved performance when correcting technical bias.ConclusionTaxaNorm both sample- and taxon- specific bias by introducing an appropriate regression framework in the microbiome data, which aids in data interpretation and visualization. The ‘TaxaNorm’ R package is freely available through the CRAN repository https://CRAN.R-project.org/package=TaxaNorm and the source code can be downloaded at https://github.com/wangziyue57/TaxaNorm.

Methanogenic Potential of Sewer Microbiomes and Its Implications for Methane Emission.

The sewer system, despite being a significant source of methane emissions, has often been overlooked in current greenhouse gas inventories due to the limited availability of quantitative data. Direct monitoring in sewers can be expensive or biased due to access limitations and internal heterogeneity of sewer networks. Fortunately, since methane is almost exclusively biogenic in sewers, we demonstrate in this study that the methanogenic potential can be estimated using known sewer microbiome data. By combining data mining techniques and bioinformatics databases, we developed the first data-driven method to analyze methanogenic potentials using a data set containing 633 observations of 53 variables obtained from literature mining. The methanogenic potential in the sewer sediment was around 250-870% higher than that in the wet biofilm on the pipe and sewage water. Additionally, k-means clustering and principal component analysis linked higher methane emission rates (9.72 ± 51.3 kgCO2eq m-3 d-1) with smaller pipe size, higher water level, and higher potentials of sulfate reduction in the wetted pipe biofilm. These findings exhibit the possibility of connecting microbiome data with biogenic greenhouse gases, further offering insights into new approaches for understanding greenhouse gas emissions from understudied sources.

Metformin increases gut multidrug resistance genes in type 2 diabetes, potentially linked to Escherichia coli

Metformin is the most commonly prescribed medication for treating type 2 diabetes (T2D). It is known that metformin can alter the gut microbiome, which influences the effectiveness of metformin treatment. We posited that if the gut microbiome, a reservoir of the resistome, is altered, then the resistome should change as well. To test this hypothesis, we reanalyzed microbiome data generated by Wu et al. (Nat Med 23(7):850–858, 2017), identifying antibiotic resistance genes (ARGs) and bacterial species. Through read-based analysis, we observed that the abundance of ARGs indeed changed in many samples treated with metformin. Moreover, the altered pattern was sufficiently heterogeneous across individual samples to allow subcategorization. We also found a strong correlation between the abundance of multidrug-resistant ARGs (MDR-ARGs) and the presence of E. coli. The contig-based analysis led to the same conclusion: an increase in MDR-ARGs due to metformin was associated with an increase in E. coli. In relation to this, we were able to confirm that the majority of MDR-ARGs are likely to originate from E. coli. These results suggest that metformin may have the potential side effect of increasing E. coli carrying ARGs, particularly MDR-ARGs, which could be a concern in T2D therapy that relies on metformin.

PSIX-30 Interaction between rumen microbiome and host feed efficiency phenotype across contrasting breeds and dietary sources

Abstract Provision of feed accounts for a high proportion of the variable cost in beef production systems, and consequently is a major determinant of overall profitability. Thus, improving the conversion of feed to animal product by identifying and breeding cattle with improved feed efficiency potential provides a method through which feed input costs may be reduced, whilst also contributing to more environmentally sustainable beef production. The rumen microbiome dictates the feed degradation capacity and subsequent nutrient supply in ruminants and is thus potentially impacted by feed efficiency phenotype. However, the relationship between rumen microbiota and host feed efficiency phenotype is yet to be fully elucidated. This is further complicated when contrasting diets, cattle breeds as well as various stages of animal development are considered. The objective of this study was to undertake network analysis on rumen microbiome data generated from two contrasting steer breed types (Charolais and Holstein-Friesian), divergent within-breed for residual feed intake (RFI), and offered contrasting diets during different stages of development: i) high-concentrate during the growing phase; ii) grass silage during the growing phase; iii) zero-grazed grass during the growing phase, and iv) high-concentrate during the finishing phase. From approximately 10 mo of age, Charolais (n = 90) and Holstein-Friesian (n = 77) steers were individually fed each of the aforementioned diets for 70 d, following an adaptation period between each diet. At the end of each dietary phase, RFI was determined for every steer and rumen fluid was sampled using a transesophageal sampling device from the 10 most-efficient (Low-RFI) and the 10 least-efficient (High-RFI) steers within each breed. Rumen fluid samples were then subjected to 16S rRNA sequencing and a network-based systems biology analysis subsequently undertaken using PCIT incorporating the 16S sequencing data and individual RFI values across each breed and dietary phase. A total of 110 microbial taxa were connected (P < 0.05; r > 0.8) to the breed and diet contrasts examined; however, no single microbe was commonly connected to all contrasts. The Pyramidobacter genus was the most represented phylotype, displaying a positive connection with the Charolais steers during the grass silage diet, as well as a negative association with Charolais steers during the first high-concentrate diet and Holstein-Friesian steers during the zero-grazed grass diet. Similarly, of the 35 microbial taxa significantly connected to more than one breed/diet contrast, 12 displayed altered direction of association, indicating a differential relationship between the rumen microbiome and RFI phenotype depending on the prevailing dietary management. Acknowledgement: The animal model used in this study was funded by the Irish Department of Agriculture, Food and the Marine (RSF13/S/519). Kate Keogh received funding for this work from the Research Leaders 2025 program (co-funded by Teagasc and the European Union’s Horizon 2020, Marie Skłodowska-Curie grant agreement number 754380).

Cultured fecal microbial community and its impact as fecal microbiota transplantation treatment in mice gut inflammation

The fecal microbiome is identical to the gut microbial communities and provides an easy access to the gut microbiome. Therefore, fecal microbial transplantation (FMT) strategies have been used to alter dysbiotic gut microbiomes with healthy fecal microbiota, successfully alleviating various metabolic disorders, such as obesity, type 2 diabetes, and inflammatory bowel disease (IBD). However, the success of FMT treatment is donor-dependent and variations in gut microbes cannot be avoided. This problem may be overcome by using a cultured fecal microbiome. In this study, a human fecal microbiome was cultured using five different media; growth in brain heart infusion (BHI) media resulted in the highest microbial community cell count. The microbiome (16S rRNA) data demonstrated that the cultured microbial communities were similar to that of the original fecal sample. Therefore, the BHI-cultured fecal microbiome was selected for cultured FMT (cFMT). Furthermore, a dextran sodium sulfate (DSS)-induced mice-IBD model was used to confirm the impact of cFMT. Results showed that cFMT effectively alleviated IBD-associated symptoms, including improved gut permeability, restoration of the inflamed gut epithelium, decreased expression of pro-inflammatory cytokines (IFN-γ, TNF-α, IL-1, IL-6, IL-12, and IL-17), and increased expression of anti-inflammatory cytokines (IL-4 and IL-10). Thus, study’s findings suggest that cFMT can be a potential alternative to nFMT.Key points• In vitro fecal microbial communities were grown in a batch culture using five different media.• Fecal microbial transplantation was performed on DSS-treated mice using cultured and normal fecal microbes.• Cultured fecal microbes effectively alleviated IBD-associated symptoms.Graphical

Toward Multilabel Classification for Multiple Disease Prediction Using Gut Microbiota Profiles.

Advancements in high-throughput technologies have yielded large-scale human gut microbiota profiles, sparking considerable interest in exploring the relationship between the gut microbiome and complex human diseases. Through extracting and integrating knowledge from complex microbiome data, existing machine learning (ML)-based studies have demonstrated their effectiveness in the precise identification of high-risk individuals. However, these approaches struggle to address the heterogeneity and sparsity of microbial features and explore the intrinsic relatedness among human diseases. In this work, we reframe human gut microbiome-based disease detection as a multilabel classification (MLC) problem and integrate a range of innovative techniques within the proposed MLC framework, aptly named GutMLC. Specifically, the entity semantic similarity as priori knowledge is incorporated into multilabel feature selection and loss functions by capturing the shared attributes and inherent associations among diseases and microbes. To tackle the issue of label imbalance, both within and between labels, we adapt the focal loss (FL) function for MLC using debiased inverse weighting. Extensive experiment results consistently demonstrate the competitive performance of GutMLC in comparison with commonly used MLC and single-label classification (SLC) algorithms. This work seeks to unlock the potential of gut microbiota as robust biomarkers for multiple disease prediction.

Meta-analysis of oral microbiome reveals sex-based diversity in biofilms during periodontitis.

Sex is an often overlooked, yet compulsory, biological variable when performing biomedical research. Periodontitis is a common yet progressively debilitating chronic inflammatory disorder affecting the tissues supporting teeth that ultimately leads to tooth loss if left untreated. The incidence of periodontitis is sex biased, with increased prevalence in males compared with females but with unknown etiology. We performed a sex-specific meta-analysis using publicly available oral microbiome data from different sampling sites of patients with periodontitis and periodontally healthy controls; sex balance was established for each periodontal health condition. Our results show sex-based diversity in oral biofilms of individuals with periodontitis but not in their saliva, with increased abundance of several periodontal pathogens in subgingival plaques from females compared with males. We devised a quantitative measure, uniquely defined as the Microsexome Index (MSI), which indicates that sexual dimorphism in subgingival bacterial composition is a distinct feature of reduced microbial diversity during periodontitis but not under healthy conditions. In addition, we found that smoking exacerbates microsexome diversity in supragingival biofilms, particularly during periodontitis. Taken together, we provide insights regarding sex-based diversity in periodontitis, a disease with multiorgan associations, and provide the rationale for further mechanistic, diagnostic, and therapeutic studies.

Predicting superagers: a machine learning approach utilizing gut microbiome features.

Cognitive decline is often considered an inevitable aspect of aging; however, recent research has identified a subset of older adults known as "superagers" who maintain cognitive abilities comparable to those of younger individuals. Investigating the neurobiological characteristics associated with superior cognitive function in superagers is essential for understanding "successful aging." Evidence suggests that the gut microbiome plays a key role in brain function, forming a bidirectional communication network known as the microbiome-gut-brain axis. Alterations in the gut microbiome have been linked to cognitive aging markers such as oxidative stress and inflammation. This study aims to investigate the unique patterns of the gut microbiome in superagers and to develop machine learning-based predictive models to differentiate superagers from typical agers. We recruited 161 cognitively unimpaired, community-dwelling volunteers aged 60 years or from dementia prevention centers in Seoul, South Korea. After applying inclusion and exclusion criteria, 115 participants were included in the study. Following the removal of microbiome data outliers, 102 participants, comprising 57 superagers and 45 typical agers, were finally analyzed. Superagers were defined based on memory performance at or above average normative values of middle-aged adults. Gut microbiome data were collected from stool samples, and microbial DNA was extracted and sequenced. Relative abundances of bacterial genera were used as features for model development. We employed the LightGBM algorithm to build predictive models and utilized SHAP analysis for feature importance and interpretability. The predictive model achieved an AUC of 0.832 and accuracy of 0.764 in the training dataset, and an AUC of 0.861 and accuracy of 0.762 in the test dataset. Significant microbiome features for distinguishing superagers included Alistipes, PAC001137_g, PAC001138_g, Leuconostoc, and PAC001115_g. SHAP analysis revealed that higher abundances of certain genera, such as PAC001138_g and PAC001115_g, positively influenced the likelihood of being classified as superagers. Our findings demonstrate the machine learning-based predictive models using gut-microbiome features can differentiate superagers from typical agers with a reasonable performance.

Stable biomarker discovery in multi-omics data via canonical correlation analysis.

Multi-omics analysis offers a promising avenue to a better understanding of complex biological phenomena. In particular, untangling the pathophysiology of multifactorial health conditions such as the inflammatory bowel disease (IBD) could benefit from simultaneous consideration of several omics levels. However, taking full advantage of multi-omics data requires the adoption of suitable new tools. Multi-view learning, a machine learning technique that natively joins together heterogeneous data, is a natural source for such methods. Here we present a new approach to variable selection in unsupervised multi-view learning by applying stability selection to canonical correlation analysis (CCA). We apply our method, StabilityCCA, to simulated and real multi-omics data, and demonstrate its ability to find relevant variables and improve the stability of variable selection. In a case study on an IBD microbiome data set, we link together metagenomics and metabolomics, revealing a connection between their joint structure and the disease, and identifying potential biomarkers. Our results showcase the usefulness of multi-view learning in multi-omics analysis and demonstrate StabilityCCA as a powerful tool for biomarker discovery.

Using Large Language Models for Microbiome Findings Reports in Laboratory Diagnostics

Background: Advancements in genomic technologies are rapidly evolving, with the potential to transform laboratory diagnostics by enabling high-throughput analysis of complex biological data, such as microbiome data. Large Language Models (LLMs) have shown significant promise in extracting actionable insights from vast datasets, but their application in generating microbiome findings reports with clinical interpretations and lifestyle recommendations has not been explored yet. Methods: This article introduces an innovative framework that utilizes LLMs to automate the generation of findings reports in the context of microbiome diagnostics. The proposed model integrates LLMs within an event-driven, workflow-based architecture, designed to enhance scalability and adaptability in clinical laboratory environments. Special focus is given to aligning the model with clinical standards and regulatory guidelines such as the In-Vitro Diagnostic Regulation (IVDR) and the guidelines published by the High-Level Expert Group on Artificial Intelligence (HLEG AI). The implementation of this model was demonstrated through a prototype called “MicroFlow”. Results: The implementation of MicroFlow indicates the viability of automating findings report generation using LLMs. Initial evaluation by laboratory expert users indicated that the integration of LLMs is promising, with the generated reports being plausible and useful, although further testing on real-world data is necessary to assess the model’s accuracy and reliability. Conclusions: This work presents a potential approach for using LLMs to support the generation of findings reports in microbiome diagnostics. While the initial results seem promising, further evaluation and refinement are needed to ensure the model’s effectiveness and adherence to clinical standards. Future efforts will focus on improvements based on feedback from laboratory experts and comprehensive testing on real patient data.

Combining genomics and semen microbiome increases the accuracy of predicting bull prolificacy.

Commercial livestock producers need to prioritize genetic progress for health and efficiency traits to address productivity, welfare, and environmental concerns but face challenges due to limited pedigree information in extensive multi-sire breeding scenarios. Utilizing pooled DNA for genotyping and integrating seminal microbiome information into genomic models could enhance predictions of male fertility traits, thus addressing complexities in reproductive performance and inbreeding effects. Using the Angus Australia database comprising genotypes and pedigree data for 78,555 animals, we simulated percentage of normal sperm (PNS) and prolificacy of sires, resulting in 713 sires and 27,557 progeny in the final dataset. Publicly available microbiome data from 45 bulls was used to simulate data for the 713 sires. By incorporating both genomic and microbiome information our models were able to explain a larger proportion of phenotypic variation in both PNS (0.94) and prolificacy (0.56) compared to models using a single data source (e.g., 0.36 and 0.41, respectively, using only genomic information). Additionally, models containing both genomic and microbiome data revealed larger phenotypic differences between animals in the top and bottom quartile of predictions, indicating potential for improved productivity and sustainability in livestock farming systems. Inbreeding depression was observed to affect fertility traits, which makes the incorporation of microbiome information on the prediction of fertility traits even more actionable. Crucially, our inferences demonstrate the potential of the semen microbiome to contribute to the improvement of fertility traits in cattle and pave the way for the development of targeted microbiome interventions to improve reproductive performance in livestock.

Integrating genome-and transcriptome-wide association studies to uncover the host-microbiome interactions in bovine rumen methanogenesis.

The ruminal microbiota generates biogenic methane in ruminants. However, the role of host genetics in modifying ruminal microbiota-mediated methane emissions remains mysterious, which has severely hindered the emission control of this notorious greenhouse gas. Here, we uncover the host genetic basis of rumen microorganisms by genome-and transcriptome-wide association studies with matched genome, rumen transcriptome, and microbiome data from a cohort of 574 Holstein cattle. Heritability estimation revealed that approximately 70% of microbial taxa had significant heritability, but only 43 genetic variants with significant association with 22 microbial taxa were identified through a genome-wide association study (GWAS). In contrast, the transcriptome-wide association study (TWAS) of rumen microbiota detected 28,260 significant gene-microbe associations, involving 210 taxa and 4652 unique genes. On average, host genetic factors explained approximately 28% of the microbial abundance variance, while rumen gene expression explained 43%. In addition, we highlighted that TWAS exhibits a strong advantage in detecting gene expression and phenotypic trait associations in direct effector organs. For methanogenic archaea, only one significant signal was detected by GWAS, whereas the TWAS obtained 1703 significant associated host genes. By combining multiple correlation analyses based on these host TWAS genes, rumen microbiota, and volatile fatty acids, we observed that substrate hydrogen metabolism is an essential factor linking host-microbe interactions in methanogenesis. Overall, these findings provide valuable guidelines for mitigating methane emissions through genetic regulation and microbial management strategies in ruminants.

Technical note: A biological anthropologist's guide for applying microbiome science to studies of human and non-human primates.

A central goal of biological anthropology is connecting environmental variation to differences in host physiology, biology, health, and evolution. The microbiome represents a valuable pathway for studying how variation in host environments impacts health outcomes. While there are many resources for learning about methods related to microbiome sample collection, laboratory analyses, and genetic sequencing, there are fewer dedicated to helping researchers navigate the dense portfolio of bioinformatics and statistical approaches for analyzing microbiome data. Those that do exist are rarely related to questions in biological anthropology and instead are often focused on human biomedicine. To address this gap, we expand on existing tutorials and provide a "road map" to aid biological anthropologists in understanding, selecting, and deploying the data analysis and visualization methods that are most appropriate for their specific research questions. Leveraging an existing dataset of fecal samples and survey data collected from wild geladas living in Simien Mountains National Park in Ethiopia (Baniel et al., 2021), this paper guides researchers toward answering three questions related to variation in the gut microbiome across host and environmental factors. By providing explanations, examples, and a reproducible workflow for different analytic methods, we move beyond the theoretical benefits of considering the microbiome within anthropological research and instead present researchers with a guide for applying microbiome science to their work. This paper makes microbiome science more accessible to biological anthropologists and paves the way for continued research into the microbiome's role in the ecology, evolution, and health of human and non-human primates.

Bacterial microbiome diversity along poultry slaughtering lines: insights from chicken carcasses and environmental sources

Abstract Introduction This study aimed to determine the bacterial diversity of chicken carcasses and their surrounding environment at various stages along a poultry slaughter line. Material and Methods Amplicon sequencing of the 16S rRNA gene was employed to assess the shifts in bacterial community diversity at both phylum and genus levels. Samples were collected from September to November 2021, targeting carcass surfaces at various operational stages (post-defeathering, post-evisceration, post-water chilling, and post-cooling), as well as from the internal environments and air of these units. The study took place in a vertically integrated poultry slaughterhouse in Konya, Turkey. Results Microbial diversity increased after the chilling and storage stages as a result of redistribution of the microorganisms after the physical effect of the slaughtering stages. The final product sample taken after storage had the highest bacterial abundance. The abundance at this stage was found to be strongly correlated with that at other slaughtering stages, as well as with the abundance in chilling water and on the personnel’s hands. The common genera in chicken carcasses during slaughter stages were Macrococcus, Acinetobacter, Enterococcus, Escherichia-Shigella, Psychrobacter, Streptococcus, Lactococcus and Ligilactobacillus. Microbiome data in environmental samples indicated that the genera in highest relative abundance were Bacillus, Anoxybacillus, Acinetobacter and Psychrobacter. In air samples, the storage room had the highest diversity and in this place Bacillus spp. and Staphylococcus spp. were in the majority. Conclusion This study may provide some useful information to pinpoint the critical contamination sources in the poultry slaughtering process.

Gut Microbiomics of Sustained Knee Pain in Knee Osteoarthritis Patients.

To examine whether gut microbes were associated with post-surgery sustained knee pain in knee osteoarthritis (OA) patients by a gut microbiomics approach. Total knee replacement (TKR) patients due to primary knee OA were recruited. Sustained knee pain status at least one year after TKR was defined by the Western Ontario and McMaster University Arthritis Index (WOMAC). Fasting plasma sample and fecal sample were collected. Metabolomic profiling was performed on fasting plasma. 16S rRNA sequencing was performed on fecal samples to determine microbial composition. Twenty TKR patients due to primary knee OA were included in the study with 10 experiencing sustained post-surgery pain and 10 without such pain. Age, sex, and BMI were matched. Linear discriminant analysis of microbiome data identified 13 bacterial taxa that were highly abundant in the pain group, and 5 that were highly abundant in the non-pain group (all P < 0.05). Plasma metabolomic profiling measured 622 metabolites. The correlation analysis indicated that the 18 taxa were significantly correlated with 231 metabolites (all P < 0.05). sPLS-DA analysis showed that 30 out of the 231 metabolites explained 29% of total variance and can be used to clearly separate sustained knee pain patients from non-pain patients. Pathway enrichment analysis showed that these significant metabolites were enriched in arachidonic acid metabolic pathway, bile acid biosynthesis, and linoleic acid metabolism. The gut microbes may play a significant role in sustained knee pain in knee OA patients after TKR potentially through their activation of inflammatory pathways, lipid metabolism, and central sensitization.

Rank-based correlation matrix estimation for high dimensional microbiome data

In recent years, the progress made in high-throughout sequencing techniques supports the analysis of microbial communities greatly. The fact that real-world microbiome data lie in a high-dimensional simplex makes conventional statistical analysis unreliable owing to the sum-to-one constraint. Motivated by the urgent requirements of directly inferring the dependence pattern and dependence intensity simultaneously among microbial taxa, we address the challenge of correlation matrix estimation for this kind of data. Assuming that the unobserved log-basis random vectors follow the elliptical distribution, a robust method built on the Kendall's tau statistic is developed to estimate the basis generalized correlation matrix. Regardless of the existence of the covariance, the basis generalized correlation matrix is always a reliable measure of dependence structure. Theoretically, we establish the convergence rate of the proposed estimator under the spectral norm. Support recovery is discussed as well. Empirically, simulation studies demonstrate the outstanding numerical performance of our method. We also apply our method to analyse a microbiome dataset from human gut. We provide all the codes at https://github.com/lwfwhunanhero/CRCE.

Host diet drives gut microbiome convergence between coral reef fishes and mammals.

Animal gut microbiomes are critical to host physiology and fitness. The gut microbiomes of fishes-the most abundant and diverse vertebrate clade-have received little attention relative to other clades. Coral reef fishes, in particular, make up a wide range of evolutionary histories and feeding ecologies that are likely associated with gut microbiome diversity. The repeated evolution of herbivory in fishes and mammals also allows us to examine microbiome similarity in relationship to diet across the entire vertebrate tree of life. Here, we generate a large coral reef fish gut microbiome dataset (n = 499 samples, 19 species) and combine it with a diverse aggregation of public microbiome data (n = 447) to show that host diet drives significant convergence between coral reef fish and mammalian gut microbiomes. We demonstrate that this similarity is largely driven by carnivory and herbivory and that herbivorous and carnivorous hosts exhibit distinct microbial compositions across fish and mammals. We also show that fish and mammal gut microbiomes share prominent microbial taxa, including Ruminoccocus spp. and Akkermansia spp., and predicted metabolic pathways. Despite the major evolutionary and ecological differences between fishes and mammals, our results reveal that their gut microbiomes undergo similar dietary selective pressures. Thus, diet, in addition to phylosymbiosis must be considered even when comparing the gut microbiomes of distantly related hosts.

Machine learning surveillance of foodborne infectious diseases using wastewater microbiome, crowdsourced, and environmental data

Clostridium perfringens (CP) is a common cause of foodborne infection, leading to significant human health risks and a high economic burden. Thus, effective CP disease surveillance is essential for preventive and therapeutic interventions; however, conventional practices often entail complex, resource-intensive, and costly procedures. This study introduced a data-driven machine learning (ML) modeling framework for CP-related disease surveillance. It leveraged an integrated dataset of municipal wastewater microbiome (e.g., CP abundance), crowdsourced (CP-related web search keywords), and environmental data. Various optimization strategies, including data integration, data normalization, model selection, and hyperparameter tuning, were implemented to improve the ML modeling performance, leading to enhanced predictions of CP cases over time. Explainable artificial intelligence methods identified CP abundance as the most reliable predictor of CP disease cases. Multi-omics subsequently revealed the presence of CP and its genotypes/toxinotypes in wastewater, validating the utility of microbiome-data-enabled ML surveillance for foodborne diseases. This ML-based framework thus exhibits significant potential for complementing and reinforcing existing disease surveillance systems.