HMP16SData: Efficient Access to the Human Microbiome Project Through Bioconductor

A quantitative approach to measure and predict microbiome response to antibiotics

Although antibiotics induce sizable perturbations in the human microbiome, we lack a systematic and quantitative method to measure and predict the microbiome's response to specific antibiotics. Here, we introduce such a method, which takes the form of a microbiome response index (MiRIx) for each antibiotic. Antibiotic-specific MiRIx values quantify the overall susceptibility of the microbiota to an antibiotic, based on databases of bacterial phenotypes and published data on intrinsic antibiotic susceptibility. We applied our approach to five published microbiome studies that carried out antibiotic interventions with vancomycin, metronidazole, ciprofloxacin, amoxicillin, and doxycycline. We show how MiRIx can be used in conjunction with existing microbiome analytical approaches to gain a deeper understanding of the microbiome response to antibiotics. Finally, we generate antibiotic response predictions for the oral, skin, and gut microbiome in healthy humans. Our approach is implemented as open-source software and is readily applied to microbiome data sets generated by 16S rRNA marker gene sequencing or shotgun metagenomics.

IMPORTANCE

Antibiotics are potent influencers of the human microbiome and can be a source for enduring dysbiosis and antibiotic resistance in healthcare. Existing microbiome data analysis methods can quantify perturbations of bacterial communities but cannot evaluate whether the differences are aligned with the expected activity of a specific antibiotic. Here, we present a novel method to quantify and predict antibiotic-specific microbiome changes, implemented in a ready-to-use software package. This has the potential to be a critical tool to broaden our understanding of the relationship between the microbiome and antibiotics.

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

BACKGROUND

In 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.

RESULTS

We 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.

CONCLUSION

TaxaNorm 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 .

Obesity-associated microbiomes instigate visceral adipose tissue inflammation by recruitment of distinct neutrophils

Neutrophils are increasingly implicated in chronic inflammation and metabolic disorders. Here, we show that visceral adipose tissue (VAT) from individuals with obesity contains more neutrophils than in those without obesity and is associated with a distinct bacterial community. Exploring the mechanism, we gavaged microbiome-depleted mice with stool from patients with and without obesity during high-fat or normal diet administration. Only mice receiving high-fat diet and stool from subjects with obesity show enrichment of VAT neutrophils, suggesting donor microbiome and recipient diet determine VAT neutrophilia. A rise in pro-inflammatory CD4+ Th1 cells and a drop in immunoregulatory T cells in VAT only follows if there is a transient spike in neutrophils. Human VAT neutrophils exhibit a distinct gene expression pattern that is found in different human tissues, including tumors. VAT neutrophils and bacteria may be a novel therapeutic target for treating inflammatory-driven complications of obesity, including insulin resistance and colon cancer.

ADAPT: Analysis of Microbiome Differential Abundance by Pooling Tobit Models

Microbiome differential abundance analysis remains a challenging problem despite multiple methods proposed in the literature. The excessive zeros and compositionality of metagenomics data are two main challenges for differential abundance analysis. We propose a novel method called "analysis of differential abundance by pooling Tobit models" (ADAPT) to overcome these two challenges. ADAPT uniquely treats zero counts as left-censored observations to facilitate computation and enhance interpretation. ADAPT also encompasses a theoretically justified way of selecting non-differentially abundant microbiome taxa as a reference for hypothesis testing. We generate synthetic data using independent simulation frameworks to show that ADAPT has more consistent false discovery rate control and higher statistical power than competitors. We use ADAPT to analyze 16S rRNA sequencing of saliva samples and shotgun metagenomics sequencing of plaque samples collected from infants in the COHRA2 study. The results provide novel insights into the association between the oral microbiome and early childhood dental caries.

Microbial Biosurfactant as an Alternate to Chemical Surfactants for Application in Cosmetics Industries in Personal and Skin Care Products: A Critical Review

Cosmetics and personal care items are used worldwide and administered straight to the skin. The hazardous nature of the chemical surfactant utilized in the production of cosmetics has caused alarm on a global scale. Therefore, bacterial biosurfactants (BS) are becoming increasingly popular in industrial product production as a biocompatible, low-toxic alternative surfactant. Chemical surfactants can induce allergic responses and skin irritations; thus, they should be replaced with less harmful substances for skin health. The cosmetic industry seeks novel biological alternatives to replace chemical compounds and improve product qualities. Most of these chemicals have a biological origin and can be obtained from plant, bacterial, fungal, and algal sources. Various biological molecules have intriguing capabilities, such as biosurfactants, vitamins, antioxidants, pigments, enzymes, and peptides. These are safe, biodegradable, and environmentally friendly than chemical options. Plant-based biosurfactants, such as saponins, offer numerous advantages over synthetic surfactants, i.e., biodegradable, nontoxic, and environmentally friendly nature. Saponins are a promising source of natural biosurfactants for various industrial and academic applications. However, microbial glycolipids and lipopeptides have been used in biotechnology and cosmetics due to their multifunctional character, including detergency, emulsifying, foaming, and skin moisturizing capabilities. In addition, some of them have the potential to be used as antibacterial agents. In this review, we like to enlighten the application of microbial biosurfactants for replacing chemical surfactants in existing cosmetic and personal skincare pharmaceutical formulations due to their antibacterial, skin surface moisturizing, and low toxicity characteristics.

Stochastic variational variable selection for high-dimensional microbiome data

BACKGROUND

The rapid and accurate identification of a minimal-size core set of representative microbial species plays an important role in the clustering of microbial community data and interpretation of clustering results. However, the huge dimensionality of microbial metagenomics datasets is a major challenge for the existing methods such as Dirichlet multinomial mixture (DMM) models. In the approach of the existing methods, the computational burden of identifying a small number of representative species from a large number of observed species remains a challenge.

RESULTS

We propose a novel approach to improve the performance of the widely used DMM approach by combining three ideas: (i) we propose an indicator variable to identify representative operational taxonomic units that substantially contribute to the differentiation among clusters; (ii) to address the computational burden of high-dimensional microbiome data, we propose a stochastic variational inference, which approximates the posterior distribution using a controllable distribution called variational distribution, and stochastic optimization algorithms for fast computation; and (iii) we extend the finite DMM model to an infinite case by considering Dirichlet process mixtures and estimating the number of clusters as a variational parameter. Using the proposed method, stochastic variational variable selection (SVVS), we analyzed the root microbiome data collected in our soybean field experiment, the human gut microbiome data from three published datasets of large-scale case-control studies and the healthy human microbiome data from the Human Microbiome Project.

CONCLUSIONS

SVVS demonstrates a better performance and significantly faster computation than those of the existing methods in all cases of testing datasets. In particular, SVVS is the only method that can analyze massive high-dimensional microbial data with more than 50,000 microbial species and 1000 samples. Furthermore, a core set of representative microbial species is identified using SVVS that can improve the interpretability of Bayesian mixture models for a wide range of microbiome studies. Video Abstract.

Human Microbiome Mixture Analysis Using Weighted Quantile Sum Regression

Studies of the health effects of the microbiome often measure overall associations by using diversity metrics, and individual taxa associations in separate analyses, but do not consider the correlated relationships between taxa in the microbiome. In this study, we applied random subset weighted quantile sum regression with repeated holdouts (WQSRSRH), a mixture method successfully applied to 'omic data to account for relationships between many predictors, to processed amplicon sequencing data from the Human Microbiome Project. We simulated a binary variable associated with 20 operational taxonomic units (OTUs). WQSRSRH was used to test for the association between the microbiome and the simulated variable, adjusted for sex, and sensitivity and specificity were calculated. The WQSRSRH method was also compared to other standard methods for microbiome analysis. The method was further illustrated using real data from the Growth and Obesity Cohort in Chile to assess the association between the gut microbiome and body mass index. In the analysis with simulated data, WQSRSRH predicted the correct directionality of association between the microbiome and the simulated variable, with an average sensitivity and specificity of 75% and 70%, respectively, in identifying the 20 associated OTUs. WQSRSRH performed better than all other comparison methods. In the illustration analysis of the gut microbiome and obesity, the WQSRSRH analysis identified an inverse association between body mass index and the gut microbe mixture, identifying Bacteroides, Clostridium, Prevotella, and Ruminococcus as important genera in the negative association. The application of WQSRSRH to the microbiome allows for analysis of the mixture effect of all the taxa in the microbiome, while simultaneously identifying the most important to the mixture, and allowing for covariate adjustment. It outperformed other methods when using simulated data, and in analysis with real data found results consistent with other study findings.

A comprehensive evaluation of microbial differential abundance analysis methods: current status and potential solutions

BACKGROUND

Differential abundance analysis (DAA) is one central statistical task in microbiome data analysis. A robust and powerful DAA tool can help identify highly confident microbial candidates for further biological validation. Numerous DAA tools have been proposed in the past decade addressing the special characteristics of microbiome data such as zero inflation and compositional effects. Disturbingly, different DAA tools could sometimes produce quite discordant results, opening to the possibility of cherry-picking the tool in favor of one's own hypothesis. To recommend the best DAA tool or practice to the field, a comprehensive evaluation, which covers as many biologically relevant scenarios as possible, is critically needed.

RESULTS

We performed by far the most comprehensive evaluation of existing DAA tools using real data-based simulations. We found that DAA methods explicitly addressing compositional effects such as ANCOM-BC, Aldex2, metagenomeSeq (fitFeatureModel), and DACOMP did have improved performance in false-positive control. But they are still not optimal: type 1 error inflation or low statistical power has been observed in many settings. The recent LDM method generally had the best power, but its false-positive control in the presence of strong compositional effects was not satisfactory. Overall, none of the evaluated methods is simultaneously robust, powerful, and flexible, which makes the selection of the best DAA tool difficult. To meet the analysis needs, we designed an optimized procedure, ZicoSeq, drawing on the strength of the existing DAA methods. We show that ZicoSeq generally controlled for false positives across settings, and the power was among the highest. Application of DAA methods to a large collection of real datasets revealed a similar pattern observed in simulation studies.

CONCLUSIONS

Based on the benchmarking study, we conclude that none of the existing DAA methods evaluated can be applied blindly to any real microbiome dataset. The applicability of an existing DAA method depends on specific settings, which are usually unknown a priori. To circumvent the difficulty of selecting the best DAA tool in practice, we design ZicoSeq, which addresses the major challenges in DAA and remedies the drawbacks of existing DAA methods. ZicoSeq can be applied to microbiome datasets from diverse settings and is a useful DAA tool for robust microbiome biomarker discovery. Video Abstract.

CBEA: Competitive balances for taxonomic enrichment analysis

Research in human-associated microbiomes often involves the analysis of taxonomic count tables generated via high-throughput sequencing. It is difficult to apply statistical tools as the data is high-dimensional, sparse, and compositional. An approachable way to alleviate high-dimensionality and sparsity is to aggregate variables into pre-defined sets. Set-based analysis is ubiquitous in the genomics literature and has demonstrable impact on improving interpretability and power of downstream analysis. Unfortunately, there is a lack of sophisticated set-based analysis methods specific to microbiome taxonomic data, where current practice often employs abundance summation as a technique for aggregation. This approach prevents comparison across sets of different sizes, does not preserve inter-sample distances, and amplifies protocol bias. Here, we attempt to fill this gap with a new single-sample taxon enrichment method that uses a novel log-ratio formulation based on the competitive null hypothesis commonly used in the enrichment analysis literature. Our approach, titled competitive balances for taxonomic enrichment analysis (CBEA), generates sample-specific enrichment scores as the scaled log-ratio of the subcomposition defined by taxa within a set and the subcomposition defined by its complement. We provide sample-level significance testing by estimating an empirical null distribution of our test statistic with valid p-values. Herein, we demonstrate, using both real data applications and simulations, that CBEA controls for type I error, even under high sparsity and high inter-taxa correlation scenarios. Additionally, CBEA provides informative scores that can be inputs to downstream analyses such as prediction tasks.

Beyond samples: A metric revealing more connections of gut microbiota between individuals

Studies of gut microbiota explore their complicated connections between individuals of different characteristics by applying different metrics to abundance data obtained from fecal samples. Although classic metrics are capable to quantify differences between samples, the microbiome of fecal sample is not a good surrogate for the gut microbiome of individuals because the microbial populations of the distal colon does not adequately represent that of the entire gastrointestinal tract. To overcome the deficiency of classic metrics in which the differences can be measured between the samples analyzed, but not the corresponding populations, we propose a metric for representing composition differences in the gut microbiota of individuals. Our investigation shows this metric outperforms traditional measures for multiple scenarios. For gut microbiota in diverse geographic populations, this metric presents more explainable data variance than others, not only in regular variance analysis but also in principle component analysis and partition analysis of biologic characteristics. With time-series data, the metric further presents a strong correlation with the time interval of serial sampling. Our findings suggest that the metric is robust and powerfully detects the intrinsic variations in gut microbiota. The metric holds promise for revealing more relations between gut microbiota and human health.

mbImpute: an accurate and robust imputation method for microbiome data

A critical challenge in microbiome data analysis is the existence of many non-biological zeros, which distort taxon abundance distributions, complicate data analysis, and jeopardize the reliability of scientific discoveries. To address this issue, we propose the first imputation method for microbiome data-mbImpute-to identify and recover likely non-biological zeros by borrowing information jointly from similar samples, similar taxa, and optional metadata including sample covariates and taxon phylogeny. We demonstrate that mbImpute improves the power of identifying disease-related taxa from microbiome data of type 2 diabetes and colorectal cancer, and mbImpute preserves non-zero distributions of taxa abundances.

treeclimbR pinpoints the data-dependent resolution of hierarchical hypotheses

treeclimbR is for analyzing hierarchical trees of entities, such as phylogenies or cell types, at different resolutions. It proposes multiple candidates that capture the latent signal and pinpoints branches or leaves that contain features of interest, in a data-driven way. It outperforms currently available methods on synthetic data, and we highlight the approach on various applications, including microbiome and microRNA surveys as well as single-cell cytometry and RNA-seq datasets. With the emergence of various multi-resolution genomic datasets, treeclimbR provides a thorough inspection on entities across resolutions and gives additional flexibility to uncover biological associations.

Direct interaction network inference for compositional data via codaloss

16S rRNA gene sequencing and whole microbiome sequencing make it possible and stable to quantitatively analyze the composition of microbial communities and the relationship among microbial communities, microbes, and hosts. One essential step in the analysis of microbiome compositional data is inferring the direct interaction network among microbial species, bringing to light the potential underlying mechanism that regulates interaction in their communities. However, standard statistical analysis may obtain spurious results due to compositional nature of microbiome data; therefore, network recovery of microbial communities remains challenging. Here, we propose a novel loss function called codaloss for direct microbes interaction network estimation under the sparsity assumptions. We develop an alternating direction optimization algorithm to obtain sparse solution of codaloss as estimator. Compared to other state-of-the-art methods, our model makes less assumptions about the microbial networks. The simulation and real microbiome data results show that our method outperforms other methods in network inference. An implementation of codaloss is available from https://github.com/xuebaliang/Codaloss.

TreeSummarizedExperiment: a S4 class for data with hierarchical structure

Data organized into hierarchical structures (e.g., phylogenies or cell types) arises in several biological fields. It is therefore of interest to have data containers that store the hierarchical structure together with the biological profile data, and provide functions to easily access or manipulate data at different resolutions. Here, we present TreeSummarizedExperiment, a R/S4 class that extends the commonly used SingleCellExperiment class by incorporating tree representations of rows and/or columns (represented by objects of the phylo class). It follows the convention of the SummarizedExperiment class, while providing links between the assays and the nodes of a tree to allow data manipulation at arbitrary levels of the tree. The package is designed to be extensible, allowing new functions on the tree (phylo) to be contributed. As the work is based on the SingleCellExperiment class and the phylo class, both of which are popular classes used in many R packages, it is expected to be able to interact seamlessly with many other tools.

TreeSummarizedExperiment: a S4 class for data with hierarchical structure

Data organized into hierarchical structures (e.g., phylogenies or cell types) arises in several biological fields. It is therefore of interest to have data containers that store the hierarchical structure together with the biological profile data, and provide functions to easily access or manipulate data at different resolutions. Here, we present TreeSummarizedExperiment, a R/S4 class that extends the commonly used SingleCellExperiment class by incorporating tree representations of rows and/or columns (represented by objects of the phylo class). It follows the convention of the SummarizedExperiment class, while providing links between the assays and the nodes of a tree to allow data manipulation at arbitrary levels of the tree. The package is designed to be extensible, allowing new functions on the tree (phylo) to be contributed. As the work is based on the SingleCellExperiment class and the phylo class, both of which are popular classes used in many R packages, it is expected to be able to interact seamlessly with many other tools.

Assessment of statistical methods from single cell, bulk RNA-seq, and metagenomics applied to microbiome data

BACKGROUND

The correct identification of differentially abundant microbial taxa between experimental conditions is a methodological and computational challenge. Recent work has produced methods to deal with the high sparsity and compositionality characteristic of microbiome data, but independent benchmarks comparing these to alternatives developed for RNA-seq data analysis are lacking.

RESULTS

We compare methods developed for single-cell and bulk RNA-seq, and specifically for microbiome data, in terms of suitability of distributional assumptions, ability to control false discoveries, concordance, power, and correct identification of differentially abundant genera. We benchmark these methods using 100 manually curated datasets from 16S and whole metagenome shotgun sequencing.

CONCLUSIONS

The multivariate and compositional methods developed specifically for microbiome analysis did not outperform univariate methods developed for differential expression analysis of RNA-seq data. We recommend a careful exploratory data analysis prior to application of any inferential model and we present a framework to help scientists make an informed choice of analysis methods in a dataset-specific manner.

The food-gut axis: lactic acid bacteria and their link to food, the gut microbiome and human health

Lactic acid bacteria (LAB) are present in foods, the environment and the animal gut, although fermented foods (FFs) are recognized as the primary niche of LAB activity. Several LAB strains have been studied for their health-promoting properties and are employed as probiotics. FFs are recognized for their potential beneficial effects, which we review in this article. They are also an important source of LAB, which are ingested daily upon FF consumption. In this review, we describe the diversity of LAB and their occurrence in food as well as the gut microbiome. We discuss the opportunities to study LAB diversity and functional properties by considering the availability of both genomic and metagenomic data in public repositories, as well as the different latest computational tools for data analysis. In addition, we discuss the role of LAB as potential probiotics by reporting the prevalence of key genomic features in public genomes and by surveying the outcomes of LAB use in clinical trials involving human subjects. Finally, we highlight the need for further studies aimed at improving our knowledge of the link between LAB-fermented foods and the human gut from the perspective of health promotion.

Comparison of lung microbiota between antineutrophil cytoplasmic antibody-associated vasculitis and sarcoidosis

Microbial involvement in the pathogenesis have been suggested in both antineutrophil cytoplasmic antibody-associated vasculitis (AAV) and sarcoidosis, both of which have lung involvement. However, exhaustive research to assess the bacteria in the lung in AAV and in sarcoidosis have not been performed. We sought to elucidate the distinct dysbiotic lung microbiota between AAV and sarcoidosis. We used 16S rRNA gene high-throughput sequencing to obtain the bacterial community composition of bronchoalveolar lavage fluid (BALF) in patients with AAV (n = 16) compared to patients with sarcoidosis (n = 21). The patients had not undergone therapy with immunosuppressive medication when their BALF was acquired. No difference was observed in α-diversity between patients with AAV and patients with sarcoidosis when using all the detected taxa. We defined the taxa of the oral cavity by using the data of oral microbiota of healthy individuals from the Human Microbiome Project (HMP). The analysis using only oral taxa made the difference in α-diversity between AAV and sarcoidosis clearer compared with those using all the detected taxa. Besides, the analysis using detected taxa except for oral taxa also made the difference in α-diversity between AAV and sarcoidosis clearer compared with those using all the detected taxa. A linear negative relationship between the α-diversity and Birmingham vasculitis activity score (BVAS) was detected in the AAV group. The observed p-value for the effect of the disease groups on the ß-diversity was small while the effect of other factors including sex and smoking status did not have small p-values. By excluding oral taxa from all the detected taxa, we found a cluster mainly consisted of sarcoidosis patients which was characterized with microbial community monopolized by Erythrobacteraceae family. Our results suggested the importance of considering the influence of oral microbiota in evaluating lung microbiota.

IPCO: Inference of Pathways from Co-variance analysis

Background

Key aspects of microbiome research are the accurate identification of taxa and the profiling of their functionality. Amplicon profiling based on the 16S ribosomal DNA sequence is a ubiquitous technique to identify and profile the abundance of the various taxa. However, it does not provide information on their encoded functionality. Predictive tools that can accurately extrapolate the functional information of a microbiome based on taxonomic profile composition are essential. At present, the applicability of these tools is limited due to requirement of reference genomes from known species. We present IPCO (Inference of Pathways from Co-variance analysis), a new method of inferring functionality for 16S-based microbiome profiles independent of reference genomes. IPCO utilises the biological co-variance observed between paired taxonomic and functional profiles and co-varies it with the queried dataset.

Results

IPCO outperforms other established methods both in terms of sample and feature profile prediction. Validation results confirmed that IPCO can replicate observed biological associations between shotgun and metabolite profiles. Comparative analysis of predicted functionality profiles with other popular 16S-based functional prediction tools showed significantly lower performances with predicted functionality showing little to no correlation with paired shotgun features across samples.

Conclusions

IPCO can infer functionality from 16S datasets and significantly outperforms existing tools. IPCO is implemented in R and available from https://github.com/IPCO-Rlibrary/IPCO.

Microbiome data science

Best practices from open data science are spreading across research fields, providing new opportunities for research and education. Open data science emphasizes the view that digitalization is enabling new forms of resource sharing, collaboration and outreach. This has the potential to improve the overall transparency and efficiency of research. Microbiome bioinformatics is a rapidly developing area that can greatly benefit from this progress. The concept of microbiome data science refers to the application of best practices from open data science to microbiome bioinformatics. The increasing availability of open data and new opportunities to collaborate online are greatly facilitating the development of this field. A microbiome data science ecosystem combines experimental research data with open data processing and analysis and reproducible tutorials that can also serve as an educational resource. Here, we provide an overview of the current status of microbiome data science from a community developer perspective and propose directions for future development of the field.

Invited Commentary: Improving the Accessibility of Human Microbiome Project Data Through Integration With R/Bioconductor

Alterations in the composition of the microbiota have been implicated in many diseases. The Human Microbiome Project (HMP) provides a comprehensive reference data set of the "normal" human microbiome of 242 healthy adults at 5 major body sites. The HMP used both 16S ribosomal RNA gene sequencing and whole-genome metagenomic sequencing to profile the subjects' microbial communities. However, accessing and analyzing the HMP data set still presents technical and bioinformatic challenges, given that researchers must import the microbiome data, integrate phylogenetic trees, and access and merge public and restricted metadata. The HMP16SData R/Bioconductor package developed by Schiffer et al. (Am J Epidemiol. 2019;188(6):1023-1026) greatly simplifies access to the HMP data by combining 16S taxonomic abundance data, public patient metadata, and phylogenetic trees as a single data object. The authors also provide an interface for users with approved Database of Genotypes and Phenotypes (dbGaP) projects to easily retrieve and merge the controlled-access HMP metadata. This package has a broad range of appeal to researchers across disciplines and with various levels of expertise in using R and/or other statistical tools, which translates to improved data accessibility for public health research, with data from healthy individuals serving as a reference for disease-associated studies.

Statistical evaluation of diet-microbe associations

Background

Statistical evaluation of the association between microbial abundance and dietary variables can be done in various ways. Currently, there is no consensus on which methods are to be preferred in which circumstances. Application of particular methods seems to be based on the tradition of a particular research group, availability of experience with particular software, or depending on the outcomes of the analysis.

Results

We applied four popular methods including edgeR, limma, metagenomeSeq and shotgunFunctionalizeR, to evaluate the association between dietary variables and abundance of microbes. We found large difference in results between the methods. Our simulation studies revealed that no single method was optimal.

Conclusions

We advise researchers to run multiple analyses and focus on the significant findings identified by multiple methods in order to achieve a better control of false discovery rate, although the false discovery rate can still be substantial.