Kernel-based genetic association analysis for microbiome phenotypes identifies host genetic drivers of beta-diversity.

BACKGROUND: Understanding human genetic influences on the gut microbiota helps elucidate the mechanisms by which genetics may influence health outcomes. Typical microbiome genome-wide association studies (GWAS) marginally assess the association between individual genetic variants and individual microbial taxa. We propose a novel approach, the covariate-adjusted kernel RV (KRV) framework, to map genetic variants associated with microbiome beta-diversity, which focuses on overall shifts in the microbiota. The KRV framework evaluates the association between genetics and microbes by comparing similarity in genetic profiles, based on groups of variants at the gene level, to similarity in microbiome profiles, based on the overall microbiome composition, across all pairs of individuals. By reducing the multiple-testing burden and capturing intrinsic structure within the genetic and microbiome data, the KRV framework has the potential of improving statistical power in microbiome GWAS. RESULTS: We apply the covariate-adjusted KRV to the Hispanic Community Health Study/Study of Latinos (HCHS/SOL) in a two-stage (first gene-level, then variant-level) genome-wide association analysis for gut microbiome beta-diversity. We have identified an immunity-related gene, IL23R, reported in a previous microbiome genetic association study and discovered 3 other novel genes, 2 of which are involved in immune functions or autoimmune disorders. In addition, simulation studies show that the covariate-adjusted KRV has a greater power than other microbiome GWAS methods that rely on univariate microbiome phenotypes across a range of scenarios. CONCLUSIONS: Our findings highlight the value of the covariate-adjusted KRV as a powerful microbiome GWAS approach and support an important role of immunity-related genes in shaping the gut microbiome composition. Video Abstract.

Exploration of associations among dietary tryptophan, microbiome composition and function, and symptom severity in irritable bowel syndrome.

BACKGROUND: Imbalance of the tryptophan (TRP) pathway may influence symptoms among patients with irritable bowel syndrome (IBS). This study explored relationships among different components that contribute to TRP metabolism (dietary intake, stool metabolite levels, predicted microbiome metabolic capability) in females with IBS and healthy controls (HCs). Within the IBS group, we also investigated relationships between TRP metabolic determinants, Bifidobacterium abundance, and symptoms of IBS. METHODS: Participants with IBS (Rome III) and HCs completed a 28-day diary of gastrointestinal symptoms and a 3-day food record for TRP intake. They provided a stool sample for shotgun metagenomics, 16 S rRNA analyses, and quantitative measurement of TRP by mass spectrometry. RESULTS: Our cohort included 115 females, 69 with IBS and 46 HCs, with a mean age of 28.5 years (SD 7.4). TRP intake (p = 0.71) and stool TRP level (p = 0.27) did not differ between IBS and HC. Bifidobacterium abundance was lower in the IBS group than in HCs (p = 0.004). Predicted TRP metabolism gene content was higher in IBS than HCs (FDR-corrected q = 0.006), whereas predicted biosynthesis gene content was lower (q = 0.045). Within the IBS group, there was no association between symptom severity and TRP intake or stool TRP, but there was a significant interaction between Bifidobacterium abundance and TRP intake (q = 0.029) in predicting stool character. CONCLUSIONS: Dietary TRP intake, microbiome composition, and differences in TRP metabolism constitute a complex interplay of factors that could modulate IBS symptom severity.

Impact of Data and Study Characteristics on Microbiome Volatility Estimates.

The human microbiome is a dynamic community of bacteria, viruses, fungi, and other microorganisms. Both the composition of the microbiome (the microbes that are present and their relative abundances) and the temporal variability of the microbiome (the magnitude of changes in their composition across time, called volatility) has been associated with human health. However, the effect of unbalanced sampling intervals and differential read depth on the estimates of microbiome volatility has not been thoroughly assessed. Using four publicly available gut and vaginal microbiome time series, we subsampled the datasets to several sampling intervals and read depths and then compared additive, multiplicative, centered log ratio (CLR)-based, qualitative, and distance-based measures of microbiome volatility between the conditions. We find that longer sampling intervals are associated with larger quantitative measures of change (particularly for common taxa), but not with qualitative measures of change or distance-based volatility quantification. A lower sequencing read depth is associated with smaller multiplicative, CLR-based, and qualitative measures of change (particularly for less common taxa). Strategic subsampling may serve as a useful sensitivity analysis in unbalanced longitudinal studies investigating clinical associations with microbiome volatility.

Accommodating multiple potential normalizations in microbiome associations studies.

BACKGROUND: Microbial communities are known to be closely related to many diseases, such as obesity and HIV, and it is of interest to identify differentially abundant microbial species between two or more environments. Since the abundances or counts of microbial species usually have different scales and suffer from zero-inflation or over-dispersion, normalization is a critical step before conducting differential abundance analysis. Several normalization approaches have been proposed, but it is difficult to optimize the characterization of the true relationship between taxa and interesting outcomes.  RESULTS: To avoid the challenge of picking an optimal normalization and accommodate the advantages of several normalization strategies, we propose an omnibus approach. Our approach is based on a Cauchy combination test, which is flexible and powerful by aggregating individual p values. We also consider a truncated test statistic to prevent substantial power loss. We experiment with a basic linear regression model as well as recently proposed powerful association tests for microbiome data and compare the performance of the omnibus approach with individual normalization approaches. Experimental results show that, regardless of simulation settings, the new approach exhibits power that is close to the best normalization strategy, while controling the type I error well.  CONCLUSIONS: The proposed omnibus test releases researchers from choosing among various normalization methods and it is an aggregated method that provides the powerful result to the underlying optimal normalization, which requires tedious trial and error. While the power may not exceed the best normalization, it is always much better than using a poor choice of normalization.

Batch effects removal for microbiome data via conditional quantile regression.

Batch effects in microbiome data arise from differential processing of specimens and can lead to spurious findings and obscure true signals. Strategies designed for genomic data to mitigate batch effects usually fail to address the zero-inflated and over-dispersed microbiome data. Most strategies tailored for microbiome data are restricted to association testing or specialized study designs, failing to allow other analytic goals or general designs. Here, we develop the Conditional Quantile Regression (ConQuR) approach to remove microbiome batch effects using a two-part quantile regression model. ConQuR is a comprehensive method that accommodates the complex distributions of microbial read counts by non-parametric modeling, and it generates batch-removed zero-inflated read counts that can be used in and benefit usual subsequent analyses. We apply ConQuR to simulated and real microbiome datasets and demonstrate its advantages in removing batch effects while preserving the signals of interest.

Testing microbiome association using integrated quantile regression models.

MOTIVATION: Most existing microbiome association analyses focus on the association between microbiome and conditional mean of health or disease-related outcomes, and within this vein, vast computational tools and methods have been devised for standard binary or continuous outcomes. However, these methods tend to be limited either when the underlying microbiome-outcome association occurs somewhere other than the mean level, or when distribution of the outcome variable is irregular (e.g. zero-inflated or mixtures) such that conditional outcome mean is less meaningful. We address this gap by investigating association analysis between microbiome compositions and conditional outcome quantiles. RESULTS: We introduce a new association analysis tool named MiRKAT-IQ within the Microbiome Regression-based Kernel Association Test framework using Integrated Quantile regression models to examine the association between microbiome and the distribution of outcome. For an individual quantile, we utilize the existing kernel machine regression framework to examine the association between that conditional outcome quantile and a group of microbial features (e.g. microbiome community compositions). Then, the goal of examining microbiome association with the whole outcome distribution is achieved by integrating all outcome conditional quantiles over a process, and thus our new MiRKAT-IQ test is robust to both the location of association signals (e.g. mean, variance, median) and the heterogeneous distribution of the outcome. Extensive numerical simulation studies have been conducted to show the validity of the new MiRKAT-IQ test. We demonstrate the potential usefulness of MiRKAT-IQ with applications to actual biological data collected from a previous microbiome study. AVAILABILITY AND IMPLEMENTATION: R codes to implement the proposed methodology is provided in the MiRKAT package, which is available on CRAN. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Powerful and robust non-parametric association testing for microbiome data via a zero-inflated quantile approach (ZINQ).

BACKGROUND: Identification of bacterial taxa associated with diseases, exposures, and other variables of interest offers a more comprehensive understanding of the role of microbes in many conditions. However, despite considerable research in statistical methods for association testing with microbiome data, approaches that are generally applicable remain elusive. Classical tests often do not accommodate the realities of microbiome data, leading to power loss. Approaches tailored for microbiome data depend highly upon the normalization strategies used to handle differential read depth and other data characteristics, and they often have unacceptably high false positive rates, generally due to unsatisfied distributional assumptions. On the other hand, many non-parametric tests suffer from loss of power and may also present difficulties in adjusting for potential covariates. Most extant approaches also fail in the presence of heterogeneous effects. The field needs new non-parametric approaches that are tailored to microbiome data, robust to distributional assumptions, and powerful under heterogeneous effects, while permitting adjustment for covariates. METHODS: As an alternative to existing approaches, we propose a zero-inflated quantile approach (ZINQ), which uses a two-part quantile regression model to accommodate the zero inflation in microbiome data. For a given taxon, ZINQ consists of a valid test in logistic regression to model the zero counts, followed by a series of quantile rank-score based tests on multiple quantiles of the non-zero part with adjustment for the zero inflation. As a regression and quantile-based approach, the method is non-parametric and robust to irregular distributions, while providing an allowance for covariate adjustment. Since no distributional assumptions are made, ZINQ can be applied to data that has been processed under any normalization strategy. RESULTS: Thorough simulations based on real data across a range of scenarios and application to real data sets show that ZINQ often has equivalent or higher power compared to existing tests even as it offers better control of false positives. CONCLUSIONS: We present ZINQ, a quantile-based association test between microbiota and dichotomous or quantitative clinical variables, providing a powerful and robust alternative for the current microbiome differential abundance analysis. Video Abstract.

A Comprehensive Self-Management Program With Diet Education Does Not Alter Microbiome Characteristics in Women With Irritable Bowel Syndrome.

BACKGROUND AND PURPOSE: Changes in diet and lifestyle factors are frequently recommended for persons with irritable bowel syndrome (IBS). It is unknown whether these recommendations alter the gut microbiome and/or whether baseline microbiome predicts improvement in symptoms and quality of life following treatment. Therefore, the purpose of this study was to explore if baseline gut microbiome composition predicted response to a Comprehensive Self-Management (CSM) intervention and if the intervention resulted in a different gut microbiome composition compared to usual care. METHODS: Individuals aged 18-70 years with IBS symptoms ≥6 months were recruited using convenience sampling. Individuals were excluded if medication use or comorbidities would influence symptoms or microbiome. Participants completed a baseline assessment and were randomized into the eight-session CSM intervention which included dietary education and cognitive behavioral therapy versus usual care. Questionnaires included demographics, quality of life, and symptom diaries. Fecal samples were collected at baseline and 3-month post-randomization for 16S rRNA-based microbiome analysis. RESULTS: Within the CSM intervention group (n = 30), Shannon diversity, richness, and beta diversity measures at baseline did not predict benefit from the CSM intervention at 3 months, as measured by change in abdominal pain and quality of life. Based on both alpha and beta diversity, the change from baseline to follow-up microbiome bacterial taxa did not differ between CSM (n = 25) and usual care (n = 25). CONCLUSIONS AND INFERENCES: Baseline microbiome does not predict symptom improvement with CSM intervention. We do not find evidence that the CSM intervention influences gut microbiome diversity or composition over the course of 3 months.

A Kernel-based Test of Independence for Cluster-correlated Data.

The Hilbert-Schmidt Independence Criterion (HSIC) is a powerful kernel-based statistic for assessing the generalized dependence between two multivariate variables. However, independence testing based on the HSIC is not directly possible for cluster-correlated data. Such a correlation pattern among the observations arises in many practical situations, e.g., family-based and longitudinal data, and requires proper accommodation. Therefore, we propose a novel HSIC-based independence test to evaluate the dependence between two multivariate variables based on cluster-correlated data. Using the previously proposed empirical HSIC as our test statistic, we derive its asymptotic distribution under the null hypothesis of independence between the two variables but in the presence of sample correlation. Based on both simulation studies and real data analysis, we show that, with clustered data, our approach effectively controls type I error and has a higher statistical power than competing methods.

MiRKAT: kernel machine regression-based global association tests for the microbiome.

SUMMARY: Distance-based tests of microbiome beta diversity are an integral part of many microbiome analyses. MiRKAT enables distance-based association testing with a wide variety of outcome types, including continuous, binary, censored time-to-event, multivariate, correlated and high-dimensional outcomes. Omnibus tests allow simultaneous consideration of multiple distance and dissimilarity measures, providing higher power across a range of simulation scenarios. Two measures of effect size, a modified R-squared coefficient and a kernel RV coefficient, are incorporated to allow comparison of effect sizes across multiple kernels. AVAILABILITY AND IMPLEMENTATION: MiRKAT is available on CRAN as an R package. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

pldist: ecological dissimilarities for paired and longitudinal microbiome association analysis.

MOTIVATION: The human microbiome is notoriously variable across individuals, with a wide range of 'healthy' microbiomes. Paired and longitudinal studies of the microbiome have become increasingly popular as a way to reduce unmeasured confounding and to increase statistical power by reducing large inter-subject variability. Statistical methods for analyzing such datasets are scarce. RESULTS: We introduce a paired UniFrac dissimilarity that summarizes within-individual (or within-pair) shifts in microbiome composition and then compares these compositional shifts across individuals (or pairs). This dissimilarity depends on a novel transformation of relative abundances, which we then extend to more than two time points and incorporate into several phylogenetic and non-phylogenetic dissimilarities. The data transformation and resulting dissimilarities may be used in a wide variety of downstream analyses, including ordination analysis and distance-based hypothesis testing. Simulations demonstrate that tests based on these dissimilarities retain appropriate type 1 error and high power. We apply the method in two real datasets. AVAILABILITY AND IMPLEMENTATION: The R package pldist is available on GitHub at https://github.com/aplantin/pldist. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.