A Review and Tutorial of Machine Learning Methods for Microbiome Host Trait Prediction

Explainable machine learning model for identifying key gut microbes and metabolites biomarkers associated with myasthenia gravis

Diagnostic markers for myasthenia gravis (MG) are limited; thus, innovative approaches are required for supportive diagnosis and personalized care. Gut microbes are associated with MG pathogenesis; however, few studies have adopted machine learning (ML) to identify the associations among MG, gut microbiota, and metabolites. In this study, we developed an explainable ML model to predict biomarkers for MG diagnosis. We enrolled 19 MG patients and 10 non-MG individuals. Stool samples were collected and microbiome assessment was performed using 16S rRNA sequencing. Untargeted metabolic profiling was conducted to identify fecal amplicon significant variants (ASVs) and metabolites. We developed an explainable ML model in which the top ASVs and metabolites are combined to identify the best predictive performance. This model uses the SHapley Additive exPlanations method to generate both global and personalized explanations. Fecal microbe-metabolite composition differed significantly between groups. The key bacterial families were Lachnospiraceae and Ruminococcaceae, and the top three features were Lachnospiraceae, inosine, and methylhistidine. An ML model trained with the top 1 % ASVs and top 15 % metabolites combined outperformed all other models. Personalized explanations revealed different patterns of microbe-metabolite contributions in patients with MG. The integration of the microbiota-metabolite features and the development of an explainable ML framework can accurately identify MG and provide personalized explanations, revealing the associations between gut microbiota, metabolites, and MG. An online calculator employing this algorithm was developed that provides a streamlined interface for MG diagnosis screening and conducting personalized evaluations.

Network impact of a single-time-point microbial sample

The human microbiome plays a crucial role in determining our well-being and can significantly influence human health. The individualized nature of the microbiome may reveal host-specific information about the health state of the subject. In particular, the microbiome is an ecosystem shaped by a tangled network of species-species and host-species interactions. Thus, analysis of the ecological balance of microbial communities can provide insights into these underlying interrelations. However, traditional methods for network analysis require many samples, while in practice only a single-time-point microbial sample is available in clinical screening. Recently, a method for the analysis of a single-time-point sample, which evaluates its 'network impact' with respect to a reference cohort, has been applied to analyze microbial samples from women with Gestational Diabetes Mellitus. Here, we introduce different variations of the network impact approach and systematically study their performance using simulated 'samples' fabricated via the Generalized Lotka-Volttera model of ecological dynamics. We show that the network impact of a single sample captures the effect of the interactions between the species, and thus can be applied to anomaly detection of shuffled samples, which are 'normal' in terms of species abundance but 'abnormal' in terms of species-species interrelations. In addition, we demonstrate the use of the network impact in binary and multiclass classifications, where the reference cohorts have similar abundance profiles but different species-species interactions. Individualized analysis of the human microbiome has the potential to improve diagnosis and personalized treatments.

A review of machine learning methods for cancer characterization from microbiome data

Recent studies have shown that the microbiome can impact cancer development, progression, and response to therapies suggesting microbiome-based approaches for cancer characterization. As cancer-related signatures are complex and implicate many taxa, their discovery often requires Machine Learning approaches. This review discusses Machine Learning methods for cancer characterization from microbiome data. It focuses on the implications of choices undertaken during sample collection, feature selection and pre-processing. It also discusses ML model selection, guiding how to choose an ML model, and model validation. Finally, it enumerates current limitations and how these may be surpassed. Proposed methods, often based on Random Forests, show promising results, however insufficient for widespread clinical usage. Studies often report conflicting results mainly due to ML models with poor generalizability. We expect that evaluating models with expanded, hold-out datasets, removing technical artifacts, exploring representations of the microbiome other than taxonomical profiles, leveraging advances in deep learning, and developing ML models better adapted to the characteristics of microbiome data will improve the performance and generalizability of models and enable their usage in the clinic.

Interpretable machine learning decodes soil microbiome's response to drought stress

BACKGROUND

Extreme weather events induced by climate change, particularly droughts, have detrimental consequences for crop yields and food security. Concurrently, these conditions provoke substantial changes in the soil bacterial microbiota and affect plant health. Early recognition of soil affected by drought enables farmers to implement appropriate agricultural management practices. In this context, interpretable machine learning holds immense potential for drought stress classification of soil based on marker taxa.

RESULTS

This study demonstrates that the 16S rRNA-based metagenomic approach of Differential Abundance Analysis methods and machine learning-based Shapley Additive Explanation values provide similar information. They exhibit their potential as complementary approaches for identifying marker taxa and investigating their enrichment or depletion under drought stress in grass lineages. Additionally, the Random Forest Classifier trained on a diverse range of relative abundance data from the soil bacterial micobiome of various plant species achieves a high accuracy of 92.3 % at the genus rank for drought stress prediction. It demonstrates its generalization capacity for the lineages tested.

CONCLUSIONS

In the detection of drought stress in soil bacterial microbiota, this study emphasizes the potential of an optimized and generalized location-based ML classifier. By identifying marker taxa, this approach holds promising implications for microbe-assisted plant breeding programs and contributes to the development of sustainable agriculture practices. These findings are crucial for preserving global food security in the face of climate change.

Meta-analysis of wastewater microbiome for antibiotic resistance profiling

The microbial composition and stress molecules are main drivers influencing the development and spread of antibiotic resistance bacteria (ARBs) and genes (ARGs) in the environment. A reliable and rapid method for identifying associations between microbiome composition and resistome remains challenging. In the present study, secondary metagenome data of sewage and hospital wastewaters were assessed for differential taxonomic and ARG profiling. Subsequently, Random Forest (RF)-based ML models were used to predict ARG profiles based on taxonomic composition and model validation on hospital wastewaters. Total ARG abundance was significantly higher in hospital wastewaters (15 ppm) than sewage (5 ppm), while the resistance towards methicillin, carbapenem, and fluoroquinolone were predominant. Although, Pseudomonas constituted major fraction, Streptomyces, Enterobacter, and Klebsiella were characteristic of hospital wastewaters. Prediction modeling showed that the relative abundance of pathogenic genera Escherichia, Vibrio, and Pseudomonas contributed most towards variations in total ARG count. Moreover, the model was able to identify host-specific patterns for contributing taxa and related ARGs with >90% accuracy in predicting the ARG subtype abundance. More than >80% accuracy was obtained for hospital wastewaters, demonstrating that the model can be validly extrapolated to different types of wastewater systems. Findings from the study showed that the ML approach could identify ARG profile based on bacterial composition including 16S rDNA amplicon data, and can serve as a viable alternative to metagenomic binning for identification of potential hosts of ARGs. Overall, this study demonstrates the promising application of ML techniques for predicting the spread of ARGs and provides guidance for early warning of ARBs emergence.

Interpretable metric learning in comparative metagenomics: The adaptive Haar-like distance

Random forests have emerged as a promising tool in comparative metagenomics because they can predict environmental characteristics based on microbial composition in datasets where β-diversity metrics fall short of revealing meaningful relationships between samples. Nevertheless, despite this efficacy, they lack biological insight in tandem with their predictions, potentially hindering scientific advancement. To overcome this limitation, we leverage a geometric characterization of random forests to introduce a data-driven phylogenetic β-diversity metric, the adaptive Haar-like distance. This new metric assigns a weight to each internal node (i.e., split or bifurcation) of a reference phylogeny, indicating the relative importance of that node in discerning environmental samples based on their microbial composition. Alongside this, a weighted nearest-neighbors classifier, constructed using the adaptive metric, can be used as a proxy for the random forest while maintaining accuracy on par with that of the original forest and another state-of-the-art classifier, CoDaCoRe. As shown in datasets from diverse microbial environments, however, the new metric and classifier significantly enhance the biological interpretability and visualization of high-dimensional metagenomic samples.

Proportion-based normalizations outperform compositional data transformations in machine learning applications

BACKGROUND

Normalization, as a pre-processing step, can significantly affect the resolution of machine learning analysis for microbiome studies. There are countless options for normalization scheme selection. In this study, we examined compositionally aware algorithms including the additive log ratio (alr), the centered log ratio (clr), and a recent evolution of the isometric log ratio (ilr) in the form of balance trees made with the PhILR R package. We also looked at compositionally naïve transformations such as raw counts tables and several transformations that are based on relative abundance, such as proportions, the Hellinger transformation, and a transformation based on the logarithm of proportions (which we call "lognorm").

RESULTS

In our evaluation, we used 65 metadata variables culled from four publicly available datasets at the amplicon sequence variant (ASV) level with a random forest machine learning algorithm. We found that different common pre-processing steps in the creation of the balance trees made very little difference in overall performance. Overall, we found that the compositionally aware data transformations such as alr, clr, and ilr (PhILR) performed generally slightly worse or only as well as compositionally naïve transformations. However, relative abundance-based transformations outperformed most other transformations by a small but reliably statistically significant margin.

CONCLUSIONS

Our results suggest that minimizing the complexity of transformations while correcting for read depth may be a generally preferable strategy in preparing data for machine learning compared to more sophisticated, but more complex, transformations that attempt to better correct for compositionality. Video Abstract.

Fast detection of bacterial gut pathogens on miniaturized devices: an overview

INTRODUCTION

Gut microbes pose challenges like colon inflammation, deadly diarrhea, antimicrobial resistance dissemination, and chronic disease onset. Development of early, rapid and specific diagnosis tools is essential for improving infection control. Point-of-care testing (POCT) systems offer rapid, sensitive, low-cost and sample-to-answer methods for microbe detection from various clinical and environmental samples, bringing the advantages of portability, automation, and simple operation.

AREAS COVERED

Rapid detection of gut microbes can be done using a wide array of techniques including biosensors, immunological assays, electrochemical impedance spectroscopy, mass spectrometry and molecular biology. Inclusion of Internet of Things, machine learning, and smartphone-based point-of-care applications is an important aspect of POCT. In this review, the authors discuss various fast diagnostic platforms for gut pathogens and their main challenges.

EXPERT OPINION

Developing effective assays for microbe detection can be complex. Assay design must consider factors like target selection, real-time and multiplex detection, sample type, reagent stability and storage, primer/probe design, and optimizing reaction conditions for accuracy and sensitivity. Mitigating these challenges requires interdisciplinary collaboration among scientists, clinicians, engineers, and industry partners. Future efforts are essential to enhance sensitivity, specificity, and versatility of POCT systems for gut microbe detection and quantification, advancing infectious disease diagnostics and management.

Pre-challenge gut microbial signature predicts RhCMV/SIV vaccine efficacy in rhesus macaques

Background

RhCMV/SIV vaccines protect ∼59% of vaccinated rhesus macaques against repeated limiting-dose intra-rectal exposure with highly pathogenic SIVmac239M, but the exact mechanism responsible for the vaccine efficacy is not known. It is becoming evident that complex interactions exist between gut microbiota and the host immune system. Here we aimed to investigate if the rhesus gut microbiome impacts RhCMV/SIV vaccine-induced protection.

Methods

Three groups of 15 rhesus macaques naturally pre-exposed to RhCMV were vaccinated with RhCMV/SIV vaccines. Rectal swabs were collected longitudinally both before SIV challenge (after vaccination) and post challenge and were profiled using 16S rRNA based microbiome analysis.

Results

We identified ∼2,400 16S rRNA amplicon sequence variants (ASVs), representing potential bacterial species/strains. Global gut microbial profiles were strongly associated with each of the three vaccination groups, and all animals tended to maintain consistent profiles throughout the pre-challenge phase. Despite vaccination group differences, using newly developed compositional data analysis techniques we identified a common gut microbial signature predictive of vaccine protection outcome across the three vaccination groups. Part of this microbial signature persisted even after SIV challenge. We also observed a strong correlation between this microbial signature and an early signature derived from whole blood transcriptomes in the same animals.

Conclusions

Our findings indicate that changes in gut microbiomes are associated with RhCMV/SIV vaccine-induced protection and early host response to vaccination in rhesus macaques.

Harnessing of Artificial Intelligence for the Diagnosis and Prevention of Hospital-Acquired Infections: A Systematic Review

Healthcare-associated infections (HAIs) are the most common adverse events in healthcare and constitute a major global public health concern. Surveillance represents the foundation for the effective prevention and control of HAIs, yet conventional surveillance is costly and labor intensive. Artificial intelligence (AI) and machine learning (ML) have the potential to support the development of HAI surveillance algorithms for the understanding of HAI risk factors, the improvement of patient risk stratification as well as the prediction and timely detection and prevention of infections. AI-supported systems have so far been explored for clinical laboratory testing and imaging diagnosis, antimicrobial resistance profiling, antibiotic discovery and prediction-based clinical decision support tools in terms of HAIs. This review aims to provide a comprehensive summary of the current literature on AI applications in the field of HAIs and discuss the future potentials of this emerging technology in infection practice. Following the PRISMA guidelines, this study examined the articles in databases including PubMed and Scopus until November 2023, which were screened based on the inclusion and exclusion criteria, resulting in 162 included articles. By elucidating the advancements in the field, we aim to highlight the potential applications of AI in the field, report related issues and shortcomings and discuss the future directions.

Nine (not so simple) steps: a practical guide to using machine learning in microbial ecology

Due to the complex nature of microbiome data, the field of microbial ecology has many current and potential uses for machine learning (ML) modeling. With the increased use of predictive ML models across many disciplines, including microbial ecology, there is extensive published information on the specific ML algorithms available and how those algorithms have been applied. Thus, our goal is not to summarize the breadth of ML models available or compare their performances. Rather, our goal is to provide more concrete and actionable information to guide microbial ecologists in how to select, run, and interpret ML algorithms to predict the taxa or genes associated with particular sample categories or environmental gradients of interest. Such microbial data often have unique characteristics that require careful consideration of how to apply ML models and how to interpret the associated results. This review is intended for practicing microbial ecologists who may be unfamiliar with some of the intricacies of ML models. We provide examples and discuss common opportunities and pitfalls specific to applying ML models to the types of data sets most frequently collected by microbial ecologists.

Consistent effects of independent domestication events on the plant microbiota

The effect of plant domestication on plant-microbe interactions remains difficult to prove. In this study, we provide evidence of a domestication effect on the composition and abundance of the plant microbiota. We focused on the genus Phaseolus, which underwent four independent domestication events within two species (P. vulgaris and P. lunatus), providing multiple replicates of a process spanning thousands of years. We targeted Phaseolus seeds to identify a link between domesticated traits and bacterial community composition as Phaseolus seeds have been subject to large and consistent phenotypic changes during these independent domestication events. The seed bacterial communities of representative plant accessions from subpopulations descended from each domestication event were analyzed under controlled and field conditions. The results showed that independent domestication events led to similar seed bacterial community signatures in independently domesticated plant populations, which could be partially explained by selection for common domesticated plant phenotypes. Our results therefore provide evidence of a consistent effect of plant domestication on seed microbial community composition and abundance and offer avenues for applying knowledge of the impact of plant domestication on the plant microbiota to improve microbial applications in agriculture.

Research progress on the application of 16S rRNA gene sequencing and machine learning in forensic microbiome individual identification

Forensic microbiome research is a field with a wide range of applications and a number of protocols have been developed for its use in this area of research. As individuals host radically different microbiota, the human microbiome is expected to become a new biomarker for forensic identification. To achieve an effective use of this procedure an understanding of factors which can alter the human microbiome and determinations of stable and changing elements will be critical in selecting appropriate targets for investigation. The 16S rRNA gene, which is notable for its conservation and specificity, represents a potentially ideal marker for forensic microbiome identification. Gene sequencing involving 16S rRNA is currently the method of choice for use in investigating microbiomes. While the sequencing involved with microbiome determinations can generate large multi-dimensional datasets that can be difficult to analyze and interpret, machine learning methods can be useful in surmounting this analytical challenge. In this review, we describe the research methods and related sequencing technologies currently available for application of 16S rRNA gene sequencing and machine learning in the field of forensic identification. In addition, we assess the potential value of 16S rRNA and machine learning in forensic microbiome science.

Methodology for biomarker discovery with reproducibility in microbiome data using machine learning

BACKGROUND

In recent years, human microbiome studies have received increasing attention as this field is considered a potential source for clinical applications. With the advancements in omics technologies and AI, research focused on the discovery for potential biomarkers in the human microbiome using machine learning tools has produced positive outcomes. Despite the promising results, several issues can still be found in these studies such as datasets with small number of samples, inconsistent results, lack of uniform processing and methodologies, and other additional factors lead to lack of reproducibility in biomedical research. In this work, we propose a methodology that combines the DADA2 pipeline for 16s rRNA sequences processing and the Recursive Ensemble Feature Selection (REFS) in multiple datasets to increase reproducibility and obtain robust and reliable results in biomedical research.

RESULTS

Three experiments were performed analyzing microbiome data from patients/cases in Inflammatory Bowel Disease (IBD), Autism Spectrum Disorder (ASD), and Type 2 Diabetes (T2D). In each experiment, we found a biomarker signature in one dataset and applied to 2 other as further validation. The effectiveness of the proposed methodology was compared with other feature selection methods such as K-Best with F-score and random selection as a base line. The Area Under the Curve (AUC) was employed as a measure of diagnostic accuracy and used as a metric for comparing the results of the proposed methodology with other feature selection methods. Additionally, we use the Matthews Correlation Coefficient (MCC) as a metric to evaluate the performance of the methodology as well as for comparison with other feature selection methods.

CONCLUSIONS

We developed a methodology for reproducible biomarker discovery for 16s rRNA microbiome sequence analysis, addressing the issues related with data dimensionality, inconsistent results and validation across independent datasets. The findings from the three experiments, across 9 different datasets, show that the proposed methodology achieved higher accuracy compared to other feature selection methods. This methodology is a first approach to increase reproducibility, to provide robust and reliable results.

Machine learning modeling for the prediction of phosphorus and nitrogen removal efficiency and screening of crucial microorganisms in wastewater treatment plants

The effectiveness of wastewater treatment plants (WWTPs) is largely determined by the microbial community structure in their activated sludge (AS). Interactions among microbial communities in AS systems and their indirect effects on water quality changes are crucial for WWTP performance. However, there is currently no quantitative method to evaluate the contribution of microorganisms to the operating efficiency of WWTPs. Traditional assessments of WWTP performance are limited by experimental conditions, methods, and other factors, resulting in increased costs and experimental pollutants. Therefore, an effective method is needed to predict WWTP efficiency based on AS community structure and quantitatively evaluate the contribution of microorganisms in the AS system. This study evaluated and compared microbial communities and water quality changes from WWTPs worldwide by meta-analysis of published high-throughput sequencing data. Six machine learning (ML) models were utilized to predict the efficiency of phosphorus and nitrogen removal in WWTPs; among them, XGBoost showed the highest prediction accuracy. Cross-entropy was used to screen the crucial microorganisms related to phosphorus and nitrogen removal efficiency, and the modeling confirmed the reasonableness of the results. Thirteen genera with nitrogen and phosphorus cycling pathways obtained from the screening were considered highly appropriate for the simultaneous removal of phosphorus and nitrogen. The results showed that the microbes Haliangium, Vicinamibacteraceae, Tolumonas, and SWB02 are potentially crucial for phosphorus and nitrogen removal, as they may be involved in the process of phosphorus and nitrogen removal in sewage treatment plants. Overall, these findings have deepened our understanding of the relationship between microbial community structure and performance of WWTPs, indicating that microbial data should play a critical role in the future design of sewage treatment plants. The ML model of this study can efficiently screen crucial microbes associated with WWTP system performance, and it is promising for the discovery of potential microbial metabolic pathways.

A robust microbiome signature for autism spectrum disorder across different studies using machine learning

Autism spectrum disorder (ASD) is a highly complex neurodevelopmental disorder characterized by deficits in sociability and repetitive behaviour, however there is a great heterogeneity within other comorbidities that accompany ASD. Recently, gut microbiome has been pointed out as a plausible contributing factor for ASD development as individuals diagnosed with ASD often suffer from intestinal problems and show a differentiated intestinal microbial composition. Nevertheless, gut microbiome studies in ASD rarely agree on the specific bacterial taxa involved in this disorder. Regarding the potential role of gut microbiome in ASD pathophysiology, our aim is to investigate whether there is a set of bacterial taxa relevant for ASD classification by using a sibling-controlled dataset. Additionally, we aim to validate these results across two independent cohorts as several confounding factors, such as lifestyle, influence both ASD and gut microbiome studies. A machine learning approach, recursive ensemble feature selection (REFS), was applied to 16S rRNA gene sequencing data from 117 subjects (60 ASD cases and 57 siblings) identifying 26 bacterial taxa that discriminate ASD cases from controls. The average area under the curve (AUC) of this specific set of bacteria in the sibling-controlled dataset was 81.6%. Moreover, we applied the selected bacterial taxa in a tenfold cross-validation scheme using two independent cohorts (a total of 223 samples-125 ASD cases and 98 controls). We obtained average AUCs of 74.8% and 74%, respectively. Analysis of the gut microbiome using REFS identified a set of bacterial taxa that can be used to predict the ASD status of children in three distinct cohorts with AUC over 80% for the best-performing classifiers. Our results indicate that the gut microbiome has a strong association with ASD and should not be disregarded as a potential target for therapeutic interventions. Furthermore, our work can contribute to use the proposed approach for identifying microbiome signatures across other 16S rRNA gene sequencing datasets.

Methodological Considerations in Longitudinal Analyses of Microbiome Data: A Comprehensive Review

Biological processes underlying health and disease are inherently dynamic and are best understood when characterized in a time-informed manner. In this comprehensive review, we discuss challenges inherent in time-series microbiome data analyses and compare available approaches and methods to overcome them. Appropriate handling of longitudinal microbiome data can shed light on important roles, functions, patterns, and potential interactions between large numbers of microbial taxa or genes in the context of health, disease, or interventions. We present a comprehensive review and comparison of existing microbiome time-series analysis methods, for both preprocessing and downstream analyses, including differential analysis, clustering, network inference, and trait classification. We posit that the careful selection and appropriate utilization of computational tools for longitudinal microbiome analyses can help advance our understanding of the dynamic host-microbiome relationships that underlie health-maintaining homeostases, progressions to disease-promoting dysbioses, as well as phases of physiologic development like those encountered in childhood.

Machine-learning analysis of cross-study samples according to the gut microbiome in 12 infant cohorts

IMPORTANCE

There are challenges in merging microbiome data from diverse research groups due to the intricate and multifaceted nature of such data. To address this, we utilized a combination of machine-learning (ML) models to analyze 16S sequencing data from a substantial set of gut microbiome samples, sourced from 12 distinct infant cohorts that were gathered prospectively. Our initial focus was on the mode of delivery due to its prior association with changes in infant gut microbiomes. Through ML analysis, we demonstrated the effective merging and comparison of various gut microbiome data sets, facilitating the identification of robust microbiome biomarkers applicable across varied study populations.

Microbiome Metabolome Integration Platform (MMIP): a web-based platform for microbiome and metabolome data integration and feature identification

A microbial community maintains its ecological dynamics via metabolite crosstalk. Hence, knowledge of the metabolome, alongside its populace, would help us understand the functionality of a community and also predict how it will change in atypical conditions. Methods that employ low-cost metagenomic sequencing data can predict the metabolic potential of a community, that is, its ability to produce or utilize specific metabolites. These, in turn, can potentially serve as markers of biochemical pathways that are associated with different communities. We developed MMIP (Microbiome Metabolome Integration Platform), a web-based analytical and predictive tool that can be used to compare the taxonomic content, diversity variation and the metabolic potential between two sets of microbial communities from targeted amplicon sequencing data. MMIP is capable of highlighting statistically significant taxonomic, enzymatic and metabolic attributes as well as learning-based features associated with one group in comparison with another. Furthermore, MMIP can predict linkages among species or groups of microbes in the community, specific enzyme profiles, compounds or metabolites associated with such a group of organisms. With MMIP, we aim to provide a user-friendly, online web server for performing key microbiome-associated analyses of targeted amplicon sequencing data, predicting metabolite signature, and using learning-based linkage analysis, without the need for initial metabolomic analysis, and thereby helping in hypothesis generation.

Supervised learning and model analysis with compositional data

Supervised learning, such as regression and classification, is an essential tool for analyzing modern high-throughput sequencing data, for example in microbiome research. However, due to the compositionality and sparsity, existing techniques are often inadequate. Either they rely on extensions of the linear log-contrast model (which adjust for compositionality but cannot account for complex signals or sparsity) or they are based on black-box machine learning methods (which may capture useful signals, but lack interpretability due to the compositionality). We propose KernelBiome, a kernel-based nonparametric regression and classification framework for compositional data. It is tailored to sparse compositional data and is able to incorporate prior knowledge, such as phylogenetic structure. KernelBiome captures complex signals, including in the zero-structure, while automatically adapting model complexity. We demonstrate on par or improved predictive performance compared with state-of-the-art machine learning methods on 33 publicly available microbiome datasets. Additionally, our framework provides two key advantages: (i) We propose two novel quantities to interpret contributions of individual components and prove that they consistently estimate average perturbation effects of the conditional mean, extending the interpretability of linear log-contrast coefficients to nonparametric models. (ii) We show that the connection between kernels and distances aids interpretability and provides a data-driven embedding that can augment further analysis. KernelBiome is available as an open-source Python package on PyPI and at https://github.com/shimenghuang/KernelBiome.

Dysbiosis signatures of gut microbiota and the progression of type 2 diabetes: a machine learning approach in a Mexican cohort

Introduction

The gut microbiota (GM) dysbiosis is one of the causal factors for the progression of different chronic metabolic diseases, including type 2 diabetes mellitus (T2D). Understanding the basis that laid this association may lead to developing new therapeutic strategies for preventing and treating T2D, such as probiotics, prebiotics, and fecal microbiota transplants. It may also help identify potential early detection biomarkers and develop personalized interventions based on an individual's gut microbiota profile. Here, we explore how supervised Machine Learning (ML) methods help to distinguish taxa for individuals with prediabetes (prediabetes) or T2D.

Methods

To this aim, we analyzed the GM profile (16s rRNA gene sequencing) in a cohort of 410 Mexican naïve patients stratified into normoglycemic, prediabetes, and T2D individuals. Then, we compared six different ML algorithms and found that Random Forest had the highest predictive performance in classifying T2D and prediabetes patients versus controls.

Results

We identified a set of taxa for predicting patients with T2D compared to normoglycemic individuals, including Allisonella, Slackia, Ruminococus_2, Megaspgaera, Escherichia/Shigella, and Prevotella, among them. Besides, we concluded that Anaerostipes, Intestinibacter, Prevotella_9, Blautia, Granulicatella, and Veillonella were the relevant genus in patients with prediabetes compared to normoglycemic subjects.

Discussion

These findings allow us to postulate that GM is a distinctive signature in prediabetes and T2D patients during the development and progression of the disease. Our study highlights the role of GM and opens a window toward the rational design of new preventive and personalized strategies against the control of this disease.

MetaPhlAn 4 profiling of unknown species-level genome bins improves the characterization of diet-associated microbiome changes in mice

Mouse models are key tools for investigating host-microbiome interactions. However, shotgun metagenomics can only profile a limited fraction of the mouse gut microbiome. Here, we employ a metagenomic profiling method, MetaPhlAn 4, which exploits a large catalog of metagenome-assembled genomes (including 22,718 metagenome-assembled genomes from mice) to improve the profiling of the mouse gut microbiome. We combine 622 samples from eight public datasets and an additional cohort of 97 mouse microbiomes, and we assess the potential of MetaPhlAn 4 to better identify diet-related changes in the host microbiome using a meta-analysis approach. We find multiple, strong, and reproducible diet-related microbial biomarkers, largely increasing those identifiable by other available methods relying only on reference information. The strongest drivers of the diet-induced changes are uncharacterized and previously undetected taxa, confirming the importance of adopting metagenomic methods integrating metagenomic assemblies for comprehensive profiling.

Rumen Microbiota Predicts Feed Efficiency of Primiparous Nordic Red Dairy Cows

Efficient feed utilization in dairy cows is crucial for economic and environmental reasons. The rumen microbiota plays a significant role in feed efficiency, but studies utilizing microbial data to predict host phenotype are limited. In this study, 87 primiparous Nordic Red dairy cows were ranked for feed efficiency during their early lactation based on residual energy intake, and the rumen liquid microbial ecosystem was subsequently evaluated using 16S rRNA amplicon and metagenome sequencing. The study used amplicon data to build an extreme gradient boosting model, demonstrating that taxonomic microbial variation can predict efficiency (r_test = 0.55). Prediction interpreters and microbial network revealed that predictions were based on microbial consortia and the efficient animals had more of the highly interacting microbes and consortia. Rumen metagenome data was used to evaluate carbohydrate-active enzymes and metabolic pathway differences between efficiency phenotypes. The study showed that an efficient rumen had a higher abundance of glycoside hydrolases, while an inefficient rumen had more glycosyl transferases. Enrichment of metabolic pathways was observed in the inefficient group, while efficient animals emphasized bacterial environmental sensing and motility over microbial growth. The results suggest that inter-kingdom interactions should be further analyzed to understand their association with the feed efficiency of animals.

New perspectives into the vaginal microbiome with systems biology

The vaginal microbiome (VMB) is critical to female reproductive health; however, the mechanisms associated with optimal and non-optimal states remain poorly understood due to the complex community structure and dynamic nature. Quantitative systems biology techniques applied to the VMB have improved understanding of community composition and function using primarily statistical methods. In contrast, fewer mechanistic models that use a priori knowledge of VMB features to develop predictive models have been implemented despite their use for microbiomes at other sites, including the gastrointestinal tract. Here, we explore systems biology approaches that have been applied in the VMB, highlighting successful techniques and discussing new directions that hold promise for improving understanding of health and disease.

Conditional Forest Models Built Using Metagenomic Data Accurately Predicted Salmonella Contamination in Northeastern Streams

The use of water contaminated with Salmonella for produce production contributes to foodborne disease burden. To reduce human health risks, there is a need for novel, targeted approaches for assessing the pathogen status of agricultural water. We investigated the utility of water microbiome data for predicting Salmonella contamination of streams used to source water for produce production. Grab samples were collected from 60 New York streams in 2018 and tested for Salmonella. Separately, DNA was extracted from the samples and used for Illumina shotgun metagenomic sequencing. Reads were trimmed and used to assign taxonomy with Kraken2. Conditional forest (CF), regularized random forest (RRF), and support vector machine (SVM) models were implemented to predict Salmonella contamination. Model performance was assessed using 10-fold cross-validation repeated 10 times to quantify area under the curve (AUC) and Kappa score. CF models outperformed the other two algorithms based on AUC (0.86, CF; 0.81, RRF; 0.65, SVM) and Kappa score (0.53, CF; 0.41, RRF; 0.12, SVM). The taxa that were most informative for accurately predicting Salmonella contamination based on CF were compared to taxa identified by ALDEx2 as being differentially abundant between Salmonella-positive and -negative samples. CF and differential abundance tests both identified Aeromonas salmonicida (variable importance [VI] = 0.012) and Aeromonas sp. strain CA23 (VI = 0.025) as the two most informative taxa for predicting Salmonella contamination. Our findings suggest that microbiome-based models may provide an alternative to or complement existing water monitoring strategies. Similarly, the informative taxa identified in this study warrant further investigation as potential indicators of Salmonella contamination of agricultural water. IMPORTANCE Understanding the associations between surface water microbiome composition and the presence of foodborne pathogens, such as Salmonella, can facilitate the identification of novel indicators of Salmonella contamination. This study assessed the utility of microbiome data and three machine learning algorithms for predicting Salmonella contamination of Northeastern streams. The research reported here both expanded the knowledge on the microbiome composition of surface waters and identified putative novel indicators (i.e., Aeromonas species) for Salmonella in Northeastern streams. These putative indicators warrant further research to assess whether they are consistent indicators of Salmonella contamination across regions, waterways, and years not represented in the data set used in this study. Validated indicators identified using microbiome data may be used as targets in the development of rapid (e.g., PCR-based) detection assays for the assessment of microbial safety of agricultural surface waters.

EnsDeepDP: An Ensemble Deep Learning Approach for Disease Prediction Through Metagenomics

A growing number of studies show that the human microbiome plays a vital role in human health and can be a crucial factor in predicting certain human diseases. However, microbiome data are often characterized by the limited samples and high-dimensional features, which pose a great challenge for machine learning methods. Therefore, this paper proposes a novel ensemble deep learning disease prediction method that combines unsupervised and supervised learning paradigms. First, unsupervised deep learning methods are used to learn the potential representation of the sample. Afterwards, the disease scoring strategy is developed based on the deep representations as the informative features for ensemble analysis. To ensure the optimal ensemble, a score selection mechanism is constructed, and performance boosting features are engaged with the original sample. Finally, the composite features are trained with gradient boosting classifier for health status decision. For case study, the ensemble deep learning flowchart has been demonstrated on six public datasets extracted from the human microbiome profiling. The results show that compared with the existing algorithms, our framework achieves better performance on disease prediction.

Leveraging Scheme for Cross-Study Microbiome Machine Learning Prediction and Feature Evaluations

The microbiota has proved to be one of the critical factors for many diseases, and researchers have been using microbiome data for disease prediction. However, models trained on one independent microbiome study may not be easily applicable to other independent studies due to the high level of variability in microbiome data. In this study, we developed a method for improving the generalizability and interpretability of machine learning models for predicting three different diseases (colorectal cancer, Crohn's disease, and immunotherapy response) using nine independent microbiome datasets. Our method involves combining a smaller dataset with a larger dataset, and we found that using at least 25% of the target samples in the source data resulted in improved model performance. We determined random forest as our top model and employed feature selection to identify common and important taxa for disease prediction across the different studies. Our results suggest that this leveraging scheme is a promising approach for improving the accuracy and interpretability of machine learning models for predicting diseases based on microbiome data.

Compositionality, sparsity, spurious heterogeneity, and other data-driven challenges for machine learning algorithms within plant microbiome studies

The plant-associated microbiome is a key component of plant systems, contributing to their health, growth, and productivity. The application of machine learning (ML) in this field promises to help untangle the relationships involved. However, measurements of microbial communities by high-throughput sequencing pose challenges for ML. Noise from low sample sizes, soil heterogeneity, and technical factors can impact the performance of ML. Additionally, the compositional and sparse nature of these datasets can impact the predictive accuracy of ML. We review recent literature from plant studies to illustrate that these properties often go unmentioned. We expand our analysis to other fields to quantify the degree to which mitigation approaches improve the performance of ML and describe the mathematical basis for this. With the advent of accessible analytical packages for microbiome data including learning models, researchers must be familiar with the nature of their datasets.

Interpreting tree ensemble machine learning models with endoR

Tree ensemble machine learning models are increasingly used in microbiome science as they are compatible with the compositional, high-dimensional, and sparse structure of sequence-based microbiome data. While such models are often good at predicting phenotypes based on microbiome data, they only yield limited insights into how microbial taxa may be associated. We developed endoR, a method to interpret tree ensemble models. First, endoR simplifies the fitted model into a decision ensemble. Then, it extracts information on the importance of individual features and their pairwise interactions, displaying them as an interpretable network. Both the endoR network and importance scores provide insights into how features, and interactions between them, contribute to the predictive performance of the fitted model. Adjustable regularization and bootstrapping help reduce the complexity and ensure that only essential parts of the model are retained. We assessed endoR on both simulated and real metagenomic data. We found endoR to have comparable accuracy to other common approaches while easing and enhancing model interpretation. Using endoR, we also confirmed published results on gut microbiome differences between cirrhotic and healthy individuals. Finally, we utilized endoR to explore associations between human gut methanogens and microbiome components. Indeed, these hydrogen consumers are expected to interact with fermenting bacteria in a complex syntrophic network. Specifically, we analyzed a global metagenome dataset of 2203 individuals and confirmed the previously reported association between Methanobacteriaceae and Christensenellales. Additionally, we observed that Methanobacteriaceae are associated with a network of hydrogen-producing bacteria. Our method accurately captures how tree ensembles use features and interactions between them to predict a response. As demonstrated by our applications, the resultant visualizations and summary outputs facilitate model interpretation and enable the generation of novel hypotheses about complex systems.

Sampling from four geographically divergent young female populations demonstrates forensic geolocation potential in microbiomes

Studies of human microbiomes using new sequencing techniques have increasingly demonstrated that their ecologies are partly determined by the lifestyle and habits of individuals. As such, significant forensic information could be obtained from high throughput sequencing of the human microbiome. This approach, combined with multiple analytical techniques demonstrates that bacterial DNA can be used to uniquely identify an individual and to provide information about their life and behavioral patterns. However, the transformation of these findings into actionable forensic information, including the geolocation of the samples, remains limited by incomplete understanding of the effects of confounding factors and the paucity of diverse sequences. We obtained 16S rRNA sequences of stool and oral microbiomes collected from 206 young and healthy females from four globally diverse populations, in addition to supporting metadata, including dietary and medical information. Analysis of these microbiomes revealed detectable geolocation signals between the populations, even for populations living within the same city. Accounting for other lifestyle variables, such as diet and smoking, lessened but does not remove the geolocation signal.

Microbiome in Precision Psychiatry: An Overview of the Ethical Challenges Regarding Microbiome Big Data and Microbiome-Based Interventions

There has been a spurt in both fundamental and translational research that examines the underlying mechanisms of the human microbiome in psychiatric disorders. The personalized and dynamic features of the human microbiome suggest the potential of its manipulation for precision psychiatry in ways to improve mental health and avoid disease. However, findings in the field of microbiome also raise philosophical and ethical questions. From a philosophical point of view, they may yet be another attempt at providing a biological cause for phenomena that ultimately cannot be so easily localized. From an ethical point of view, it is relevant that the human gut microbiome comprises data on the individual's lifestyle, disease history, previous medications, and mental health. Massive datasets of microbiome sequences are collected to facilitate comparative studies to identify specific links between the microbiome and mental health. Although this emerging research domain may show promise for psychiatric patients, it is surrounded by ethical challenges regarding patient privacy, health risks, effects on personal identity, and concerns about responsibility. This narrative overview displays the roles and advances of microbiome research in psychiatry and discusses the philosophical and ethical implications of microbiome big data and microbiome-based interventions for psychiatric patients. We also investigate whether these issues are really "new," or "old wine in new bottles."

Machine Learning Data Analysis Highlights the Role of Parasutterella and Alloprevotella in Autism Spectrum Disorders

In recent years, the involvement of the gut microbiota in disease and health has been investigated by sequencing the 16S gene from fecal samples. Dysbiotic gut microbiota was also observed in Autism Spectrum Disorder (ASD), a neurodevelopmental disorder characterized by gastrointestinal symptoms. However, despite the relevant number of studies, it is still difficult to identify a typical dysbiotic profile in ASD patients. The discrepancies among these studies are due to technical factors (i.e., experimental procedures) and external parameters (i.e., dietary habits). In this paper, we collected 959 samples from eight available projects (540 ASD and 419 Healthy Controls, HC) and reduced the observed bias among studies. Then, we applied a Machine Learning (ML) approach to create a predictor able to discriminate between ASD and HC. We tested and optimized three algorithms: Random Forest, Support Vector Machine and Gradient Boosting Machine. All three algorithms confirmed the importance of five different genera, including Parasutterella and Alloprevotella. Furthermore, our results show that ML algorithms could identify common taxonomic features by comparing datasets obtained from countries characterized by latent confounding variables.

Evaluation of Prebiotics through an In Vitro Gastrointestinal Digestion and Fecal Fermentation Experiment: Further Idea on the Implementation of Machine Learning Technique

Prebiotics are non-digestible food ingredients that promote the growth of beneficial gut microorganisms and foster their activities. The performance of prebiotics has often been tested in mouse models in which the gut ecology differs from that of humans. In this study, we instead performed an in vitro gastrointestinal digestion and fecal fermentation experiment to evaluate the efficiency of eight different prebiotics. Feces obtained from 11 different individuals were used to ferment digested prebiotics. The total DNA from each sample was extracted and sequenced through Illumina MiSeq for microbial community analysis. The amount of short-chain fatty acids was assessed through gas chromatography. We found links between community shifts and the increased amount of short-chain fatty acids after prebiotics treatment. The results from differential abundance analysis showed increases in beneficial gut microorganisms, such as Bifidobacterium, Faeclibacterium, and Agathobacter, after prebiotics treatment. We were also able to construct well-performing machine-learning models that could predict the amount of short-chain fatty acids based on the gut microbial community structure. Finally, we provide an idea for further implementation of machine-learning techniques to find customized prebiotics.

Improve the Colorectal Cancer Diagnosis Using Gut Microbiome Data

In the United States, colorectal cancer is the second largest cause of cancer death, and accurate early detection and identification of high-risk patients is a high priority. Although fecal screening tests are available, the close relationship between colorectal cancer and the gut microbiome has generated considerable interest. We describe a machine learning method for gut microbiome data to assist in diagnosing colorectal cancer. Our methodology integrates feature engineering, mediation analysis, statistical modeling, and network analysis into a novel unified pipeline. Simulation results illustrate the value of the method in comparison to existing methods. For predicting colorectal cancer in two real datasets, this pipeline showed an 8.7% higher prediction accuracy and 13% higher area under the receiver operator characteristic curve than other published work. Additionally, the approach highlights important colorectal cancer-related taxa for prioritization, such as high levels of Bacteroides fragilis, which can help elucidate disease pathology. Our algorithms and approach can be widely applied for Colorectal cancer prediction using either 16 S rRNA or shotgun metagenomics data.

Predicting the abundance of metal resistance genes in subtropical estuaries using amplicon sequencing and machine learning

Heavy metals are a group of anthropogenic contaminants in estuary ecosystems. Bacteria in estuaries counteract the highly concentrated metal toxicity through metal resistance genes (MRGs). Presently, metagenomic technology is popularly used to study MRGs. However, an easier and less expensive method of acquiring MRG information is needed to deepen our understanding of the fate of MRGs. Thus, this study explores the feasibility of using a machine learning approach-namely, random forests (RF)-to predict MRG abundance based on the 16S rRNA amplicon sequenced datasets from subtropical estuaries in China. Our results showed that the total MRG abundance could be predicted by RF models using bacterial composition at different taxonomic levels. Among them, the relative abundance of bacterial phyla had the highest predicted accuracy (71.7 %). In addition, the RF models constructed by bacterial phyla predicted the abundance of six MRG types and nine MRG subtypes with substantial accuracy (R² > 0.600). Five bacterial phyla (Firmicutes, Bacteroidetes, Patescibacteria, Armatimonadetes, and Nitrospirae) substantially determined the variations in MRG abundance. Our findings prove that RF models can predict MRG abundance in South China estuaries during the wet season by using the bacterial composition obtained by 16S rRNA amplicon sequencing.

New-Generation Sequencing Technology in Diagnosis of Fungal Plant Pathogens: A Dream Comes True?

The fast and continued progress of high-throughput sequencing (HTS) and the drastic reduction of its costs have boosted new and unpredictable developments in the field of plant pathology. The cost of whole-genome sequencing, which, until few years ago, was prohibitive for many projects, is now so affordable that a new branch, phylogenomics, is being developed. Fungal taxonomy is being deeply influenced by genome comparison, too. It is now easier to discover new genes as potential targets for an accurate diagnosis of new or emerging pathogens, notably those of quarantine concern. Similarly, with the development of metabarcoding and metagenomics techniques, it is now possible to unravel complex diseases or answer crucial questions, such as "What's in my soil?", to a good approximation, including fungi, bacteria, nematodes, etc. The new technologies allow to redraw the approach for disease control strategies considering the pathogens within their environment and deciphering the complex interactions between microorganisms and the cultivated crops. This kind of analysis usually generates big data that need sophisticated bioinformatic tools (machine learning, artificial intelligence) for their management. Herein, examples of the use of new technologies for research in fungal diversity and diagnosis of some fungal pathogens are reported.

Metagenomics and artificial intelligence in the context of human health

Human microbiome is ubiquitous, dynamic, and site-specific consortia of microbial communities. The pathogenic nature of microorganisms within human tissues has led to an increase in microbial studies. Characterization of genera, like Streptococcus, Cutibacterium, Staphylococcus, Bifidobacterium, Lactococcus and Lactobacillus through culture-dependent and culture-independent techniques has been reported. However, due to the unique environment within human tissues, it is difficult to culture these microorganisms making their molecular studies strenuous. MGs offer a gateway to explore and characterize hidden microbial communities through a culture-independent mode by direct DNA isolation. By function and sequence-based MGs, Scientists can explore the mechanistic details of numerous microbes and their interaction with the niche. Since the data generated from MGs studies is highly complex and multi-dimensional, it requires accurate analytical tools to evaluate and interpret the data. Artificial intelligence (AI) provides the luxury to automatically learn the data dimensionality and ease its complexity that makes the disease diagnosis and disease response easy, accurate and timely. This review provides insight into the human microbiota and its exploration and expansion through MG studies. The review elucidates the significance of MGs in studying the changing microbiota during disease conditions besides highlighting the role of AI in computational analysis of MG data.

Multi-Omics Strategies for Investigating the Microbiome in Toxicology Research

Microbial communities on and within the host contact environmental pollutants, toxic compounds, and other xenobiotic compounds. These communities of bacteria, fungi, viruses, and archaea possess diverse metabolic potential to catabolize compounds and produce new metabolites. Microbes alter chemical disposition thus making the microbiome a natural subject of interest for toxicology. Sequencing and metabolomics technologies permit the study of microbiomes altered by acute or long-term exposure to xenobiotics. These investigations have already contributed to and are helping to re-interpret traditional understandings of toxicology. The purpose of this review is to provide a survey of the current methods used to characterize microbes within the context of toxicology. This will include discussion of commonly used techniques for conducting omic-based experiments, their respective strengths and deficiencies, and how forward-looking techniques may address present shortcomings. Finally, a perspective will be provided regarding common assumptions that currently impede microbiome studies from producing causal explanations of toxicologic mechanisms.

Human disease prediction from microbiome data by multiple feature fusion and deep learning

Human disease prediction from microbiome data has broad implications in metagenomics. It is rare for the existing methods to consider abundance profiles from both known and unknown microbial organisms, or capture the taxonomic relationships among microbial taxa, leading to significant information loss. On the other hand, deep learning has shown unprecedented advantages in classification tasks for its feature-learning ability. However, it encounters the opposite situation in metagenome-based disease prediction since high-dimensional low-sample-size metagenomic datasets can lead to severe overfitting; and black-box model fails in providing biological explanations. To circumvent the related problems, we developed MetaDR, a comprehensive machine learning-based framework that integrates various information and deep learning to predict human diseases. Experimental results indicate that MetaDR achieves competitive prediction performance with a reduction in running time, and effectively discovers the informative features with biological insights.

LinDA: linear models for differential abundance analysis of microbiome compositional data

Differential abundance analysis is at the core of statistical analysis of microbiome data. The compositional nature of microbiome sequencing data makes false positive control challenging. Here, we show that the compositional effects can be addressed by a simple, yet highly flexible and scalable, approach. The proposed method, LinDA, only requires fitting linear regression models on the centered log-ratio transformed data, and correcting the bias due to compositional effects. We show that LinDA enjoys asymptotic FDR control and can be extended to mixed-effect models for correlated microbiome data. Using simulations and real examples, we demonstrate the effectiveness of LinDA.

EPIMUTESTR: a nearest neighbor machine learning approach to predict cancer driver genes from the evolutionary action of coding variants

Discovering rare cancer driver genes is difficult because their mutational frequency is too low for statistical detection by computational methods. EPIMUTESTR is an integrative nearest-neighbor machine learning algorithm that identifies such marginal genes by modeling the fitness of their mutations with the phylogenetic Evolutionary Action (EA) score. Over cohorts of sequenced patients from The Cancer Genome Atlas representing 33 tumor types, EPIMUTESTR detected 214 previously inferred cancer driver genes and 137 new candidates never identified computationally before of which seven genes are supported in the COSMIC Cancer Gene Census. EPIMUTESTR achieved better robustness and specificity than existing methods in a number of benchmark methods and datasets.

Host phenotype classification from human microbiome data is mainly driven by the presence of microbial taxa

Machine learning-based classification approaches are widely used to predict host phenotypes from microbiome data. Classifiers are typically employed by considering operational taxonomic units or relative abundance profiles as input features. Such types of data are intrinsically sparse, which opens the opportunity to make predictions from the presence/absence rather than the relative abundance of microbial taxa. This also poses the question whether it is the presence rather than the abundance of particular taxa to be relevant for discrimination purposes, an aspect that has been so far overlooked in the literature. In this paper, we aim at filling this gap by performing a meta-analysis on 4,128 publicly available metagenomes associated with multiple case-control studies. At species-level taxonomic resolution, we show that it is the presence rather than the relative abundance of specific microbial taxa to be important when building classification models. Such findings are robust to the choice of the classifier and confirmed by statistical tests applied to identifying differentially abundant/present taxa. Results are further confirmed at coarser taxonomic resolutions and validated on 4,026 additional 16S rRNA samples coming from 30 public case-control studies.

The composition of the human microbiome has been linked to a large number of different diseases. In this context, classification methodologies based on machine learning approaches have represented a promising tool for diagnostic purposes from metagenomics data. The link between microbial population composition and host phenotypes has been usually performed by considering taxonomic profiles represented by relative abundances of microbial species. In this study, we show that it is more the presence rather than the relative abundance of microbial taxa to be relevant to maximize classification accuracy. This is accomplished by conducting a meta-analysis on more than 4,000 shotgun metagenomes coming from 25 case-control studies and in which original relative abundance data are degraded to presence/absence profiles. Findings are also extended to 16S rRNA data and advance the research field in building prediction models directly from human microbiome data.

Machine learning-assisted identification of bioindicators predicts medium-chain carboxylate production performance of an anaerobic mixed culture

BACKGROUND

The ability to quantitatively predict ecophysiological functions of microbial communities provides an important step to engineer microbiota for desired functions related to specific biochemical conversions. Here, we present the quantitative prediction of medium-chain carboxylate production in two continuous anaerobic bioreactors from 16S rRNA gene dynamics in enriched communities.

RESULTS

By progressively shortening the hydraulic retention time (HRT) from 8 to 2 days with different temporal schemes in two bioreactors operated for 211 days, we achieved higher productivities and yields of the target products n-caproate and n-caprylate. The datasets generated from each bioreactor were applied independently for training and testing machine learning algorithms using 16S rRNA genes to predict n-caproate and n-caprylate productivities. Our dataset consisted of 14 and 40 samples from HRT of 8 and 2 days, respectively. Because of the size and balance of our dataset, we compared linear regression, support vector machine and random forest regression algorithms using the original and balanced datasets generated using synthetic minority oversampling. Further, we performed cross-validation to estimate model stability. The random forest regression was the best algorithm producing more consistent results with median of error rates below 8%. More than 90% accuracy in the prediction of n-caproate and n-caprylate productivities was achieved. Four inferred bioindicators belonging to the genera Olsenella, Lactobacillus, Syntrophococcus and Clostridium IV suggest their relevance to the higher carboxylate productivity at shorter HRT. The recovery of metagenome-assembled genomes of these bioindicators confirmed their genetic potential to perform key steps of medium-chain carboxylate production.

CONCLUSIONS

Shortening the hydraulic retention time of the continuous bioreactor systems allows to shape the communities with desired chain elongation functions. Using machine learning, we demonstrated that 16S rRNA amplicon sequencing data can be used to predict bioreactor process performance quantitatively and accurately. Characterizing and harnessing bioindicators holds promise to manage reactor microbiota towards selection of the target processes. Our mathematical framework is transferrable to other ecosystem processes and microbial systems where community dynamics is linked to key functions. The general methodology used here can be adapted to data types of other functional categories such as genes, transcripts, proteins or metabolites. Video Abstract.

Revealing General Patterns of Microbiomes That Transcend Systems: Potential and Challenges of Deep Transfer Learning

A growing body of research has established that the microbiome can mediate the dynamics and functional capacities of diverse biological systems. Yet, we understand little about what governs the response of these microbial communities to host or environmental changes. Most efforts to model microbiomes focus on defining the relationships between the microbiome, host, and environmental features within a specified study system and therefore fail to capture those that may be evident across multiple systems. In parallel with these developments in microbiome research, computer scientists have developed a variety of machine learning tools that can identify subtle, but informative, patterns from complex data. Here, we recommend using deep transfer learning to resolve microbiome patterns that transcend study systems. By leveraging diverse public data sets in an unsupervised way, such models can learn contextual relationships between features and build on those patterns to perform subsequent tasks (e.g., classification) within specific biological contexts.

Mian: interactive web-based microbiome data table visualization and machine learning platform

SUMMARY

Mian is a web application to interactively visualize, run statistical tools and train machine learning models on operational taxonomic unit (OTU) or amplicon sequence variant (ASV) datasets to identify key taxonomic groups, diversity trends or taxonomic composition shifts in the context of provided categorical or numerical sample metadata. Tools, including Fisher's exact test, Boruta feature selection, alpha and beta diversity, and random forest and deep neural network classifiers, facilitate open-ended data exploration and hypothesis generation on microbial datasets.

AVAILABILITY

Mian is freely available at: miandata.org. Mian is an open-source platform licensed under the MIT license with source code available at github.com/tbj128/mian.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Incorporating genome-based phylogeny and functional similarity into diversity assessments helps to resolve a global collection of human gut metagenomes

Tree-based diversity measures incorporate phylogenetic or functional relatedness into comparisons of microbial communities. This can improve the identification of explanatory factors compared to tree-agnostic diversity measures. However, applying tree-based diversity measures to metagenome data is more challenging than for single-locus sequencing (e.g., 16S rRNA gene). Utilizing the Genome Taxonomy Database (GTDB) for species-level metagenome profiling allows for functional diversity measures based on genomic content or traits inferred from it. Still, it is unclear how metagenome-based assessments of microbiome diversity benefit from incorporating phylogeny or function into measures of diversity. We assessed this by measuring phylogeny-based, function-based, and tree-agnostic diversity measures from a large, global collection of human gut metagenomes composed of 30 studies and 2943 samples. We found tree-based measures to explain phenotypic variation (e.g., westernization, disease status, and gender) better or equivalent to tree-agnostic measures. Ecophylogenetic and functional diversity measures provided unique insight into how microbiome diversity was partitioned by phenotype. Tree-based measures greatly improved machine learning model performance for predicting westernization, disease status, and gender, relative to models trained solely on tree-agnostic measures. Our findings illustrate the usefulness of tree- and function-based measures for metagenomic assessments of microbial diversity, which is a fundamental component of microbiome science. This article is protected by copyright. All rights reserved.

Multimodal deep learning applied to classify healthy and disease states of human microbiome

Metagenomic sequencing methods provide considerable genomic information regarding human microbiomes, enabling us to discover and understand microbial diseases. Compositional differences have been reported between patients and healthy people, which could be used in the diagnosis of patients. Despite significant progress in this regard, the accuracy of these tools needs to be improved for applications in diagnostics and therapeutics. MDL4Microbiome, the method developed herein, demonstrated high accuracy in predicting disease status by using various features from metagenome sequences and a multimodal deep learning model. We propose combining three different features, i.e., conventional taxonomic profiles, genome-level relative abundance, and metabolic functional characteristics, to enhance classification accuracy. This deep learning model enabled the construction of a classifier that combines these various modalities encoded in the human microbiome. We achieved accuracies of 0.98, 0.76, 0.84, and 0.97 for predicting patients with inflammatory bowel disease, type 2 diabetes, liver cirrhosis, and colorectal cancer, respectively; these are comparable or higher than classical machine learning methods. A deeper analysis was also performed on the resulting sets of selected features to understand the contribution of their different characteristics. MDL4Microbiome is a classifier with higher or comparable accuracy compared with other machine learning methods, which offers perspectives on feature generation with metagenome sequences in deep learning models and their advantages in the classification of host disease status.

Application of Deep Learning in Plant-Microbiota Association Analysis

Unraveling the association between microbiome and plant phenotype can illustrate the effect of microbiome on host and then guide the agriculture management. Adequate identification of species and appropriate choice of models are two challenges in microbiome data analysis. Computational models of microbiome data could help in association analysis between the microbiome and plant host. The deep learning methods have been widely used to learn the microbiome data due to their powerful strength of handling the complex, sparse, noisy, and high-dimensional data. Here, we review the analytic strategies in the microbiome data analysis and describe the applications of deep learning models for plant–microbiome correlation studies. We also introduce the application cases of different models in plant–microbiome correlation analysis and discuss how to adapt the models on the critical steps in data processing. From the aspect of data processing manner, model structure, and operating principle, most deep learning models are suitable for the plant microbiome data analysis. The ability of feature representation and pattern recognition is the advantage of deep learning methods in modeling and interpretation for association analysis. Based on published computational experiments, the convolutional neural network and graph neural networks could be recommended for plant microbiome analysis.