GutBalance: a server for the human gut microbiome-based disease prediction and biomarker discovery with compositionality addressed

M3HOGAT: A Multi-View Multi-Modal Multi-Scale High-Order Graph Attention Network for Microbe-Disease Association Prediction

Numerous scientific studies have found a link between diverse microorganisms in the human body and complex human diseases. Because traditional experimental approaches are time-consuming and expensive, using computational methods to identify microbes correlated with diseases is critical. In this paper, a new microbe-disease association prediction model is proposed that combines a multi-view multi-modal network and a multi-scale feature fusion mechanism, called M3HOGAT. Firstly, a microbe-disease association network and multiple similarity views are constructed based on multi-source information. Then, consider that neighbor information from disparate orders might be more adept at learning node representations. Consequently, the higher-order graph attention network (HOGAT) is devised to aggregate neighbor information from disparate orders to extract microbe and disease features from different networks and views. Given that the embedding features of microbe and disease from different views possess varying importance, a multi-scale feature fusion mechanism is employed to learn their interaction information, thereby generating the final feature of microbes and diseases. Finally, an inner product decoder is used to reconstruct the microbe-disease association matrix. Compared with five state-of-the-art methods on the HMDAD and Disbiome datasets, the results of 5-fold cross-validations show that M3HOGAT achieves the best performance. Furthermore, case studies on asthma and obesity confirm the effectiveness of M3HOGAT in identifying potential disease-related microbes.

MicroHDF: predicting host phenotypes with metagenomic data using a deep forest-based framework

The gut microbiota plays a vital role in human health, and significant effort has been made to predict human phenotypes, especially diseases, with the microbiota as a promising indicator or predictor with machine learning (ML) methods. However, the accuracy is impacted by a lot of factors when predicting host phenotypes with the metagenomic data, e.g. small sample size, class imbalance, high-dimensional features, etc. To address these challenges, we propose MicroHDF, an interpretable deep learning framework to predict host phenotypes, where a cascade layers of deep forest units is designed for handling sample class imbalance and high dimensional features. The experimental results show that the performance of MicroHDF is competitive with that of existing state-of-the-art methods on 13 publicly available datasets of six different diseases. In particular, it performs best with the area under the receiver operating characteristic curve of 0.9182 ± 0.0098 and 0.9469 ± 0.0076 for inflammatory bowel disease (IBD) and liver cirrhosis, respectively. Our MicroHDF also shows better performance and robustness in cross-study validation. Furthermore, MicroHDF is applied to two high-risk diseases, IBD and autism spectrum disorder, as case studies to identify potential biomarkers. In conclusion, our method provides an effective and reliable prediction of the host phenotype and discovers informative features with biological insights.

Overview of data preprocessing for machine learning applications in human microbiome research

Although metagenomic sequencing is now the preferred technique to study microbiome-host interactions, analyzing and interpreting microbiome sequencing data presents challenges primarily attributed to the statistical specificities of the data (e.g., sparse, over-dispersed, compositional, inter-variable dependency). This mini review explores preprocessing and transformation methods applied in recent human microbiome studies to address microbiome data analysis challenges. Our results indicate a limited adoption of transformation methods targeting the statistical characteristics of microbiome sequencing data. Instead, there is a prevalent usage of relative and normalization-based transformations that do not specifically account for the specific attributes of microbiome data. The information on preprocessing and transformations applied to the data before analysis was incomplete or missing in many publications, leading to reproducibility concerns, comparability issues, and questionable results. We hope this mini review will provide researchers and newcomers to the field of human microbiome research with an up-to-date point of reference for various data transformation tools and assist them in choosing the most suitable transformation method based on their research questions, objectives, and data characteristics.

'Multi-omics' data integration: applications in probiotics studies

The concept of probiotics is witnessing increasing attention due to its benefits in influencing the host microbiome and the modulation of host immunity through the strengthening of the gut barrier and stimulation of antibodies. These benefits, combined with the need for improved nutraceuticals, have resulted in the extensive characterization of probiotics leading to an outburst of data generated using several 'omics' technologies. The recent development in system biology approaches to microbial science is paving the way for integrating data generated from different omics techniques for understanding the flow of molecular information from one 'omics' level to the other with clear information on regulatory features and phenotypes. The limitations and tendencies of a 'single omics' application to ignore the influence of other molecular processes justify the need for 'multi-omics' application in probiotics selections and understanding its action on the host. Different omics techniques, including genomics, transcriptomics, proteomics, metabolomics and lipidomics, used for studying probiotics and their influence on the host and the microbiome are discussed in this review. Furthermore, the rationale for 'multi-omics' and multi-omics data integration platforms supporting probiotics and microbiome analyses was also elucidated. This review showed that multi-omics application is useful in selecting probiotics and understanding their functions on the host microbiome. Hence, recommend a multi-omics approach for holistically understanding probiotics and the microbiome.

Changes in the microbiota in different intestinal segments of mice with sepsis

Introduction

The small intestine, as the main digestion and absorption site of the gastrointestinal tract, is often overlooked in studies, and the overall microbiota does not reflect the makeup of the microbiota in different segments of the intestine. Therefore, we aimed to exclude the influence of routine ICU treatment measures on sepsis patients and observed changes in the diversity and abundance of gut microbiota in different intestinal segments of septic mice.

Methods

The mice were randomly divided into the CLP6h group and the sham group. The contents of the colon and small intestine of the experimental group and the control group were collected after 6 h.

Results

After CLP, the number and structure of the gut microbiota in the colon changed most obviously, among which Bacteroidetes had the most significant changes. Akkermansia, D.Firmicutes_bacterium_M10_2, Blautia, Bifidobacterium, Lactobacillus, Candidatus_Arthromitus, and Muribaculaceae were changed in the colon. Lactobacillus, Bifidobacterium, Akkermansia, Blautia, Candidatus_Arthromitus, and Lachnospiraceae_NK4A136_group were changed in the small intestine.

Discussion

Our experiment found that there were different numbers of unique and common gut microbiota in the small intestine and colon after sepsis, and the gut microbiota of the colon changed more drastically after sepsis than the small intestine. Thus, we should focus on protective gut microbiota and mucin-degrading microbes. We hope that these results will provide help for sepsis treatment in the future.

Investigating causal relationships between the gut microbiota and allergic diseases: A mendelian randomization study

Observational studies revealed altered gut microbial composition in patients with allergic diseases, which illustrated a strong association between the gut microbiome and the risk of allergies. However, whether such associations reflect causality remains to be well-documented. Two-sample mendelian randomization (2SMR) was performed to estimate the potential causal effect between the gut microbiota and the risk of allergic diseases. 3, 12, and 16 SNPs at the species, genus, and family levels respectively of 15 microbiome features were obtained as the genetic instruments of the exposure dataset from a previous study. GWAS summary data of a total of 17 independent studies related to allergic diseases were collected from the IEU GWAS database for the outcome dataset. Significant causal relationships were obtained between gut microbiome features including Ruminococcaceae, Eggerthella, Bifidobacterium, Faecalibacterium, and Bacteroides and the risk of allergic diseases. Furthermore, our results also pointed out a number of putative associations between the gut microbiome and allergic diseases. Taken together, this study was the first study using the approach of 2SMR to elucidate the association between gut microbiome and allergic diseases.

Principal microbial groups: compositional alternative to phylogenetic grouping of microbiome data

Statistical and machine learning techniques based on relative abundances have been used to predict health conditions and to identify microbial biomarkers. However, high dimensionality, sparsity and the compositional nature of microbiome data represent statistical challenges. On the other hand, the taxon grouping allows summarizing microbiome abundance with a coarser resolution in a lower dimension, but it presents new challenges when correlating taxa with a disease. In this work, we present a novel approach that groups Operational Taxonomical Units (OTUs) based only on relative abundances as an alternative to taxon grouping. The proposed procedure acknowledges the compositional data making use of principal balances. The identified groups are called Principal Microbial Groups (PMGs). The procedure reduces the need for user-defined aggregation of $\textrm{OTU}$s and offers the possibility of working with coarse group of $\textrm{OTU}$s, which are not present in a phylogenetic tree. PMGs can be used for two different goals: (1) as a dimensionality reduction method for compositional data, (2) as an aggregation procedure that provides an alternative to taxon grouping for construction of microbial balances afterward used for disease prediction. We illustrate the procedure with a cirrhosis study data. PMGs provide a coherent data analysis for the search of biomarkers in human microbiota. The source code and demo data for PMGs are available at: https://github.com/asliboyraz/PMGs.

iProbiotics: a machine learning platform for rapid identification of probiotic properties from whole-genome primary sequences

Lactic acid bacteria consortia are commonly present in food, and some of these bacteria possess probiotic properties. However, discovery and experimental validation of probiotics require extensive time and effort. Therefore, it is of great interest to develop effective screening methods for identifying probiotics. Advances in sequencing technology have generated massive genomic data, enabling us to create a machine learning-based platform for such purpose in this work. This study first selected a comprehensive probiotics genome dataset from the probiotic database (PROBIO) and literature surveys. Then, k-mer (from 2 to 8) compositional analysis was performed, revealing diverse oligonucleotide composition in strain genomes and apparently more probiotic (P-) features in probiotic genomes than non-probiotic genomes. To reduce noise and improve computational efficiency, 87 376 k-mers were refined by an incremental feature selection (IFS) method, and the model achieved the maximum accuracy level at 184 core features, with a high prediction accuracy (97.77%) and area under the curve (98.00%). Functional genomic analysis using annotations from gene ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) and Rapid Annotation using Subsystem Technology (RAST) databases, as well as analysis of genes associated with host gastrointestinal survival/settlement, carbohydrate utilization, drug resistance and virulence factors, revealed that the distribution of P-features was biased toward genes/pathways related to probiotic function. Our results suggest that the role of probiotics is not determined by a single gene, but by a combination of k-mer genomic components, providing new insights into the identification and underlying mechanisms of probiotics. This work created a novel and free online bioinformatic tool, iProbiotics, which would facilitate rapid screening for probiotics.

Enhancing Physical and Thermodynamic Properties of DNA Storage Sets with End-constraint

With the explosion of data, DNA is considered as an ideal carrier for storage due to its high storage density. However, low-quality DNA sets hamper the widespread use of DNA storage. This work proposes a new method to design high-quality DNA storage sets. Firstly, random switch and double-weight offspring strategies are introduced in Double-strategy Black Widow Optimization Algorithm (DBWO). Experimental results of 26 benchmark functions show that the exploration and exploitation abilities of DBWO are greatly improved from previous work. Secondly, DBWO is applied in designing DNA storage sets, and compared with previous work, the lower bounds of storage sets are boosted by 9%-37%. Finally, to improve the poor stabilities of sequences, the End-constraint is proposed in designing DNA storage sets. By measuring the number of hairpin structures, melting temperature, and minimum free energy, it is evaluated that with our innovative constraint, DBWO can construct not only a larger number of storage sets, but also enhance physical and thermodynamic properties of DNA storage sets.

919 Syrup Alleviates Postpartum Depression by Modulating the Structure and Metabolism of Gut Microbes and Affecting the Function of the Hippocampal GABA/Glutamate System

Postpartum depression (PPD) is a mental disorder that affects pregnant women around the world, with serious consequences for mothers, families, and children. Its pathogenesis remains unclear, and medications for treating PPD that can be used during lactation remain to be identified. 919 syrup (919 TJ) is a Chinese herbal medicine that has been shown to be beneficial in the treatment of postpartum depression in both clinical and experimental studies. The mechanism of action of 919 TJ is unclear. 919 syrup is ingested orally, making the potential interaction between the drug and the gut microbiome impossible to ignore. We therefore hypothesized that 919 syrup could improve the symptoms of postpartum depression by affecting the structure and function of the intestinal flora, thereby altering hippocampal metabolism. We compared changes in hippocampal metabolism, fecal metabolism, and intestinal microflora of control BALB/c mice, mice with induced untreated PPD, and mice with induced PPD treated with 919 TJ, and found that 4-aminobutyric acid (GABA) in the hippocampus corresponded with PPD behaviors. Based on changes in GABA levels, multiple key gut bacterial species (Mucispirillum schaedleri, Bifidobacterium pseudolongum, Desulfovibrio piger, Alloprevotella tannerae, Bacteroides sp.2.1.33B and Prevotella sp. CAG:755) were associated with PPD. Metabolic markers that may represent the function of the intestinal microbiota in mice with PPD were identified (Met-Arg, urocanic acid, thioetheramide-PC, L-pipecolic acid, and linoleoyl ethanolamide). The relationship between these factors is not a simple one-to-one correspondence, but more likely a network of staggered functions. We therefore believe that the composition and function of the entire intestinal flora should be emphasized in research studying the gut and PPD, rather than changes in the abundance of individual bacterial species. The introduction of this concept of “GutBalance” may help clarify the relationship between gut bacteria and systemic disease.

DisBalance: a platform to automatically build balance-based disease prediction models and discover microbial biomarkers from microbiome data

How best to utilize the microbial taxonomic abundances in regard to the prediction and explanation of human diseases remains appealing and challenging, and the relative nature of microbiome data necessitates a proper feature selection method to resolve the compositional problem. In this study, we developed an all-in-one platform to address a series of issues in microbiome-based human disease prediction and taxonomic biomarkers discovery. We prioritize the interpretation, runtime and classification accuracy of the distal discriminative balances analysis (DBA-distal) method in selecting a set of distal discriminative balances, and develop DisBalance, a comprehensive platform, to integrate and streamline the workflows of disease model building, disease risk prediction and disease-related biomarker discovery for microbiome-based binary classifications. DisBalance allows the de novo model-building and disease risk prediction in a very fast and convenient way. To facilitate the model-driven and knowledge-driven discoveries, DisBalance dedicates multiple strategies for the mining of microbial biomarkers. The independent validation of the models constructed by the DisBalance pipeline is performed on seven microbiome datasets from the original article of DBA-distal. The implementation of the DisBalance platform is demonstrated by a complete analysis of a shotgun metagenomic dataset of Ulcerative Colitis (UC). As a free and open-source, DisBlance can be accessed at http://lab.malab.cn/soft/DisBalance. The source code and demo data for Disbalance are available at https://github.com/yangfenglong/DisBalance.

Quantitative Analysis and Visualization of the Interaction Between Intestinal Microbiota and Type 1 Diabetes in Children Based on Multi-Databases

Background: As an important autoimmune disease, type 1 diabetes (T1D) is often diagnosed in children, but due to the complexity of the etiology of diabetes and many other factors, the disease pathogenesis of diabetes is still unclear. The intestinal microbiota has been proved to have close relationships with T1D in recent years, which is one of the most important molecular bases of pathogenesis and prognosis factors for T1D. Using the multi-omics and multicenter sample analysis method, a number of intestinal microbiota in T1D have been discovered and explained, which has provided comprehensive and rich information. However, how to find more useful information and get an intuitive understanding that people need conveniently in the huge data sea has become the focus of attention. Therefore, quantitative analysis and visualization of the interaction between intestinal microbiota and T1D in children are urgently needed. Methods: We retrieved the detailed original data from the National Center for Biotechnology Information, GMREPO, and gutMEGA databases and other authoritative multiple projects with related research; the ranking of intestinal microbiota abundance from healthy people, overall T1D patients, and T1D in children (0-18 years old) were detailed analyzed, classified, and visualized. Results: A total of 515 bacterial species and 161 related genera were fully analyzed. Also, Prevotella copri was led by 21.25% average abundance, followed by Clostridium tertium of 10.39% in all-cross T1D patients. For children with T1D, Bacteroides vulgatus has high abundance in all age periods, whereas the abundance of each intestinal microbiota was more uniform in female samples, with the ranking from high to low as Bacteroides dorei 9.56%, P. copri 9.53%, Streptococcus pasteurianus 8.15%, and C. tertium 7.53%, whereas in male samples, P. copri was accounted for the largest by 22.72%. The interaction between intestinal microbiota and comparison between healthy people and children with T1D was also detailed analyzed. Conclusions: This study provides a new method and comprehensive perspectives for the evaluation of the interaction between intestinal microbiota and T1D in children. A set of useful information of intestinal microbiota with its internal interaction and connections has been presented, which could be a compact, immediate, and practical scientific reference for further molecular biological and clinical translational research of T1D in children.