Usefulness of Machine Learning-Based Gut Microbiome Analysis for Identifying Patients with Irritable Bowels Syndrome.
ABSTRACT: Irritable bowel syndrome (IBS) is diagnosed by subjective clinical symptoms. We aimed to establish an objective IBS prediction model based on gut microbiome analyses employing machine learning. We collected fecal samples and clinical data from 85 adult patients who met the Rome III criteria for IBS, as well as from 26 healthy controls. The fecal gut microbiome profiles were analyzed by 16S ribosomal RNA sequencing, and the determination of short-chain fatty acids was performed by gas chromatography-mass spectrometry. The IBS prediction model based on gut microbiome data after machine learning was validated for its consistency for clinical diagnosis. The fecal microbiome alpha-diversity indices were significantly smaller in the IBS group than in the healthy controls. The amount of propionic acid and the difference between butyric acid and valerate were significantly higher in the IBS group than in the healthy controls (p < 0.05). Using LASSO logistic regression, we extracted a featured group of bacteria to distinguish IBS patients from healthy controls. Using the data for these featured bacteria, we established a prediction model for identifying IBS patients by machine learning (sensitivity >80%; specificity >90%). Gut microbiome analysis using machine learning is useful for identifying patients with IBS.
Project description:Irritable Bowel Syndrome (IBS) is a disorder of the gut-brain axis, characterized by altered gut function and frequent psychiatric co-morbidity. Although altered intestinal microbiome profiles have been documented, their relevance to the clinical expression of IBS is unknown. To evaluate a functional role of the microbiota, we colonized germ-free mice with fecal microbiota from healthy controls or IBS patients with accompanying anxiety, and monitored gut function and behavior. Mouse microbiota profiles clustered according to their human donors. Despite having taxonomically similar composition as controls, mice with IBS microbiota had distinct serum metabolomic profiles related to neuro- and immunomodulation. Mice with IBS, but not control microbiota, exhibited faster gastrointestinal transit, intestinal barrier dysfunction, innate immune activation and anxiety-like behavior. These results support the notion that the microbiota contributes to both intestinal and behavioral manifestations of IBS and rationalize the use of microbiota-directed therapies in ameliorating IBS. Overall design: We extracted total RNA of colonic tissues extracted from germ-free NIH swiss mice (10-12 weeks) colonized with healthy human gut microbiota (10 mice per fecal sample, 5 fecal samples, total=50 mice), and mice colonized with human microbiota of IBS patients (10 mice per sample, 6 fecal samples, total =60 mice). We then pooled together all the RNA samples of all the mice colonized with the same microbiota, to form a representative sample for the colonization with that specific human fecal sample. We finally ran 11 samples in total through the NanoString nCounter® Gene Expression CodeSet for mouse inflammation genes.
Project description:Accurate diagnosis and stratification of children with irritable bowel syndrome (IBS) remain challenging. Given the central role of recurrent abdominal pain in IBS, we evaluated the relationships of pediatric IBS and abdominal pain with intestinal microbes and fecal metabolites using a comprehensive clinical characterization and multiomics strategy. Using rigorous clinical phenotyping, we identified preadolescent children (aged 7 to 12 years) with Rome III IBS (n = 23) and healthy controls (n = 22) and characterized their fecal microbial communities using whole-genome shotgun metagenomics and global unbiased fecal metabolomic profiling. Correlation-based approaches and machine learning algorithms identified associations between microbes, metabolites, and abdominal pain. IBS cases differed from controls with respect to key bacterial taxa (eg, Flavonifractor plautii and Lachnospiraceae bacterium 7_1_58FAA), metagenomic functions (eg, carbohydrate metabolism and amino acid metabolism), and higher-order metabolites (eg, secondary bile acids, sterols, and steroid-like compounds). Significant associations between abdominal pain frequency and severity and intestinal microbial features were identified. A random forest classifier built on metagenomic and metabolic markers successfully distinguished IBS cases from controls (area under the curve, 0.93). Leveraging multiple lines of evidence, intestinal microbes, genes/pathways, and metabolites were associated with IBS, and these features were capable of distinguishing children with IBS from healthy children. These multi-omics features, and their links to childhood IBS coupled with nutritional interventions, may lead to new microbiome-guided diagnostic and therapeutic strategies.
Project description:Effects of the microbiome associated with diarrhea-predominant irritable bowel syndrome (IBS-D) on the gut have been reported, but no study has reported the effects of the IBS-D gut microbiome on the liver. We transplanted the fecal microbiota from an IBS-D patient and from a healthy volunteer to GF rats. The hepatic inflammation, serum biochemical parameters and metabolome, fecal microbiota profile, fecal short-chain fatty acids (SCFAs), and correlations among them before and after berberine intervention were assessed. Compared with the healthy control fecal microbiome transplantation (FMT) rats, the fecal microbiota of IBS-D patients induces significant Kupffer cell hyperplasia, hepatic sinusoid hypertrophy, and elevated levels of hepatic tumor necrosis factor-<i>?</i> and interferon-<i>?</i> and decreases the synthesis of ALB in GF rats. This is possibly related to <i>Faecalibacterium</i> and <i>Bifidobacterium</i> attributable to fecal formate, acetate, and propionate levels, which are associated with the host linoleic acid pathway. Berberine can partially reverse the Kupffer cell hyperplasia, <i>Faecalibacterium</i>, fecal formate, acetate, and propionate by modulating the gut microbiome composition. These results may imply that IBS-D not only is an intestinal functional disorder but can cause liver inflammation, thus providing some implications regarding the clinical cognition and treatment of IBS-D.
Project description:BACKGROUND:Irritable bowel syndrome (IBS) is reported associated with the alteration of gut microbial composition termed as dysbiosis. However, the pathogenic mechanism of IBS remains unclear, while the studies of Chinese individuals are scarce. This study aimed to understand the concept of dysbiosis among patients with Chinese diarrhea-predominant IBS (IBS-D), as a degree of variance between the gut microbiomes of IBS-D population and that of a healthy population. METHODS:The patients with IBS-D were recruited (assessed according to the Rome III criteria, by IBS symptom severity score) from the Outpatient Department of Gastroenterology of Peking University Third Hospital, and volunteers as healthy controls (HCs) were enrolled, during 2013. The 16S rRNA sequences were extracted from fecal samples. Ribosomal database project resources, basic local alignment search tool, and SparCC software were used to obtain the phylotype composition of samples and the internal interactions of the microbial community. Herein, the non-parametric test, Wilcoxon rank-sum test was carried out to find the statistical significance between HC and IBS-D groups. All the P values were adjusted to q values to decrease the error rate. RESULTS:The study characterized the gut microbiomes of Chinese patients with IBS-D, and demonstrated that the dysbiosis could be characterized as directed alteration of the microbiome composition leading to greater disparity between relative abundance of two phyla, Bacteroidetes (Z = 4.77, q = 1.59 × 10) and Firmicutes (Z = -3.87, q = 5.83 × 10). Moreover, it indicated that the IBS symptom features were associated with the dysbiosis of whole gut microbiome, instead of one or several certain genera even they were dominating. Two genera, Bacteroides and Lachnospiracea incertae sedis, were identified as the core genera, meanwhile, the non-core genera contribute to a larger pan-microbiome of the gut microbiome. Furthermore, the dysbiosis in patients with IBS-D was associated with a reduction of network complexity of the interacted microbial community (HC vs. IBS-D: 639 vs. 154). The disordered metabolic functions of patients with IBS-D were identified as the potential influence of gut microbiome on the host (significant difference with q < 0.01 between HC and IBS-D). CONCLUSIONS:This study supported the view of the potential influence of gut microbiome on the symptom of Chinese patients with IBS-D, and further characterized dysbiosis in Chinese patients with IBS-D, thus provided more pathological evidences for IBS-D with the further understanding of dysbiosis.
Project description:Alterations in brain-gut-microbiome (BGM) interactions have been implicated in the pathogenesis of irritable bowel syndrome (IBS). Here, we apply a systems biology approach, leveraging neuroimaging and fecal metabolite data, to characterize BGM interactions that are driving IBS pathophysiology. Fecal samples and resting state fMRI images were obtained from 138 female subjects (99 IBS, 39 healthy controls (HCs)). Partial least-squares discriminant analysis (PLS-DA) was conducted to explore group differences, and partial correlation analysis explored significantly changed metabolites and neuroimaging data. All correlational tests were performed controlling for age, body mass index, and diet; results are reported after FDR correction, with q?<?0.05 as significant. Compared to HCs, IBS showed increased connectivity of the putamen with regions of the default mode and somatosensory networks. Metabolite pathways involved in nucleic acid and amino acid metabolism differentiated the two groups. Only a subset of metabolites, primarily amino acids, were associated with IBS-specific brain changes, including tryptophan, glutamate, and histidine. Histidine was the only metabolite positively associated with both IBS-specific alterations in brain connectivity. Our findings suggest a role for several amino acid metabolites in modulating brain function in IBS. These metabolites may alter brain connectivity directly, by crossing the blood-brain-barrier, or indirectly through peripheral mechanisms. This is the first study to integrate both neuroimaging and fecal metabolite data supporting the BGM model of IBS, building the foundation for future mechanistic studies on the influence of gut microbial metabolites on brain function in IBS.
Project description:Irritable bowel syndrome (IBS) is a disorder involving dysfunctional brain-gut interactions characterized by chronic recurrent abdominal pain, altered bowel habits, and negative emotion. Previous studies have linked the habenula to the pathophysiology of negative emotion and pain. However, no studies to date have investigated habenular function in IBS patients. In this study, we investigated the resting-state functional connectivity (rsFC) and effective connectivity of the habenula in 34 subjects with IBS and 34 healthy controls and assessed the feasibility of differentiating IBS patients from healthy controls using a machine learning method. Our results showed significantly enhanced rsFC of the habenula-left dorsolateral prefrontal cortex (dlPFC) and habenula-periaqueductal grey (PAG, dorsomedial part), as well as decreased rsFC of the habenula-right thalamus (dorsolateral part), in the IBS patients compared with the healthy controls. Habenula-thalamus rsFC was positively correlated with pain intensity (r = .467, p = .005). Dynamic causal modeling (DCM) revealed significantly decreased effective connectivity from the right habenula to the right thalamus in the IBS patients compared to the healthy controls that was negatively correlated with disease duration (r = -.407, p = .017). In addition, IBS was classified with an accuracy of 71.5% based on the rsFC of the habenula-dlPFC, habenula-thalamus, and habenula-PAG in a support vector machine (SVM), which was further validated in an independent cohort of subjects (N = 44, accuracy = 65.2%, p = .026). Taken together, these findings establish altered habenular rsFC and effective connectivity in IBS, which extends our mechanistic understanding of the habenula's role in IBS.
Project description:BACKGROUND:Psychological co-morbidities in irritable bowel syndrome (IBS) have been widely recognized, whereas less is known regarding the role of gut microbial and host metabolic changes in clinical and psychological symptoms in IBS. RESULTS:A total of 70 diarrhoea-predominant IBS (IBS-D) patients and 46 healthy controls were enrolled in this study. Stool and urine samples were collected from both groups for 16S rRNA gene sequencing and metabolomic analysis. The results showed that fecal microbiota in IBS-D featured depleted Faecalibacterium (adjusted P?= 0.034), Eubacterium rectale group (adjusted P?= 0.048), Subdoligranulum (adjusted P?= 0.041) and increased Prevotella (adjusted P?= 0.041). O-ureido-L-serine, 3,4-dihydroxybenzenesulfonic acid and (R)-2-Hydroxyglutarate demonstrated lower urinary concentrations in IBS-D patients. We further built correlation matrices between gut microbe abundance, differentiated metabolite quantities and clinical parameters. Dialister manifested negative association with IBS severity (r?=?-?0.285, P?=?0.017), anxiety (r?=?-?0.347, P?=?0.003) and depression level (r?=?-?0.308, P?=?0.010). Roseburia was negatively associated with IBS severity (r?=?-?0.298, P?=?0.012). Twenty metabolites correlated with anxiety or depression levels, including 3,4-dihydroxymandelaldehyde with SAS (r?=?-?0.383, P?=?0.001), 1-methylxanthine with SDS (r?=?-?0.347, P?=?0.004) and 1D-chiro-inositol with SAS (r?=?-?0.336, P?=?0.005). In analysis of microbe-metabolite relationship, 3,4-dihydroxymandelaldehyde and 1-methylxanthine were negatively correlated with relative abundance of Clostridiumsensu stricto. CONCLUSIONS:Our findings demonstrated altered microbial and metabolomic profiles associated with clinically and psychological symptoms in IBS-D patients, which may provide insights for further investigations.
Project description:Irritable bowel syndrome (IBS) is a chronic functional disorder and its development may be linked, directly and indirectly, to intestinal dysbiosis. Here we investigated the interactions between IBS symptoms and the gut microbiome, including the relation to rifaximin (1200 mg daily; 11.2 g per a treatment). We recruited 72 patients, including 31 with IBS-D (diarrhea), 11 with IBS-C (constipation), and 30 with IBS-M (mixed constipation and diarrhea) and 30 healthy controls (HCs). Of them, 68%, 64%, and 53% patients with IBS-D, IBS-C, and IBS-M, respectively, achieved 10-12 week-term improvement after the rifaximin treatment. Stool samples were collected before and after the treatment, and fecal microbiotic profiles were analyzed by deep sequencing of 16S rRNA, while stool metabolic profiles were studied by hydrogen 1-nuclear magnetic resonance ((1)H-NMR) and gas chromatography-mass spectrometry (GC-MS). Of 26 identified phyla, only Bacteroidetes, Firmicutes, Proteobacteria, and Actinobacteria were consistently found in all samples. Bacteroidetes was predominant in fecal samples from HCs and IBS-D and IBS-M subjects, whereas Firmicutes was predominant in samples from IBS-C subjects. Species richness, but not community diversity, differentiated all IBS patients from HCs. Metabolic fingerprinting, using NMR spectra, distinguished HCs from all IBS patients. Thirteen metabolites identified by GC-MS differed HCs and IBS patients. However, neither metagenomics nor metabolomics analyses identified significant differences between patients with and without improvement after treatment.
Project description:Altered gut microbiota are assumed to be involved in the pathogenesis of irritable bowel syndrome (IBS). However, gut microbiota alterations reported in different studies are divergent and sometimes even contradictory. To better elucidate the relationship between altered gut microbiota and IBS, we characterized fecal microbiota of diarrhea-predominant IBS (IBS-D) patients and further explored the effect of rifaximin on gut microbiota using bacterial 16S rRNA gene-targeted pyrosequencing. In our study, IBS-D patients defined by Rome III criteria and age-and-gender matched healthy controls (HC) were enrolled to investigate the fecal microbiota alterations. These IBS-D patients were then treated with rifaximin for 2 weeks and followed up for 10 weeks. Fecal microbiota alterations, small intestine bacterial overgrowth (SIBO) and gastrointestinal (GI) symptoms of IBS-D patients were analyzed before and after treatment. Our results showed fecal microbiota richness but not diversity was decreased in IBS-D patients as compared to HC and there were alterations of fecal microbiota at different taxonomy levels. The abundant phyla Firmicutes was significantly decreased and Bacteroidetes was increased in IBS-D patients. Moreover, the alterations of predominant fermenting bacteria such as Bacteroidales and Clostridiales might be involved in the pathophysiology of IBS-D. In addition, rifaximin was effective in terms of SIBO eradication and even GI symptoms of IBS-D patients achieved at least 10-week improvement after treatment. Furthermore, rifaximin induced alterations of some special bacteria rather than affected the overall composition of microbiota in IBS-D patients. Meanwhile, a potential decrease in propanoate and butanoate metabolism was found in these IBS-D patients after rifaximin treatment. Taken together, there were alterations of gut microbiota in IBS-D patients as compared to HC. Rifaximin could relieve GI symptoms, modify gut microbiota in IBS-D patients and eradicate SIBO in those patients with SIBO, suggesting an additional therapeutic mechanism of rifaximin in the treatment of IBS-D. Our findings of compositional gut microbiota alterations in IBS-D and the effect of rifaximin on the gut microbiota implied that altered gut microbiota were associated with the pathogenesis of IBS.
Project description:The pathophysiology of irritable bowel syndrome (IBS) remains unclear. Here we investigated the microbiome of a large cohort of patients to identify specific signatures for IBS subtypes. We examined the microbiome of 113 patients with IBS and 66 healthy controls. A subset of these participants provided two samples one month apart. We analyzed a total of 273 fecal samples, generating more than 20 million 16S rRNA sequences. In patients with IBS, a significantly lower microbial diversity was associated with a lower relative abundance of butyrate-producing bacteria (P = 0.002; q < 0.06), in particular in patients with IBS-D and IBS-M. IBS patients who did not receive any treatment harboured a lower abundance of Methanobacteria compared to healthy controls (P = 0.005; q = 0.05). Furthermore, significant correlations were observed between several bacterial taxa and sensation of flatulence and abdominal pain (P < 0.05). Altogether, our findings showed that IBS-M and IBS-D patients are characterized by a reduction of butyrate producing bacteria, known to improve intestinal barrier function, and a reduction of methane producing microorganisms a major mechanism of hydrogen disposal in the human colon, which could explain excess of abdominal gas in IBS.