Diagnosis of Inflammatory Bowel Disease and Colorectal Cancer through Multi-View Stacked Generalization Applied on Gut Microbiome Data.

Most of the microbiome studies suggest that using ensemble models such as Random Forest results in best predictive power. In this study, we empirically evaluate a more powerful ensemble learning algorithm, multi-view stacked generalization, on pediatric inflammatory bowel disease and adult colorectal cancer patients' cohorts. We aim to check whether stacking would lead to better results compared to using a single best machine learning algorithm. Stacking achieves the best test set Average Precision (AP) on inflammatory bowel disease dataset reaching AP = 0.69, outperforming both the best base classifier (AP = 0.61) and the baseline meta learner built on top of base classifiers (AP = 0.63). On colorectal cancer dataset, the stacked classifier also outperforms (AP = 0.81) both the best base classifier (AP = 0.79) and the baseline meta learner (AP = 0.75). Stacking achieves best predictive performance on test set outperforming the best classifiers on both patient cohorts. Application of the stacking solves the issue of choosing the most appropriate machine learning algorithm by automating the model selection procedure. Clinical application of such a model is not limited to diagnosis task only, but it also can be extended to biomarker selection thanks to feature selection procedure.

Severe COVID-19 is characterised by inflammation and immature myeloid cells early in disease progression.

SARS-CoV-2 infection causes a wide spectrum of disease severity. Identifying the immunological characteristics of severe disease and the risk factors for their development are important in the management of COVID-19. This study aimed to identify and rank clinical and immunological features associated with progression to severe COVID-19 in order to investigate an immunological signature of severe disease. One hundred and eight patients with positive SARS-CoV-2 PCR were recruited. Routine clinical and laboratory markers were measured, as well as myeloid and lymphoid whole-blood immunophenotyping and measurement of the pro-inflammatory cytokines IL-6 and soluble CD25. All analysis was carried out in a routine hospital diagnostic laboratory. Univariate analysis demonstrated that severe disease was most strongly associated with elevated CRP and IL-6, loss of DLA-DR expression on monocytes and CD10 expression on neutrophils. Unbiased machine learning demonstrated that these four features were strongly associated with severe disease, with an average prediction score for severe disease of 0.925. These results demonstrate that these four markers could be used to identify patients developing severe COVID-19 and allow timely delivery of therapeutics.

Faecal microbiota transplantation halts progression of human new-onset type 1 diabetes in a randomised controlled trial.

OBJECTIVE: Type 1 diabetes (T1D) is characterised by islet autoimmunity and beta cell destruction. A gut microbiota-immunological interplay is involved in the pathophysiology of T1D. We studied microbiota-mediated effects on disease progression in patients with type 1 diabetes using faecal microbiota transplantation (FMT). DESIGN: Patients with recent-onset (<6 weeks) T1D (18-30 years of age) were randomised into two groups to receive three autologous or allogenic (healthy donor) FMTs over a period of 4 months. Our primary endpoint was preservation of stimulated C peptide release assessed by mixed-meal tests during 12 months. Secondary outcome parameters were changes in glycaemic control, fasting plasma metabolites, T cell autoimmunity, small intestinal gene expression profile and intestinal microbiota composition. RESULTS: Stimulated C peptide levels were significantly preserved in the autologous FMT group (n=10 subjects) compared with healthy donor FMT group (n=10 subjects) at 12 months. Small intestinal Prevotella was inversely related to residual beta cell function (r=-0.55, p=0.02), whereas plasma metabolites 1-arachidonoyl-GPC and 1-myristoyl-2-arachidonoyl-GPC levels linearly correlated with residual beta cell preservation (rho=0.56, p=0.01 and rho=0.46, p=0.042, respectively). Finally, baseline CD4 +CXCR3+T cell counts, levels of small intestinal Desulfovibrio piger and CCL22 and CCL5 gene expression in duodenal biopsies predicted preserved beta cell function following FMT irrespective of donor characteristics. CONCLUSION: FMT halts decline in endogenous insulin production in recently diagnosed patients with T1D in 12 months after disease onset. Several microbiota-derived plasma metabolites and bacterial strains were linked to preserved residual beta cell function. This study provides insight into the role of the intestinal gut microbiome in T1D. TRIAL REGISTRATION NUMBER: NTR3697.

Infusion of donor feces affects the gut-brain axis in humans with metabolic syndrome.

OBJECTIVE: Increasing evidence indicates that intestinal microbiota play a role in diverse metabolic processes via intestinal butyrate production. Human bariatric surgery data suggest that the gut-brain axis is also involved in this process, but the underlying mechanisms remain unknown. METHODS: We compared the effect of fecal microbiota transfer (FMT) from post-Roux-en-Y gastric bypass (RYGB) donors vs oral butyrate supplementation on (123I-FP-CIT-determined) brain dopamine transporter (DAT) and serotonin transporter (SERT) binding as well as stable isotope-determined insulin sensitivity at baseline and after 4 weeks in 24 male and female treatment-naïve metabolic syndrome subjects. Plasma metabolites and fecal microbiota were also determined at these time points. RESULTS: We observed an increase in brain DAT after donor FMT compared to oral butyrate that reduced this binding. However, no effect on body weight and insulin sensitivity was demonstrated after post-RYGB donor feces transfer in humans with metabolic syndrome. Increases in fecal levels of Bacteroides uniformis were significantly associated with an increase in DAT, whereas increases in Prevotella spp. showed an inverse association. Changes in the plasma metabolites glycine, betaine, methionine, and lysine (associated with the S-adenosylmethionine cycle) were also associated with altered striatal DAT expression. CONCLUSIONS: Although more and larger studies are needed, our data suggest a potential gut microbiota-driven modulation of brain dopamine and serotonin transporters in human subjects with obese metabolic syndrome. These data also suggest the presence of a gut-brain axis in humans that can be modulated. NTR REGISTRATION: 4488.

Treatment with Anaerobutyricum soehngenii: a pilot study of safety and dose-response effects on glucose metabolism in human subjects with metabolic syndrome.

Dysbiosis of the intestinal microbiota has been implicated in insulin resistance, although evidence regarding causality in humans is scarce. We performed a phase I/II dose-finding and safety study on the effect of oral intake of the anaerobic butyrogenic strain Anaerobutyricum soehngenii on glucose metabolism in 24 subjects with metabolic syndrome. We found that treatment with A. soehngenii was safe and observed a significant correlation between the measured fecal abundance of administered A. soehngenii and improvement in peripheral insulin sensitivity after 4 weeks of treatment. This was accompanied by an altered microbiota composition and a change in bile acid metabolism. Finally, we show that metabolic response upon administration of A. soehngenii (defined as improved insulin sensitivity 4 weeks after A. soehngenii intake) is dependent on microbiota composition at baseline. These data in humans are promising, but additional studies are needed to reproduce our findings and to investigate long-term effects, as well as other modes of delivery.

A salivary metabolite signature that reflects gingival host-microbe interactions: instability predicts gingivitis susceptibility.

Several proteins and peptides in saliva were shown to stimulate gingival wound repair, but the role of salivary metabolites in this process remains unexplored. In vitro gingival re-epithelialization kinetics were determined using unstimulated saliva samples from healthy individuals collected during an experimental gingivitis study. Elastic net regression with stability selection identified a specific metabolite signature in a training dataset that was associated with the observed re-epithelialization kinetics and enabled its prediction for all saliva samples obtained in the clinical study. This signature encompassed ten metabolites, including plasmalogens, diacylglycerol and amino acid derivatives, which reflect enhanced host-microbe interactions. This association is in agreement with the positive correlation of the metabolite signature with the individual's gingival bleeding index. Remarkably, intra-individual signature-variation over time was associated with elevated risk for gingivitis development. Unravelling how these metabolites stimulate wound repair could provide novel avenues towards therapeutic approaches in patients with impaired wound healing capacity.

Oral butyrate does not affect innate immunity and islet autoimmunity in individuals with longstanding type 1 diabetes: a randomised controlled trial.

AIMS/HYPOTHESIS: The pathophysiology of type 1 diabetes has been linked to altered gut microbiota and more specifically to a shortage of intestinal production of the short-chain fatty acid (SCFA) butyrate, which may play key roles in maintaining intestinal epithelial integrity and in human and gut microbial metabolism. Butyrate supplementation can protect against autoimmune diabetes in mouse models. We thus set out to study the effect of oral butyrate vs placebo on glucose regulation and immune variables in human participants with longstanding type 1 diabetes. METHODS: We administered a daily oral dose of 4 g sodium butyrate or placebo for 1 month to 30 individuals with longstanding type 1 diabetes, without comorbidity or medication use, in a randomised (1:1), controlled, double-blind crossover trial, with a washout period of 1 month in between. Participants were randomly allocated to the 'oral sodium butyrate capsules first' or 'oral placebo capsules first' study arm in blocks of five. The clinical investigator received blinded medication from the clinical trial pharmacy. All participants, people doing measurements or examinations, or people assessing the outcomes were blinded to group assignment. The primary outcome was a change in the innate immune phenotype (monocyte subsets and in vitro cytokine production). Secondary outcomes were changes in blood markers of islet autoimmunity (cell counts, lymphocyte stimulation indices and CD8 quantum dot assays), glucose and lipid metabolism, beta cell function (by mixed-meal test), gut microbiota and faecal SCFA. The data was collected at the Amsterdam University Medical Centers. RESULTS: All 30 participants were analysed. Faecal butyrate and propionate levels were significantly affected by oral butyrate supplementation and butyrate treatment was safe. However, this modulation of intestinal SCFAs did not result in any significant changes in adaptive or innate immunity, or in any of the other outcome variables. In our discussion, we elaborate on this important discrepancy with previous animal work. CONCLUSIONS/INTERPRETATION: Oral butyrate supplementation does not significantly affect innate or adaptive immunity in humans with longstanding type 1 diabetes. TRIAL REGISTRATION: Netherlands Trial Register: NL4832 (www.trialregister.nl). DATA AVAILABILITY: Raw sequencing data are available in the European Nucleotide Archive repository (https://www.ebi.ac.uk/ena/browse) under study PRJEB30292. FUNDING: The study was funded by a Le Ducq consortium grant, a CVON grant, a personal ZONMW-VIDI grant and a Dutch Heart Foundation grant.

Domain intelligible models.

Mining biological information from rich "-omics" datasets is facilitated by organizing features into groups that are related to a biological phenomenon or clinical outcome. For example, microorganisms can be grouped based on a phylogenetic tree that depicts their similarities regarding genetic or physical characteristics. Here, we describe algorithms that incorporate auxiliary information in terms of groups of predictors and the relationships between them into the metagenome learning task to build intelligible models. In particular, our cost function guides the feature selection process using auxiliary information by requiring related groups of predictors to provide similar contributions to the final response. We apply the developed algorithms to a recently published dataset analyzing the effects of fecal microbiota transplantation (FMT) in order to identify factors that are associated with improved peripheral insulin sensitivity, leading to accurate predictions of the response to the FMT.

On the ecosystemic network of saliva in healthy young adults.

A dysbiotic state is believed to be a key factor in the onset of oral disease. Although oral diseases have been studied for decades, our understanding of oral health, the boundaries of a healthy oral ecosystem and ecological shift toward dysbiosis is still limited. Here, we present the ecobiological heterogeneity of the salivary ecosystem and relations between the salivary microbiome, salivary metabolome and host-related biochemical salivary parameters in 268 healthy adults after overnight fasting. Gender-specific differences in the microbiome and metabolome were observed and were associated with salivary pH and dietary protein intake. Our analysis grouped the individuals into five microbiome and four metabolome-based clusters that significantly related to biochemical parameters of saliva. Low salivary pH and high lysozyme activity were associated with high proportions of streptococcal phylotypes and increased membrane-lipid degradation products. Samples with high salivary pH displayed increased chitinase activity, higher abundance of Veillonella and Prevotella species and higher levels of amino acid fermentation products, suggesting proteolytic adaptation. An over-specialization toward either a proteolytic or a saccharolytic ecotype may indicate a shift toward a dysbiotic state. Their prognostic value and the degree to which these ecotypes are related to increased disease risk remains to be determined.

A Study of the Variation in the Salivary Peptide Profiles of Young Healthy Adults Acquired Using MALDI-TOF MS.

A cross-sectional observational study was conducted to evaluate the inter-individual variation in the MALDI-TOF MS peptide profiles of unstimulated whole saliva in a population of 268 systemically healthy adults aged 18-30 yr (150 males and 118 females) with no apparent caries lesions or periodontal disease. Using Spectral Clustering, four subgroups of individuals were identified within the study population. These subgroups were delimited by the pattern of variation in 9 peaks detected in the 2-15 kDa m/z range. An Unsupervised Feature Selection algorithm showed that P-C peptide, a 44 residue-long salivary acidic proline-rich protein, and three of its fragments (Fr. 1-25, Fr. 15-35 and Fr. 15-44) play a central role in delimiting the subgroups. Significant differences were found in the salivary biochemistry of the subgroups with regard to lysozyme and chitinase, two enzymes that are part of the salivary innate defense system (p < 0.001). These results suggest that MALDI-TOF MS salivary peptide profiles may relate information on the underlying state of the oral ecosystem and may provide a useful reference for salivary disease biomarker discovery studies.

Personalized microbial network inference via co-regularized spectral clustering.

We use Human Microbiome Project (HMP) cohort (Peterson et al., 2009) to infer personalized oral microbial networks of healthy individuals. To determine clustering of individuals with similar microbial profiles, co-regularized spectral clustering algorithm is applied to the dataset. For each cluster we discovered, we compute co-occurrence relationships among the microbial species that determine microbial network per cluster of individuals. The results of our study suggest that there are several differences in microbial interactions on personalized network level in healthy oral samples acquired from various niches. Based on the results of co-regularized spectral clustering we discover two groups of individuals with different topology of their microbial interaction network. The results of microbial network inference suggest that niche-wise interactions are different in these two groups. Our study shows that healthy individuals have different microbial clusters according to their oral microbiota. Such personalized microbial networks open a better understanding of the microbial ecology of healthy oral cavities and new possibilities for future targeted medication. The scripts written in scientific Python and in Matlab, which were used for network visualization, are provided for download on the website http://learning-machines.com/.

Interindividual variation, correlations, and sex-related differences in the salivary biochemistry of young healthy adults.

A cross-sectional observational study was conducted to evaluate interindividual biochemical variation in unstimulated whole saliva in a population of 268 systemically healthy young students, 18-30 yr of age, with no apparent caries lesions or periodontal disease. Salivary flow rate, protein content, pH, buffering capacity, mucins MUC5B and MUC7, albumin, secretory IgA, cystatin S, lactoferrin, chitinase, amylase, lysozyme, and proteases were measured using ELISAs and enzymatic activity assays. Significant differences were found between male and female subjects. Salivary pH, buffering capacity, protein content, MUC5B, secretory IgA, and chitinase activity were all lower in female subjects compared with male subjects, whereas MUC7 and lysozyme activity were higher in female subjects. There was no significant difference between sexes in salivary flow rate, albumin, cystatin S, amylase, and protease activity. Principal component analysis (PCA) and spectral clustering (SC) were used to assess intervariable relationships within the data set and to identify subgroups. Spectral clustering identified two clusters of participants, which were subsequently described. This study provides a comprehensive overview of the distribution and inter-relations of a set of important salivary biochemical variables in a systemically healthy young adult population, free of apparent caries lesions and periodontal disease. It highlights significant gender differences in salivary biochemistry.