The Importance of Objective Stool Classification in Fecal 1H-NMR Metabolomics: Exponential Increase in Stool Crosslinking Is Mirrored in Systemic Inflammation and Associated to Fecal Acetate and Methionine

A novel blood-based epigenetic biosignature in first-episode schizophrenia patients through automated machine learning

Schizophrenia (SCZ) is a chronic, severe, and complex psychiatric disorder that affects all aspects of personal functioning. While SCZ has a very strong biological component, there are still no objective diagnostic tests. Lately, special attention has been given to epigenetic biomarkers in SCZ. In this study, we introduce a three-step, automated machine learning (AutoML)-based, data-driven, biomarker discovery pipeline approach, using genome-wide DNA methylation datasets and laboratory validation, to deliver a highly performing, blood-based epigenetic biosignature of diagnostic clinical value in SCZ. Publicly available blood methylomes from SCZ patients and healthy individuals were analyzed via AutoML, to identify SCZ-specific biomarkers. The methylation of the identified genes was then analyzed by targeted qMSP assays in blood gDNA of 30 first-episode drug-naïve SCZ patients and 30 healthy controls (CTRL). Finally, AutoML was used to produce an optimized disease-specific biosignature based on patient methylation data combined with demographics. AutoML identified a SCZ-specific set of novel gene methylation biomarkers including IGF2BP1, CENPI, and PSME4. Functional analysis investigated correlations with SCZ pathology. Methylation levels of IGF2BP1 and PSME4, but not CENPI were found to differ, IGF2BP1 being higher and PSME4 lower in the SCZ group as compared to the CTRL group. Additional AutoML classification analysis of our experimental patient data led to a five-feature biosignature including all three genes, as well as age and sex, that discriminated SCZ patients from healthy individuals [AUC 0.755 (0.636, 0.862) and average precision 0.758 (0.690, 0.825)]. In conclusion, this three-step pipeline enabled the discovery of three novel genes and an epigenetic biosignature bearing potential value as promising SCZ blood-based diagnostics.

A characteristic cerebellar biosignature for bipolar disorder, identified with fully automatic machine learning

Background

Transcriptomic profile differences between patients with bipolar disorder and healthy controls can be identified using machine learning and can provide information about the potential role of the cerebellum in the pathogenesis of bipolar disorder.With this aim, user-friendly, fully automated machine learning algorithms can achieve extremely high classification scores and disease-related predictive biosignature identification, in short time frames and scaled down to small datasets.

Method

A fully automated machine learning platform, based on the most suitable algorithm selection and relevant set of hyper-parameter values, was applied on a preprocessed transcriptomics dataset, in order to produce a model for biosignature selection and to classify subjects into groups of patients and controls. The parent GEO datasets were originally produced from the cerebellar and parietal lobe tissue of deceased bipolar patients and healthy controls, using Affymetrix Human Gene 1.0 ST Array.

Results

Patients and controls were classified into two separate groups, with no close-to-the-boundary cases, and this classification was based on the cerebellar transcriptomic biosignature of 25 features (genes), with Area Under Curve 0.929 and Average Precision 0.955. The biosignature includes both genes connected before to bipolar disorder, depression, psychosis or epilepsy, as well as genes not linked before with any psychiatric disease. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis revealed participation of 4 identified features in 6 pathways which have also been associated with bipolar disorder.

Conclusion

Automated machine learning (AutoML) managed to identify accurately 25 genes that can jointly - in a multivariate-fashion - separate bipolar patients from healthy controls with high predictive power. The discovered features lead to new biological insights. Machine Learning (ML) analysis considers the features in combination (in contrast to standard differential expression analysis), removing both irrelevant as well as redundant markers, and thus, focusing to biological interpretation.

Advancing microbiome research with machine learning: key findings from the ML4Microbiome COST action

The rapid development of machine learning (ML) techniques has opened up the data-dense field of microbiome research for novel therapeutic, diagnostic, and prognostic applications targeting a wide range of disorders, which could substantially improve healthcare practices in the era of precision medicine. However, several challenges must be addressed to exploit the benefits of ML in this field fully. In particular, there is a need to establish "gold standard" protocols for conducting ML analysis experiments and improve interactions between microbiome researchers and ML experts. The Machine Learning Techniques in Human Microbiome Studies (ML4Microbiome) COST Action CA18131 is a European network established in 2019 to promote collaboration between discovery-oriented microbiome researchers and data-driven ML experts to optimize and standardize ML approaches for microbiome analysis. This perspective paper presents the key achievements of ML4Microbiome, which include identifying predictive and discriminatory 'omics' features, improving repeatability and comparability, developing automation procedures, and defining priority areas for the novel development of ML methods targeting the microbiome. The insights gained from ML4Microbiome will help to maximize the potential of ML in microbiome research and pave the way for new and improved healthcare practices.

Advancing human gut microbiota research by considering gut transit time

Accumulating evidence indicates that gut transit time is a key factor in shaping the gut microbiota composition and activity, which are linked to human health. Both population-wide and small-scale studies have identified transit time as a top covariate contributing to the large interindividual variation in the faecal microbiota composition. Despite this, transit time is still rarely being considered in the field of the human gut microbiome. Here, we review the latest research describing how and why whole gut and segmental transit times vary substantially between and within individuals, and how variations in gut transit time impact the gut microbiota composition, diversity and metabolism. Furthermore, we discuss the mechanisms by which the gut microbiota may causally affect gut motility. We argue that by taking into account the interindividual and intraindividual differences in gut transit time, we can advance our understanding of diet-microbiota interactions and disease-related microbiome signatures, since these may often be confounded by transient or persistent alterations in transit time. Altogether, a better understanding of the complex, bidirectional interactions between the gut microbiota and transit time is required to better understand gut microbiome variations in health and disease.

Just Add Data: automated predictive modeling for knowledge discovery and feature selection

Fully automated machine learning (AutoML) for predictive modeling is becoming a reality, giving rise to a whole new field. We present the basic ideas and principles of Just Add Data Bio (JADBio), an AutoML platform applicable to the low-sample, high-dimensional omics data that arise in translational medicine and bioinformatics applications. In addition to predictive and diagnostic models ready for clinical use, JADBio focuses on knowledge discovery by performing feature selection and identifying the corresponding biosignatures, i.e., minimal-size subsets of biomarkers that are jointly predictive of the outcome or phenotype of interest. It also returns a palette of useful information for interpretation, clinical use of the models, and decision making. JADBio is qualitatively and quantitatively compared against Hyper-Parameter Optimization Machine Learning libraries. Results show that in typical omics dataset analysis, JADBio manages to identify signatures comprising of just a handful of features while maintaining competitive predictive performance and accurate out-of-sample performance estimation.

Urine and Fecal 1H-NMR Metabolomes Differ Significantly between Pre-Term and Full-Term Born Physically Fit Healthy Adult Males

Preterm birth (before 37 weeks gestation) accounts for ~10% of births worldwide and remains one of the leading causes of death in children under 5 years of age. Preterm born adults have been consistently shown to be at an increased risk for chronic disorders including cardiovascular, endocrine/metabolic, respiratory, renal, neurologic, and psychiatric disorders that result in increased death risk. Oxidative stress was shown to be an important risk factor for hypertension, metabolic syndrome and lung disease (reduced pulmonary function, long-term obstructive pulmonary disease, respiratory infections, and sleep disturbances). The aim of this study was to explore the differences between preterm and full-term male participants' levels of urine and fecal proton nuclear magnetic resonance (¹H-NMR) metabolomes, during rest and exercise in normoxia and hypoxia and to assess general differences in human gut-microbiomes through metagenomics at the level of taxonomy, diversity, functional genes, enzymatic reactions, metabolic pathways and predicted gut metabolites. Significant differences existed between the two groups based on the analysis of ¹H-NMR urine and fecal metabolomes and their respective metabolic pathways, enabling the elucidation of a complex set of microbiome related metabolic biomarkers, supporting the idea of distinct host-microbiome interactions between the two groups and enabling the efficient classification of samples; however, this could not be directed to specific taxonomic characteristics.

Exercise and Interorgan Communication: Short-Term Exercise Training Blunts Differences in Consecutive Daily Urine 1H-NMR Metabolomic Signatures between Physically Active and Inactive Individuals

Physical inactivity is a worldwide health problem, an important risk for global mortality and is associated with chronic noncommunicable diseases. The aim of this study was to explore the differences in systemic urine ¹H-NMR metabolomes between physically active and inactive healthy young males enrolled in the X-Adapt project in response to controlled exercise (before and after the 3-day exercise testing and 10-day training protocol) in normoxic (21% O₂), normobaric (~1000 hPa) and normal-temperature (23 °C) conditions at 1 h of 50% maximal pedaling power output (W_peak) per day. Interrogation of the exercise database established from past X-Adapt results showed that significant multivariate differences existed in physiological traits between trained and untrained groups before and after training sessions and were mirrored in significant differences in urine pH, salinity, total dissolved solids and conductivity. Cholate, tartrate, cadaverine, lysine and N6-acetyllisine were the most important metabolites distinguishing trained and untrained groups. The relatively little effort of 1 h 50% W_peak per day invested by the untrained effectively modified their resting urine metabolome into one indistinguishable from the trained group, which hence provides a good basis for the planning of future recommendations for health maintenance in adults, irrespective of the starting fitness value. Finally, the 3-day sessions of morning urine samples represent a good candidate biological matrix for future delineations of active and inactive lifestyles detecting differences unobservable by single-day sampling due to day-to-day variability.

Liquid Biopsy in Type 2 Diabetes Mellitus Management: Building Specific Biosignatures via Machine Learning

Background: The need for minimally invasive biomarkers for the early diagnosis of type 2 diabetes (T2DM) prior to the clinical onset and monitoring of β-pancreatic cell loss is emerging. Here, we focused on studying circulating cell-free DNA (ccfDNA) as a liquid biopsy biomaterial for accurate diagnosis/monitoring of T2DM. Methods: ccfDNA levels were directly quantified in sera from 96 T2DM patients and 71 healthy individuals via fluorometry, and then fragment DNA size profiling was performed by capillary electrophoresis. Following this, ccfDNA methylation levels of five β-cell-related genes were measured via qPCR. Data were analyzed by automated machine learning to build classifying predictive models. Results: ccfDNA levels were found to be similar between groups but indicative of apoptosis in T2DM. INS (Insulin), IAPP (Islet Amyloid Polypeptide-Amylin), GCK (Glucokinase), and KCNJ11 (Potassium Inwardly Rectifying Channel Subfamily J member 11) levels differed significantly between groups. AutoML analysis delivered biosignatures including GCK, IAPP and KCNJ11 methylation, with the highest ever reported discriminating performance of T2DM from healthy individuals (AUC 0.927). Conclusions: Our data unravel the value of ccfDNA as a minimally invasive biomaterial carrying important clinical information for T2DM. Upon prospective clinical evaluation, the built biosignature can be disruptive for T2DM clinical management.

Individual variability in patterns and dynamics of fecal gluten immunogenic peptides excretion after low gluten intake

Purpose

Determination of Gluten Immunogenic Peptides (GIP) in feces is a direct tool for gluten exposure detection. The sensitivity of GIP detection methods for cases of unintentional low gluten intakes is unknown. We studied the interindividual variability in the kinetic of excretion under homogeneously controlled dietary conditions, and the sensitivity of fecal GIP tests after low amounts of punctual gluten ingestions.

Methods

Participants (n = 20) followed the same gluten-free menu for 12 days in which two separated doses of gluten (50 mg and 2 g) were ingested and all the depositions were collected. GIP from stool samples were analyzed by ELISA and lateral flow immunoassay (LFIA) tests.

Results

Most participants had detectable GIP after 50 mg and 2 g gluten ingestions using ELISA test (72.2% and 95%, respectively), whereas the LFIA test showed less sensitivity (22.2% and 80%, respectively). GIP were detected at higher either frequency or concentration in the range of 12–36 h after 50 mg intake, and 12–84 h after 2 g consumption. Considering this period, diagnostic sensitivity of GIP detection after a single 50 mg ingestion may be significatively increased analyzing three stool samples per individual. High variability among participants was found in the time and amount of GIP excretion; however, some individuals showed common patterns for both gluten intakes.

Conclusion

Sporadic gluten exposure detection may require several fecal samples to achieve level of sensitivity above 90%. Interindividual variability in the dynamic of GIP excretion may suggest patterns of gluten metabolism.