Machine Learning Applications for Mass Spectrometry-Based Metabolomics

Probe-based mass spectrometry approaches for single-cell and single-organelle measurements

Exploring the chemical content of individual cells not only reveals underlying cell-to-cell chemical heterogeneity but is also a key component in understanding how cells combine to form emergent properties of cellular networks and tissues. Recent technological advances in many analytical techniques including mass spectrometry (MS) have improved instrumental limits of detection and laser/ion probe dimensions, allowing the analysis of micron and submicron sized areas. In the case of MS, these improvements combined with MS's broad analyte detection capabilities have enabled the rise of single-cell and single-organelle chemical characterization. As the chemical coverage and throughput of single-cell measurements increase, more advanced statistical and data analysis methods have aided in data visualization and interpretation. This review focuses on secondary ion MS and matrix-assisted laser desorption/ionization MS approaches for single-cell and single-organelle characterization, which is followed by advances in mass spectral data visualization and analysis.

Automated Machine Learning and Explainable AI (AutoML-XAI) for Metabolomics: Improving Cancer Diagnostics

Metabolomics generates complex data necessitating advanced computational methods for generating biological insight. While machine learning (ML) is promising, the challenges of selecting the best algorithms and tuning hyperparameters, particularly for nonexperts, remain. Automated machine learning (AutoML) can streamline this process; however, the issue of interpretability could persist. This research introduces a unified pipeline that combines AutoML with explainable AI (XAI) techniques to optimize metabolomics analysis. We tested our approach on two data sets: renal cell carcinoma (RCC) urine metabolomics and ovarian cancer (OC) serum metabolomics. AutoML, using Auto-sklearn, surpassed standalone ML algorithms like SVM and k-Nearest Neighbors in differentiating between RCC and healthy controls, as well as OC patients and those with other gynecological cancers. The effectiveness of Auto-sklearn is highlighted by its AUC scores of 0.97 for RCC and 0.85 for OC, obtained from the unseen test sets. Importantly, on most of the metrics considered, Auto-sklearn demonstrated a better classification performance, leveraging a mix of algorithms and ensemble techniques. Shapley Additive Explanations (SHAP) provided a global ranking of feature importance, identifying dibutylamine and ganglioside GM(d34:1) as the top discriminative metabolites for RCC and OC, respectively. Waterfall plots offered local explanations by illustrating the influence of each metabolite on individual predictions. Dependence plots spotlighted metabolite interactions, such as the connection between hippuric acid and one of its derivatives in RCC, and between GM3(d34:1) and GM3(18:1_16:0) in OC, hinting at potential mechanistic relationships. Through decision plots, a detailed error analysis was conducted, contrasting feature importance for correctly versus incorrectly classified samples. In essence, our pipeline emphasizes the importance of harmonizing AutoML and XAI, facilitating both simplified ML application and improved interpretability in metabolomics data science.

Python workflow for the selection and identification of marker peptides-proof-of-principle study with heated milk

The analysis of almost holistic food profiles has developed considerably over the last years. This has also led to larger amounts of data and the ability to obtain more information about health-beneficial and adverse constituents in food than ever before. Especially in the field of proteomics, software is used for evaluation, and these do not provide specific approaches for unique monitoring questions. An additional and more comprehensive way of evaluation can be done with the programming language Python. It offers broad possibilities by a large ecosystem for mass spectrometric data analysis, but needs to be tailored for specific sets of features, the research questions behind. It also offers the applicability of various machine-learning approaches. The aim of the present study was to develop an algorithm for selecting and identifying potential marker peptides from mass spectrometric data. The workflow is divided into three steps: (I) feature engineering, (II) chemometric data analysis, and (III) feature identification. The first step is the transformation of the mass spectrometric data into a structure, which enables the application of existing data analysis packages in Python. The second step is the data analysis for selecting single features. These features are further processed in the third step, which is the feature identification. The data used exemplarily in this proof-of-principle approach was from a study on the influence of a heat treatment on the milk proteome/peptidome.

Exploring machine learning for untargeted metabolomics using molecular fingerprints

BACKGROUND

Metabolomics, the study of substrates and products of cellular metabolism, offers valuable insights into an organism's state under specific conditions and has the potential to revolutionise preventive healthcare and pharmaceutical research. However, analysing large metabolomics datasets remains challenging, with available methods relying on limited and incompletely annotated metabolic pathways.

METHODS

This study, inspired by well-established methods in drug discovery, employs machine learning on metabolite fingerprints to explore the relationship of their structure with responses in experimental conditions beyond known pathways, shedding light on metabolic processes. It evaluates fingerprinting effectiveness in representing metabolites, addressing challenges like class imbalance, data sparsity, high dimensionality, duplicate structural encoding, and interpretable features. Feature importance analysis is then applied to reveal key chemical configurations affecting classification, identifying related metabolite groups.

RESULTS

The approach is tested on two datasets: one on Ataxia Telangiectasia and another on endothelial cells under low oxygen. Machine learning on molecular fingerprints predicts metabolite responses effectively, and feature importance analysis aligns with known metabolic pathways, unveiling new affected metabolite groups for further study.

CONCLUSION

In conclusion, the presented approach leverages the strengths of drug discovery to address critical issues in metabolomics research and aims to bridge the gap between these two disciplines. This work lays the foundation for future research in this direction, possibly exploring alternative structural encodings and machine learning models.

Data Acquisition and Intraoperative Tissue Analysis on a Mobile, Battery-Operated, Orbitrap Mass Spectrometer

Mass spectrometry has been increasingly explored in intraoperative studies as a potential technology to help guide surgical decision making. Yet, intraoperative experiments using high-performance mass spectrometry instrumentation present a unique set of operational challenges. For example, standard operating rooms are often not equipped with the electrical requirements to power a commercial mass spectrometer and are not designed to accommodate their permanent installation. These obstacles can impact progress and patient enrollment in intraoperative clinical studies because implementation of MS instrumentation becomes limited to specific operating rooms that have the required electrical connections and space. To expand our intraoperative clinical studies using the MasSpec Pen technology, we explored the feasibility of transporting and acquiring data on Orbitrap mass spectrometers operating on battery power in hospital buildings. We evaluated the effect of instrument movement including acceleration and rotational speeds on signal stability and mass accuracy by acquiring data using direct infusion electrospray ionization. Data were acquired while rolling the systems in/out of operating rooms and while descending/ascending a freight elevator. Despite these movements and operating the instrument on battery power, the relative standard deviation of the total ion current was <5% and the magnitude of the mass error relative to the internal calibrant never exceeded 5.06 ppm. We further evaluated the feasibility of performing intraoperative MasSpec Pen analysis while operating the Orbitrap mass spectrometer on battery power during an ovarian cancer surgery. We observed that the rich and tissue-specific molecular profile commonly detected from ovarian tissues was conserved when running on battery power. Together, these results demonstrate that Orbitrap mass spectrometers can be operated and acquire data on battery power while in motion and in rotation without losses in signal stability or mass accuracy. Furthermore, Orbitrap mass spectrometers can be used in conjunction to the MasSpec Pen while on battery power for intraoperative tissue analysis.

Advances of machine learning-assisted small extracellular vesicles detection strategy

Detection of extracellular vesicles (EVs), particularly small EVs (sEVs), is of great significance in exploring their physiological characteristics and clinical applications. The heterogeneity of sEVs plays a crucial role in distinguishing different types of cells and diseases. Machine learning, with its exceptional data processing capabilities, offers a solution to overcome the limitations of conventional detection methods for accurately classifying sEV subtypes and sources. Principal component analysis, linear discriminant analysis, partial least squares discriminant analysis, XGBoost, support vector machine, k-nearest neighbor, and deep learning, along with some combined methods such as principal component-linear discriminant analysis, have been successfully applied in the detection and identification of sEVs. This review focuses on machine learning-assisted detection strategies for cell identification and disease prediction via sEVs, and summarizes the integration of these strategies with surface-enhanced Raman scattering, electrochemistry, inductively coupled plasma mass spectrometry and fluorescence. The performance of different machine learning-based detection strategies is compared, and the advantages and limitations of various machine learning models are also evaluated. Finally, we discuss the merits and limitations of the current approaches and briefly outline the perspective of potential research directions in the field of sEV analysis based on machine learning.

COVID-19 and Comorbidities: What Has Been Unveiled by Metabolomics?

The COVID-19 pandemic has brought about diverse impacts on the global population. Individuals with comorbidities were more susceptible to the severe symptoms caused by the virus. Within the crisis scenario, metabolomics represents a potential area of science capable of providing relevant information for understanding the metabolic pathways associated with the intricate interaction between the viral disease and previous comorbidities. This work aims to provide a comprehensive description of the scientific production pertaining to metabolomics within the specific context of COVID-19 and comorbidities, while highlighting promising areas for exploration by those interested in the subject. In this review, we highlighted the studies of metabolomics that indicated a variety of metabolites associated with comorbidities and COVID-19. Furthermore, we observed that the understanding of the metabolic processes involved between comorbidities and COVID-19 is limited due to the urgent need to report disease outcomes in individuals with comorbidities. The overlap of two or more comorbidities associated with the severity of COVID-19 hinders the comprehension of the significance of each condition. Most identified studies are observational, with a restricted number of patients, due to challenges in sample collection amidst the emergent situation.

Protective effect of FKBP12 on dextran sulfate sodium-induced ulcerative colitis in mice as a tacrolimus receptor

Ulcerative colitis (UC) is a multifactorial intestinal disease with a high incidence. In recent years, there has been an urgent need for pleiotropic drugs with a clear biosafety profile. Tacrolimus (TAC) is an immunosuppressant with stronger in vivo effects and better gastrointestinal absorption and is considered a potential treatment for UC. FKBP12 is a mediator of TAC immunosuppression; however, it is unclear whether it can participate in the development of UC in combination with TAC. The purpose of this study is to preliminarily validate the function of FKBP12 by establishing dextran sulfate sodium (DSS)-induced UC model and TAC treatment. The results revealed that TAC was effective in alleviating DSS-induced UC symptoms such as body weight and disease activity index (DAI). TAC significantly protects colonic tissue and attenuates DSS-induced histomorphological changes. In addition, FKBP12 is down-regulated in the intestinal tissue of DSS-induced UC mice and in serum samples of UC patients. In conclusion, our study revealed that FKBP12 may act as a TAC receptor to have anti-inflammatory and protective effects on DSS-induced UC in mice, which will provide a new option for the treatment of UC.

Metabolomics analysis reveals novel serum metabolite alterations in cancer cachexia

Background

Cachexia is a body wasting syndrome that significantly affects well-being and prognosis of cancer patients, without effective treatment. Serum metabolites take part in pathophysiological processes of cancer cachexia, but apart from altered levels of select serum metabolites, little is known on the global changes of the overall serum metabolome, which represents a functional readout of the whole-body metabolic state. Here, we aimed to comprehensively characterize serum metabolite alterations and analyze associated pathways in cachectic cancer patients to gain new insights that could help instruct strategies for novel interventions of greater clinical benefit.

Methods

Serum was sampled from 120 metastatic cancer patients (stage UICC IV). Patients were grouped as cachectic or non-cachectic according to the criteria for cancer cachexia agreed upon international consensus (main criterium: weight loss adjusted to body mass index). Samples were pooled by cachexia phenotype and assayed using non-targeted gas chromatography-mass spectrometry (GC-MS). Normalized metabolite levels were compared using t-test (p < 0.05, adjusted for false discovery rate) and partial least squares discriminant analysis (PLS-DA). Machine-learning models were applied to identify metabolite signatures for separating cachexia states. Significant metabolites underwent MetaboAnalyst 5.0 pathway analysis.

Results

Comparative analyses included 78 cachectic and 42 non-cachectic patients. Cachectic patients exhibited 19 annotable, significantly elevated (including glucose and fructose) or decreased (mostly amino acids) metabolites associating with aminoacyl-tRNA, glutathione and amino acid metabolism pathways. PLS-DA showed distinct clusters (accuracy: 85.6%), and machine-learning models identified metabolic signatures for separating cachectic states (accuracy: 83.2%; area under ROC: 88.0%). We newly identified altered blood levels of erythronic acid and glucuronic acid in human cancer cachexia, potentially linked to pentose-phosphate and detoxification pathways.

Conclusion

We found both known and yet unknown serum metabolite and metabolic pathway alterations in cachectic cancer patients that collectively support a whole-body metabolic state with impaired detoxification capability, altered glucose and fructose metabolism, and substrate supply for increased and/or distinct metabolic needs of cachexia-associated tumors. These findings together imply vulnerabilities, dependencies and targets for novel interventions that have potential to make a significant impact on future research in an important field of cancer patient care.

Metabolomics-based development of bioproduction processes toward industrial-scale production

Microbial biomanufacturing offers a promising, environment-friendly platform for next-generation chemical production. However, its limited industrial implementation is attributed to the slow production rates of target compounds and the time-intensive engineering of high-yield strains. This review highlights how metabolomics expedites bioproduction development, as demonstrated through case studies of its integration into microbial strain engineering, culture optimization, and model construction. The Design-Build-Test-Learn (DBTL) cycle serves as a standard workflow for strain engineering. Process development, including the optimization of culture conditions and scale-up, is crucial for industrial production. In silico models facilitate the development of strains and processes. Metabolomics is a powerful driver of the DBTL framework, process development, and model construction.

Data-Driven Compound Identification in Atmospheric Mass Spectrometry

Aerosol particles found in the atmosphere affect the climate and worsen air quality. To mitigate these adverse impacts, aerosol particle formation and aerosol chemistry in the atmosphere need to be better mapped out and understood. Currently, mass spectrometry is the single most important analytical technique in atmospheric chemistry and is used to track and identify compounds and processes. Large amounts of data are collected in each measurement of current time-of-flight and orbitrap mass spectrometers using modern rapid data acquisition practices. However, compound identification remains a major bottleneck during data analysis due to lacking reference libraries and analysis tools. Data-driven compound identification approaches could alleviate the problem, yet remain rare to non-existent in atmospheric science. In this perspective, the authors review the current state of data-driven compound identification with mass spectrometry in atmospheric science and discuss current challenges and possible future steps toward a digital era for atmospheric mass spectrometry.

Multiple and Optimal Screening Subset: a method selecting global characteristic congeners for robust foodomics analysis

Metabolomics and foodomics shed light on the molecular processes within living organisms and the complex food composition by leveraging sophisticated analytical techniques to systematically analyze the vast array of molecular features. The traditional feature-picking method often results in arbitrary selections of the model, feature ranking, and cut-off, which may lead to suboptimal results. Thus, a Multiple and Optimal Screening Subset (MOSS) approach was developed in this study to achieve a balance between a minimal number of predictors and high predictive accuracy during statistical model setup. The MOSS approach compares five commonly used models in the context of food matrix analysis, specifically bourbons. These models include Student's t-test, receiver operating characteristic curve, partial least squares-discriminant analysis (PLS-DA), random forests, and support vector machines. The approach employs cross-validation to identify promising subset feature candidates that contribute to food characteristic classification. It then determines the optimal subset size by comparing it to the corresponding top-ranked features. Finally, it selects the optimal feature subset by traversing all possible feature candidate combinations. By utilizing MOSS approach to analyze 1406 mass spectral features from a collection of 122 bourbon samples, we were able to generate a subset of features for bourbon age prediction with 88% accuracy. Additionally, MOSS increased the area under the curve performance of sweetness prediction to 0.898 with only four predictors compared with the top-ranked four features at 0.681 based on the PLS-DA model. Overall, we demonstrated that MOSS provides an efficient and effective approach for selecting optimal features compared with other frequently utilized methods.

Automated Structural Activity Screening of β-Diketonate Assemblies with High-Throughput Ion Mobility-Mass Spectrometry

Embracing complexity in design, metallo-supramolecular self-assembly presents an opportunity for fabricating materials of economic significance. The array of accessible supramolecules is alluring, along with favourable energy requirements. Implementation is hampered by an inability to efficiently characterise complex mixtures. The stoichiometry, size, shape, guest binding properties and reactivity of individual components and combinations thereof are inherently challenging to resolve. A large combinatorial library of four transition metals (Fe, Cu, Ni and Zn), and six β-diketonate ligands at different molar ratios and pH was robotically prepared and directly analysed over multiple timepoints with electrospray ionisation travelling wave ion mobility-mass spectrometry. The dataset was parsed for self-assembling activity without first attempting to structurally assign individual species. Self-assembling systems were readily categorised without manual data-handling, allowing efficient screening of self-assembly activity. This workflow clarifies solution phase supramolecular assembly processes without manual, bottom-up processing. The complex behaviour of the self-assembling systems was reduced to simpler qualities, which could be automatically processed.

Progress and challenges in exploring aquatic microbial communities using non-targeted metabolomics

Advances in bioanalytical technologies are constantly expanding our insights into complex ecosystems. Here, we highlight strategies and applications that make use of non-targeted metabolomics methods in aquatic chemical ecology research and discuss opportunities and remaining challenges of mass spectrometry-based methods to broaden our understanding of environmental systems.

Opportunities and challenges for sample preparation and enrichment in mass spectrometry for single-cell metabolomics

Single-cell heterogeneity in metabolism, drug resistance and disease type poses the need for analytical techniques for single-cell analysis. As the metabolome provides the closest view of the status quo in the cell, studying the metabolome at single-cell resolution may unravel said heterogeneity. A challenge in single-cell metabolome analysis is that metabolites cannot be amplified, so one needs to deal with picolitre volumes and a wide range of analyte concentrations. Due to high sensitivity and resolution, MS is preferred in single-cell metabolomics. Large numbers of cells need to be analysed for proper statistics; this requires high-throughput analysis, and hence automation of the analytical workflow. Significant advances in (micro)sampling methods, CE and ion mobility spectrometry have been made, some of which have been applied in high-throughput analyses. Microfluidics has enabled an automation of cell picking and metabolite extraction; image recognition has enabled automated cell identification. Many techniques have been used for data analysis, varying from conventional techniques to novel combinations of advanced chemometric approaches. Steps have been set in making data more findable, accessible, interoperable and reusable, but significant opportunities for improvement remain. Herein, advances in single-cell analysis workflows and data analysis are discussed, and recommendations are made based on the experimental goal.

Insufficient sleep and weekend recovery sleep: classification by a metabolomics-based machine learning ensemble

Although weekend recovery sleep is common, the physiological responses to weekend recovery sleep are not fully elucidated. Identifying molecular biomarkers that represent adequate versus insufficient sleep could help advance our understanding of weekend recovery sleep. Here, we identified potential molecular biomarkers of insufficient sleep and defined the impact of weekend recovery sleep on these biomarkers using metabolomics in a randomized controlled trial. Healthy adults (n = 34) were randomized into three groups: control (CON: 9-h sleep opportunities); sleep restriction (SR: 5-h sleep opportunities); or weekend recovery (WR: simulated workweek of 5-h sleep opportunities followed by ad libitum weekend recovery sleep and then 2 days with 5-h sleep opportunities). Blood for metabolomics was collected on the simulated Monday immediately following the weekend. Nine machine learning models, including a machine learning ensemble, were built to classify samples from SR versus CON. Notably, SR showed decreased glycerophospholipids and sphingolipids versus CON. The machine learning ensemble showed the highest G-mean performance and classified 50% of the WR samples as insufficient sleep. Our findings show insufficient sleep and recovery sleep influence the plasma metabolome and suggest more than one weekend of recovery sleep may be necessary for the identified biomarkers to return to healthy adequate sleep levels.

Prediction of spontaneous preterm birth using supervised machine learning on metabolomic data: A case-cohort study

OBJECTIVES

To identify and internally validate metabolites predictive of spontaneous preterm birth (sPTB) using multiple machine learning methods and sequential maternal serum samples, and to predict spontaneous early term birth (sETB) using these metabolites.

DESIGN

Case-cohort design within a prospective cohort study.

SETTING

Cambridge, UK.

POPULATION OR SAMPLE

A total of 399 Pregnancy Outcome Prediction study participants, including 98 cases of sPTB.

METHODS

An untargeted metabolomic analysis of maternal serum samples at 12, 20, 28 and 36 weeks of gestation was performed. We applied six supervised machine learning methods and a weighted Cox model to measurements at 28 weeks of gestation and sPTB, followed by feature selection. We used logistic regression with elastic net penalty, followed by best subset selection, to reduce the number of predictive metabolites further. We applied coefficients from the chosen models to measurements from different gestational ages to predict sPTB and sETB.

MAIN OUTCOME MEASURES

sPTB and sETB.

RESULTS

We identified 47 metabolites, mostly lipids, as important predictors of sPTB by two or more methods and 22 were identified by three or more methods. The best 4-predictor model had an optimism-corrected area under the receiver operating characteristics curve (AUC) of 0.703 at 28 weeks of gestation. The model also predicted sPTB in 12-week samples (0.606, 95% CI 0.544-0.667) and 20-week samples (0.657, 95% CI 0.597-0.717) and it predicted sETB in 36-week samples (0.727, 95% CI 0.606-0.849). A lysolipid, 1-palmitoleoyl-GPE (16:1)*, was the strongest predictor of sPTB at 12 weeks of gestation (0.609, 95% CI 0.548-0.670), 20 weeks (0.630, 95% CI 0.569-0.690) and 28 weeks (0.660, 95% CI 0.599-0.722), and of sETB at 36 weeks (0.739, 95% CI 0.618-0.860).

CONCLUSIONS

We identified and internally validated maternal serum metabolites predictive of sPTB. A lysolipid, 1-palmitoleoyl-GPE (16:1)*, is a novel predictor of sPTB and sETB. Further validation in external populations is required.

Sex differences in associations of plasma metabolites with blood pressure and heart rate variability: The HELIUS study

BACKGROUND AND AIMS

Since plasma metabolites can modulate blood pressure (BP) and vary between men and women, we examined sex differences in plasma metabolite profiles associated with BP and sympathicovagal balance. Our secondary aim was to investigate associations between gut microbiota composition and plasma metabolites predictive of BP and heart rate variability (HRV).

METHODS

From the HELIUS cohort, we included 196 women and 173 men. Office systolic BP and diastolic BP were recorded, and heart rate variability (HRV) and baroreceptor sensitivity (BRS) were calculated using finger photoplethysmography. Plasma metabolomics was measured using untargeted LC-MS/MS. Gut microbiota composition was determined using 16S sequencing. We used machine learning models to predict BP and HRV from metabolite profiles, and to predict metabolite levels from gut microbiota composition.

RESULTS

In women, best predicting metabolites for systolic BP included dihomo-lineoylcarnitine, 4-hydroxyphenylacetateglutamine and vanillactate. In men, top predictors included sphingomyelins, N-formylmethionine and conjugated bile acids. Best predictors for HRV in men included phenylacetate and gentisate, which were associated with lower HRV in men but not in women. Several of these metabolites were associated with gut microbiota composition, including phenylacetate, multiple sphingomyelins and gentisate.

CONCLUSIONS

Plasma metabolite profiles are associated with BP in a sex-specific manner. Catecholamine derivatives were more important predictors for BP in women, while sphingomyelins were more important in men. Several metabolites were associated with gut microbiota composition, providing potential targets for intervention.

Recent advances in LC-MS-based metabolomics for clinical biomarker discovery

The employment of liquid chromatography-mass spectrometry (LC-MS) untargeted and targeted metabolomics has led to the discovery of novel biomarkers and improved the understanding of various disease mechanisms. Numerous strategies have been reported to expand the metabolite coverage in LC-MS-untargeted and targeted metabolomics. To improve the sensitivity of low-abundance or poor-ionized metabolites for reducing the amount of clinical sample, chemical derivatization methods are used to target different functional groups. Proper sample preparation is beneficial for reducing the matrix effect, maintaining the stability of the LC-MS system, and increasing the metabolite coverage. Machine learning has recently been integrated into the workflow of LC-MS metabolomics to accelerate metabolite identification and data-processing automation, and increase the accuracy of disease classification and clinical outcome prediction. Due to the rapidly growing utility of LC-MS metabolomics in discovering disease markers, this review will address the recent advances in the field and offer perspectives on various strategies for expanding metabolite coverage, chemical derivatization, sample preparation, clinical disease markers, and machining learning for disease modeling.

Automated machine learning and explainable AI (AutoML-XAI) for metabolomics: improving cancer diagnostics

Motivation

Metabolomics generates complex data necessitating advanced computational methods for generating biological insight. While machine learning (ML) is promising, the challenges of selecting the best algorithms and tuning hyperparameters, particularly for non-experts, remain. Automated machine learning (AutoML) can streamline this process; however, the issue of interpretability could persist. This research introduces a unified pipeline that combines AutoML with explainable AI (XAI) techniques to optimize metabolomics analysis.

Results

We tested our approach on two datasets: renal cell carcinoma (RCC) urine metabolomics and ovarian cancer (OC) serum metabolomics. AutoML, using auto-sklearn, surpassed standalone ML algorithms such as SVM and random forest in differentiating between RCC and healthy controls, as well as OC patients and those with other gynecological cancers (Non-OC). Auto-sklearn employed a mix of algorithms and ensemble techniques, yielding a superior performance (AUC of 0.97 for RCC and 0.85 for OC). Shapley Additive Explanations (SHAP) provided a global ranking of feature importance, identifying dibutylamine and ganglioside GM(d34:1) as the top discriminative metabolites for RCC and OC, respectively. Waterfall plots offered local explanations by illustrating the influence of each metabolite on individual predictions. Dependence plots spotlighted metabolite interactions, such as the connection between hippuric acid and one of its derivatives in RCC, and between GM3(d34:1) and GM3(18:1_16:0) in OC, hinting at potential mechanistic relationships. Through decision plots, a detailed error analysis was conducted, contrasting feature importance for correctly versus incorrectly classified samples. In essence, our pipeline emphasizes the importance of harmonizing AutoML and XAI, facilitating both simplified ML application and improved interpretability in metabolomics data science.

Availability

https://github.com/obifarin/automl-xai-metabolomics.

Metabolomic and Lipidomic Characterization of Meningioma Grades Using LC-HRMS and Machine Learning

Meningiomas are among the most common brain tumors that arise from the leptomeningeal cover of the brain and spinal cord and account for around 37% of all central nervous system tumors. According to the World Health Organization, meningiomas are classified into three histological subtypes: benign, atypical, and anaplastic. Sometimes, meningiomas with a histological diagnosis of benign tumors show clinical characteristics and behavior of aggressive tumors. In this study, we examined the metabolomic and lipidomic profiles of meningioma tumors, focusing on comparing low-grade and high-grade tumors and identifying potential markers that can discriminate between benign and malignant tumors. High-resolution mass spectrometry coupled to liquid chromatography was used for untargeted metabolomics and lipidomics analyses of 85 tumor biopsy samples with different meningioma grades. We then applied feature selection and machine learning techniques to find the features with the highest information to aid in the diagnosis of meningioma grades. Three biomarkers were identified to differentiate low- and high-grade meningioma brain tumors. The use of mass-spectrometry-based metabolomics and lipidomics combined with machine learning analyses to prospect and characterize biomarkers associated with meningioma grades may pave the way for elucidating potential therapeutic and prognostic targets.

Mass-Suite: a novel open-source python package for high-resolution mass spectrometry data analysis

Mass-Suite (MSS) is a Python-based, open-source software package designed to analyze high-resolution mass spectrometry (HRMS)-based non-targeted analysis (NTA) data, particularly for water quality assessment and other environmental applications. MSS provides flexible, user-defined workflows for HRMS data processing and analysis, including both basic functions (e.g., feature extraction, data reduction, feature annotation, data visualization, and statistical analyses) and advanced exploratory data mining and predictive modeling capabilities that are not provided by currently available open-source software (e.g., unsupervised clustering analyses, a machine learning-based source tracking and apportionment tool). As a key advance, most core MSS functions are supported by machine learning algorithms (e.g., clustering algorithms and predictive modeling algorithms) to facilitate function accuracy and/or efficiency. MSS reliability was validated with mixed chemical standards of known composition, with 99.5% feature extraction accuracy and ~ 52% overlap of extracted features relative to other open-source software tools. Example user cases of laboratory data evaluation are provided to illustrate MSS functionalities and demonstrate reliability. MSS expands available HRMS data analysis workflows for water quality evaluation and environmental forensics, and is readily integrated with existing capabilities. As an open-source package, we anticipate further development of improved data analysis capabilities in collaboration with interested users.

Non-Targeted Metabolomic Analysis of Arabidopsis thaliana (L.) Heynh: Metabolic Adaptive Responses to Stress Caused by N Starvation

As sessile organisms, plants develop the ability to respond and survive in changing environments. Such adaptive responses maximize phenotypic and metabolic fitness, allowing plants to adjust their growth and development. In this study, we analyzed the metabolic plasticity of Arabidopsis thaliana in response to nitrate deprivation by untargeted metabolomic analysis and using wild-type (WT) genotypes and the loss-of-function nia1/nia2 double mutant. Secondary metabolites were identified using seedlings grown on a hydroponic system supplemented with optimal or limiting concentrations of N (4 or 0.2 mM, respectively) and harvested at 15 and 30 days of age. Then, spectral libraries generated from shoots and roots in both ionization modes (ESI +/-) were compared. Totals of 3407 and 4521 spectral signals (m/z_rt) were obtained in the ESI+ and ESI- modes, respectively. Of these, approximately 50 and 65% were identified as differentially synthetized/accumulated. This led to the presumptive identification of 735 KEGG codes (metabolites) belonging to 79 metabolic pathways. The metabolic responses in the shoots and roots of WT genotypes at 4 mM of N favor the synthesis/accumulation of metabolites strongly related to growth. In contrast, for the nia1/nia2 double mutant (similar as the WT genotype at 0.2 mM N), metabolites identified as differentially synthetized/accumulated help cope with stress, regulating oxidative stress and preventing programmed cell death, meaning that metabolic responses under N starvation compromise growth to prioritize a defensive response.

De Novo Cleaning of Chimeric MS/MS Spectra for LC-MS/MS-Based Metabolomics

The purity of tandem mass spectrometry (MS/MS) is essential to MS/MS-based metabolite annotation and unknown exploration. This work presents a de novo approach to cleaning chimeric MS/MS spectra generated in liquid chromatography-tandem mass spectrometry (LC-MS/MS)-based metabolomics. The assumption is that true fragments and their precursors are well correlated across the samples in a study, while false or contamination fragments are rather independent. Using data simulation, this work starts with an investigation of the negative effects of chimeric MS/MS spectra on spectral similarity analysis and molecular networking. Next, the characteristics of true and false fragments in chimeric MS/MS spectra were investigated using MS/MS of chemical standards. We recognized three fragment peak attributes indicative of whether a peak is a false fragment, including (1) intensity ratio fluctuation, (2) appearance rate, and (3) relative intensity. Using these attributes, we tested three machine learning models and identified XGBoost as the best model to achieve an area under the precision-recall curve of 0.98 for a clear separation between true and false fragments. Based on the trained model, we constructed an automated bioinformatic platform, DNMS2Purifier (short for de novo MS2Purifier), for metabolic features from metabolomics studies. DNMS2Purifier recognizes and processes chimeric MS/MS spectra without additional sample analysis or library confirmation. DNMS2Purifer was evaluated on a metabolomics data set generated with different MS/MS precursor isolation windows. It successfully captured the increase in the number of false fragments from the increased isolation window. DNMS2Purifier was also compared to MS2Purifier, an existing MS/MS spectral cleaning tool based on the addition of data-independent acquisition (DIA) analysis. Results indicated that DNMS2Purifier uniquely recognizes false fragments, which complements the previous DIA-based approach. Finally, DNMS2Purifier was demonstrated using a real experimental metabolomics study, showing improved MS/MS spectral quality and leading to an improved spectral match ratio and molecular networking outcome.

Recent advances in data- and knowledge-driven approaches to explore primary microbial metabolism

With the rapid progress in metabolomics and sequencing technologies, more data on the metabolome of single microbes and their communities become available, revealing the potential of microorganisms to metabolize a broad range of chemical compounds. The analysis of microbial metabolomics datasets remains challenging since it inherits the technical challenges of metabolomics analysis, such as compound identification and annotation, while harboring challenges in data interpretation, such as distinguishing metabolite sources in mixed samples. This review outlines the recent advances in computational methods to analyze primary microbial metabolism: knowledge-based approaches that take advantage of metabolic and molecular networks and data-driven approaches that employ machine/deep learning algorithms in combination with large-scale datasets. These methods aim at improving metabolite identification and disentangling reciprocal interactions between microbes and metabolites. We also discuss the perspective of combining these approaches and further developments required to advance the investigation of primary metabolism in mixed microbial samples.

Combined Metabolipidomic and Machine Learning Approach in a Rat Model of Stroke Reveals a Deleterious Impact of Brain Injury on Heart Metabolism

Cardiac complications are frequently found following a stroke in humans whose pathophysiological mechanism remains poorly understood. We used machine learning to analyse a large set of data from a metabolipidomic study assaying 630 metabolites in a rat stroke model to investigate metabolic changes affecting the heart within 72 h after a stroke. Twelve rats undergoing a stroke and 28 rats undergoing the sham procedure were investigated. A plasmatic signature consistent with the literature with notable lipid metabolism remodelling was identified. The post-stroke heart showed a discriminant metabolic signature, in comparison to the sham controls, involving increased collagen turnover, increased arginase activity with decreased nitric oxide synthase activity as well as an altered amino acid metabolism (including serine, asparagine, lysine and glycine). In conclusion, these results demonstrate that brain injury induces a metabolic remodelling in the heart potentially involved in the pathophysiology of stroke heart syndrome.

Atomic structures, conformers and thermodynamic properties of 32k atmospheric molecules

Low-volatile organic compounds (LVOCs) drive key atmospheric processes, such as new particle formation (NPF) and growth. Machine learning tools can accelerate studies of these phenomena, but extensive and versatile LVOC datasets relevant for the atmospheric research community are lacking. We present the GeckoQ dataset with atomic structures of 31,637 atmospherically relevant molecules resulting from the oxidation of α-pinene, toluene and decane. For each molecule, we performed comprehensive conformer sampling with the COSMOconf program and calculated thermodynamic properties with density functional theory (DFT) using the Conductor-like Screening Model (COSMO). Our dataset contains the geometries of the 7 Mio. conformers we found and their corresponding structural and thermodynamic properties, including saturation vapor pressures (p_Sat), chemical potentials and free energies. The p_Sat were compared to values calculated with the group contribution method SIMPOL. To validate the dataset, we explored the relationship between structural and thermodynamic properties, and then demonstrated a first machine-learning application with Gaussian process regression.

Empowering drug off-target discovery with metabolic and structural analysis

Elucidating intracellular drug targets is a difficult problem. While machine learning analysis of omics data has been a promising approach, going from large-scale trends to specific targets remains a challenge. Here, we develop a hierarchic workflow to focus on specific targets based on analysis of metabolomics data and growth rescue experiments. We deploy this framework to understand the intracellular molecular interactions of the multi-valent dihydrofolate reductase-targeting antibiotic compound CD15-3. We analyse global metabolomics data utilizing machine learning, metabolic modelling, and protein structural similarity to prioritize candidate drug targets. Overexpression and in vitro activity assays confirm one of the predicted candidates, HPPK (folK), as a CD15-3 off-target. This study demonstrates how established machine learning methods can be combined with mechanistic analyses to improve the resolution of drug target finding workflows for discovering off-targets of a metabolic inhibitor.

Predicting the retention time of Synthetic Cannabinoids using a combinatorial QSAR approach

Background

Abuse of Synthetic Cannabinoids (SCs) has become a serious threat to public health. Due to the various structural and chemical group modified by criminals, their detection is a major challenge in forensic toxicological identification. Therefore, rapid and efficient identification of SCs is important for forensic toxicology and drug bans. The prediction of an analyte's retention time in liquid chromatography is an important index for the qualitative analysis of compounds and can provide informatics solutions for the interpretation of chromatographic data.

Methods

In this study, experimental data from high-resolution mass spectrometry (HRMS) are used to construct a regression model for predicting the retention time of SCs using machine learning methods. The prediction ability of the model is improved by adopting a strategy that combines different descriptors in different independent machine-learning methods.

Results

The best model was obtained with a method that combined Substructure Fingerprint Count and Finger printer features and the support vector regression (SVR) method, as it exhibited an R² value of 0.81 for the validation set and 0.83 for the test set. In addition, 4 new SCs were predicted by the optimized model, with a prediction error within 3%.

Conclusions

Our study provides a model that can predict the retention time of compounds and it can be used as a filter to reduce false-positive candidates when used in combination with LC-HRMS, especially in the absence of reference standards. This can improve the confidence of identification in non-targeted analysis and the reliability of identifying unknown substances.

Interpretable Machine Learning on Metabolomics Data Reveals Biomarkers for Parkinson's Disease

The use of machine learning (ML) with metabolomics provides opportunities for the early diagnosis of disease. However, the accuracy of ML and extent of information obtained from metabolomics can be limited owing to challenges associated with interpreting disease prediction models and analyzing many chemical features with abundances that are correlated and "noisy". Here, we report an interpretable neural network (NN) framework to accurately predict disease and identify significant biomarkers using whole metabolomics data sets without a priori feature selection. The performance of the NN approach for predicting Parkinson's disease (PD) from blood plasma metabolomics data is significantly higher than other ML methods with a mean area under the curve of >0.995. PD-specific markers that predate clinical PD diagnosis and contribute significantly to early disease prediction were identified including an exogenous polyfluoroalkyl substance. It is anticipated that this accurate and interpretable NN-based approach can improve diagnostic performance for many diseases using metabolomics and other untargeted 'omics methods.

PeakDecoder enables machine learning-based metabolite annotation and accurate profiling in multidimensional mass spectrometry measurements

Multidimensional measurements using state-of-the-art separations and mass spectrometry provide advantages in untargeted metabolomics analyses for studying biological and environmental bio-chemical processes. However, the lack of rapid analytical methods and robust algorithms for these heterogeneous data has limited its application. Here, we develop and evaluate a sensitive and high-throughput analytical and computational workflow to enable accurate metabolite profiling. Our workflow combines liquid chromatography, ion mobility spectrometry and data-independent acquisition mass spectrometry with PeakDecoder, a machine learning-based algorithm that learns to distinguish true co-elution and co-mobility from raw data and calculates metabolite identification error rates. We apply PeakDecoder for metabolite profiling of various engineered strains of Aspergillus pseudoterreus, Aspergillus niger, Pseudomonas putida and Rhodosporidium toruloides. Results, validated manually and against selected reaction monitoring and gas-chromatography platforms, show that 2683 features could be confidently annotated and quantified across 116 microbial sample runs using a library built from 64 standards.

Identification of biomarkers for risk assessment of arsenicosis based on untargeted metabolomics and machine learning algorithms

BACKGROUND

Long-term exposure to inorganic arsenic may lead to arsenicosis. There are, however, currently no validated metabolic biomarkers used for the identification of arsenicosis risk. This study aims to identify metabolites associated with arsenicosis and establish prediction models for risk assessment based on untargeted metabolomics and machine learning algorithms.

METHODS

In total, 105 coal-borne arsenicosis patients, with 35 subjects in each of the mild, moderate, and severe subgroups according to their symptom severity, and 60 healthy residents were enrolled from Guizhou, China. Ultra-high performance liquid chromatography-tandem mass spectrometer (UHPLC-MS/MS) was utilized to acquire the plasma metabolic profiles of the studied subjects. Statistical analysis was used to identify disease-associated metabolites. Machine learning algorithms and the identified metabolic biomarkers were resorted to assess the arsenicosis risk.

RESULTS

A total of 143 metabolic biomarkers, with organic acids being the majority, were identified to be closely associated with arsenicosis, and the most involved pathway was glycine, serine, and threonine metabolism. Comparative analysis of metabolites in arsenicosis patients with different symptom severity revealed 422 altered molecules, where disrupted metabolism of beta-alanine and arginine demonstrated the most significance. For risk assessment, the model established by a single biomarker (L-carnosine) could undoubtedly discriminate arsenicosis patients from the healthy. For classifying arsenicosis patients with different severity, the model established using 52 metabolites and linear discriminate analysis (LDA) algorithm yielded an accuracy of 0.970-0.979 on calibration set (n = 132) and 0.818-0.848 on validation set (n = 33).

CONCLUSION

Altered metabolites and disrupted pathways are prevalent in arsenicosis patients; The disrupted metabolism of one carbon and dysfunction of antioxidant defense system may partially be causes of the systematic multi-organ damage and carcinogenesis in arsenicosis patients; Metabolic biomarkers, combined with machine learning algorithms, could be efficient for risk assessment and early identification of arsenicosis.

Biotechnology Data Analysis Training with Jupyter Notebooks

Biotechnology has experienced innovations in analytics and data processing. As the volume of data and its complexity grow, new computational procedures for extracting information are being developed. However, the rate of change outpaces the adaptation of biotechnology curricula, necessitating new teaching methodologies to equip biotechnologists with data analysis abilities. To simulate experimental data, we created a virtual organism simulator (silvio) by combining diverse cellular and subcellular microbial models. With the silvio Python package, we constructed a computer-based instructional workflow to teach growth curve data analysis, promoter sequence design, and expression rate measurement. The instructional workflow is a Jupyter Notebook with background explanations and Python-based experiment simulations combined. The data analysis is conducted either within the Notebook in Python or externally with Excel. This instructional workflow was separately implemented in two distance courses for Master's students in biology and biotechnology with assessment of the pedagogic efficiency. The concept of using virtual organism simulations that generate coherent results across different experiments can be used to construct consistent and motivating case studies for biotechnological data literacy.

Association of Metabolomic Biomarkers with Sleeve Gastrectomy Weight Loss Outcomes

This prospective observational study aimed to evaluate the association of metabolomic alterations with weight loss outcomes following sleeve gastrectomy (SG). We evaluated the metabolomic profile of serum and feces prior to SG and three months post-SG, along with weight loss outcomes in 45 adults with obesity. The percent total weight loss for the highest versus the lowest weight loss tertiles (T3 vs. T1) was 17.0 ± 1.3% and 11.1 ± 0.8%, p < 0.001. Serum metabolite alterations specific to T3 at three months included a decrease in methionine sulfoxide concentration as well as alterations to tryptophan and methionine metabolism (p < 0.03). Fecal metabolite changes specific to T3 included a decrease in taurine concentration and perturbations to arachidonic acid metabolism, and taurine and hypotaurine metabolism (p < 0.002). Preoperative metabolites were found to be highly predictive of weight loss outcomes in machine learning algorithms, with an average area under the curve of 94.6% for serum and 93.4% for feces. This comprehensive metabolomics analysis of weight loss outcome differences post-SG highlights specific metabolic alterations as well as machine learning algorithms predictive of weight loss. These findings could contribute to the development of novel therapeutic targets to enhance weight loss outcomes after SG.

Comprehensive Two-Dimensional Gas Chromatography as a Bioanalytical Platform for Drug Discovery and Analysis

Over the last decades, comprehensive two-dimensional gas chromatography (GC×GC) has emerged as a significant separation tool for high-resolution analysis of disease-associated metabolites and pharmaceutically relevant molecules. This review highlights recent advances of GC×GC with different detection modalities for drug discovery and analysis, which ideally improve the screening and identification of disease biomarkers, as well as monitoring of therapeutic responses to treatment in complex biological matrixes. Selected recent GC×GC applications that focus on such biomarkers and metabolite profiling of the effects of drug administration are covered. In particular, the technical overview of recent GC×GC implementation with hyphenation to the key mass spectrometry (MS) technologies that provide the benefit of enhanced separation dimension analysis with MS domain differentiation is discussed. We conclude by highlighting the challenges in GC×GC for drug discovery and development with perspectives on future trends.

Classification of multiple cancer types by combination of plasma-based near-infrared spectroscopy analysis and machine learning modeling

BACKGROUND AND AIM

Near-infrared spectroscopy (NIRS) is a non-invasive and convenient tool, which gains features related to chemical components in biological samples. Machine learning (ML) has been popularized in medical diagnosis. This study aimed at investigating a novel cancer diagnosis strategy using NIRS data based ML modeling.

METHODS

Plasma samples were collected from a total of 247 participants, including lung cancer, cervical cancer, nasopharyngeal cancer, and healthy control, and were randomly split into train set and test set. After performing NIRS analysis, the train dataset was utilized to train ML models, including partial least-squares (PLS), random forest (RF), gradient boosting machine (GBM), and support-vector machine (SVM). Subsequently, these models were tested for their prediction performance by the test set.

RESULTS

All ML models demonstrated high prediction performance in differentiating cancers from controls, and SVM had high prediction accuracy for different types of cancers. SVM was considered as the most suitable model for its minimal computational cost and high accuracies for both binary and quaternary classification.

CONCLUSIONS

This strategy coupling NIRS with ML is insightful that may aid in clinic cancer diagnosis, while further studies should test our results in a larger cohort with better representativeness.

Plant and microbial sciences as key drivers in the development of metabolomics research

This year marks the 25th anniversary of the coinage of the term metabolome [S. G. Oliver et al., Trends Biotech. 16, 373-378 (1998)]. As the field rapidly advances, it is important to take stock of the progress which has been made to best inform the disciplines future. While a medical-centric perspective on metabolomics has recently been published [M. Giera et al., Cell Metab. 34, 21-34 (2022)], this largely ignores the pioneering contributions made by the plant and microbial science communities. In this perspective, we provide a contemporary overview of all fields in which metabolomics is employed with particular emphasis on both methodological and application breakthroughs made in plant and microbial sciences that have shaped this evolving research discipline from the very early days of its establishment. This will not cover all types of metabolomics assays currently employed but will focus mainly on those utilizing mass spectrometry-based measurements since they are currently by far the most prominent. Having established the historical context of metabolomics, we will address the key challenges currently facing metabolomics and offer potential approaches by which these can be faced. Most salient among these is the fact that the vast majority of mass features are as yet not annotated with high confidence; what we may refer to as definitive identification. We discuss the potential of both standard compound libraries and artificial intelligence technologies to address this challenge and the use of natural variance-based approaches such as genome-wide association studies in attempt to assign specific functions to the myriad of structurally similar and complex specialized metabolites. We conclude by stating our contention that as these challenges are epic and that they will need far greater cooperative efforts from biologists, chemists, and computer scientists with an interest in all kingdoms of life than have been made to date. Ultimately, a better linkage of metabolome and genome data will likely also be needed particularly considering the Earth BioGenome Project.

Logistic Regression Analysis of LC-MS/MS Data of Monomers Eluted from Aged Dental Composites: A Supervised Machine-Learning Approach

Compound identification by database searching that matches experimental with library mass spectra is commonly used in mass spectrometric (MS) data analysis. Vendor software often outputs scores that represent the quality of each spectral match for the identified compounds. However, software-generated identification results can differ drastically depending on the initial search parameters. Machine learning is applied here to provide a statistical evaluation of software-generated compound identification results from experimental tandem MS data. This task was accomplished using the logistic regression algorithm to assign an identification probability value to each identified compound. Logistic regression is usually used for classification, but here it is used to generate identification probabilities without setting a threshold for classification. Liquid chromatography coupled with quadrupole-time-of-flight tandem MS was used to analyze the organic monomers leached from resin-based dental composites in a simulated oral environment. The collected tandem MS data were processed with vendor software, followed by statistical evaluation of these results using logistic regression. The assigned identification probability to each compound provides more confidence in identification beyond solely by database matching. A total of 21 distinct monomers were identified among all samples, including five intact monomers and chemical degradation products of bisphenol A glycidyl methacrylate (BisGMA), oligomers of bisphenol-A ethoxylate methacrylate (BisEMA), triethylene glycol dimethacrylate (TEGDMA), and urethane dimethacrylate (UDMA). The logistic regression model can be used to evaluate any database-matched liquid chromatography-tandem MS result by training a new model using analytical standards of compounds present in a chosen database and then generating identification probabilities for candidates from unknown data using the new model.

Using Artificial Intelligence to Better Predict and Develop Biomarkers

Advancements in technology have improved biomarker discovery in the field of heart failure (HF). What was once a slow and laborious process has gained efficiency through use of high-throughput omics platforms to phenotype HF at the level of genes, transcripts, proteins, and metabolites. Furthermore, improvements in artificial intelligence (AI) have made the interpretation of large omics data sets easier and improved analysis. Use of omics and AI in biomarker discovery can aid clinicians by identifying markers of risk for developing HF, monitoring care, determining prognosis, and developing druggable targets. Combined, AI has the power to improve HF patient care.

Prediction of a Large-Scale Database of Collision Cross-Section and Retention Time Using Machine Learning to Reduce False Positive Annotations in Untargeted Metabolomics

Metabolite identification in untargeted metabolomics is complex, with the risk of false positive annotations. This work aims to use machine learning to successively predict the retention time (Rt) and the collision cross-section (CCS) of an open-access database to accelerate the interpretation of metabolomic results. Standards of metabolites were tested using liquid chromatography coupled with high-resolution mass spectrometry. In CCSBase and QSRR predictor machine learning models, experimental results were used to generate predicted CCS and Rt of the Human Metabolome Database. From 542 standards, 266 and 301 compounds were detected in positive and negative electrospray ionization mode, respectively, corresponding to 380 different metabolites. CCS and Rt were then predicted using machine learning tools for almost 114,000 metabolites. R² score of the linear regression between predicted and measured data achieved 0.938 and 0.898 for CCS and Rt, respectively, demonstrating the models' reliability. A CCS and Rt index filter of mean error ± 2 standard deviations could remove most misidentifications. Its application to data generated from a toxicology study on tobacco cigarettes reduced hits by 76%. Regarding the volume of data produced by metabolomics, the practical workflow provided allows for the implementation of valuable large-scale databases to improve the biological interpretation of metabolomics data.

Impacts of Different Prenatal Supplementation Strategies on the Plasma Metabolome of Bulls in the Rearing and Finishing Phase

This study investigated the effects of maternal nutrition on the plasma metabolome of Nellore bulls in the rearing and finishing phases, and metabolic differences between these phases. For this study, three nutritional approaches were used in 126 cows during pregnancy: NP-(control) mineral supplementation; PP-protein-energy supplementation in the final third; and FP-protein-energy supplementation during the entire pregnancy. We collected blood samples from male offspring in the rearing (450 ± 28 days old) and finishing phases (660 ± 28 days old). The blood was processed, and from plasma samples, we performed the targeted metabolome analysis (AbsoluteIDQ® p180 Kit). Multiple linear regression, principal component analysis (PCA), repeated measures analysis over time, and an enrichment analysis were performed. PCA showed an overlap of treatments and time clusters in the analyses. We identified significant metabolites among the treatments (rearing phase = six metabolites; finishing phase = three metabolites) and over time (21 metabolites). No significant metabolic pathways were found in the finishing phase, however, we found significant pathways in the rearing phase (Arginine biosynthesis and Histidine metabolism). Thus, prenatal nutrition impacted on plasma metabolome of bulls during the rearing and finishing phase and the different production stages showed an effect on the metabolic levels of bulls.

Deep Learning of Dual Plasma Fingerprints for High-Performance Infection Classification

Infection classification is the key for choosing the proper treatment plans. Early determination of the causative agents is critical for disease control. Host responses analysis can detect variform and sensitive host inflammatory responses to ascertain the presence and type of the infection. However, traditional host-derived inflammatory indicators are insufficient for clinical infection classification. Fingerprints-based omic analysis has attracted increasing attention globally for analyzing the complex host systemic immune response. A single type of fingerprints is not applicable for infection classification (area under curve (AUC) of 0.550-0.617). Herein, an infection classification platform based on deep learning of dual plasma fingerprints (DPFs-DL) is developed. The DPFs with high reproducibility (coefficient of variation <15%) are obtained at low sample consumption (550 nL native plasma) using inorganic nanoparticle and organic matrix assisted laser desorption/ionization mass spectrometry. A classifier (DPFs-DL) for viral versus bacterial infection discrimination (AUC of 0.775) and coronavirus disease 2019 (COVID-2019) diagnosis (AUC of 0.917) is also built. Furthermore, a metabolic biomarker panel of two differentially regulated metabolites, which may serve as potential biomarkers for COVID-19 management (AUC of 0.677-0.883), is constructed. This study will contribute to the development of precision clinical care for infectious diseases.

Fusion of Quality Evaluation Metrics and Convolutional Neural Network Representations for ROI Filtering in LC-MS

Region of interest (ROI) extraction is a fundamental step in analyzing metabolomic datasets acquired by liquid chromatography-mass spectrometry (LC-MS). However, noises and backgrounds in LC-MS data often affect the quality of extracted ROIs. Therefore, developing effective ROI evaluation algorithms is necessary to eliminate false positives meanwhile keep the false-negative rate as low as possible. In this study, a deep fused filter of ROIs (dffROI) was proposed to improve the accuracy of ROI extraction by combining the handcrafted evaluation metrics with convolutional neural network (CNN)-learned representations. To evaluate the performance of dffROI, dffROI was compared with peakonly (CNN-learned representation) and five handcrafted metrics on three LC-MS datasets and a gas chromatography-mass spectrometry (GC-MS) dataset. Results show that dffROI can achieve higher accuracy, better true-positive rate, and lower false-positive rate. Its accuracy, true-positive rate, and false-positive rate are 0.9841, 0.9869, and 0.0186 on the test set, respectively. The classification error rate of dffROI (1.59%) is significantly reduced compared with peakonly (2.73%). The model-agnostic feature importance demonstrates the necessity of fusing handcrafted evaluation metrics with the convolutional neural network representations. dffROI is an automatic, robust, and universal method for ROI filtering by virtue of information fusion and end-to-end learning. It is implemented in Python programming language and open-sourced at https://github.com/zhanghailiangcsu/dffROI under BSD License. Furthermore, it has been integrated into the KPIC2 framework previously proposed by our group to facilitate real metabolomic LC-MS dataset analysis.

Shapley Value as a Quality Control for Mass Spectra of Human Glioblastoma Tissues

The automatic processing of high-dimensional mass spectrometry data is required for the clinical implementation of ambient ionization molecular profiling methods. However, complex algorithms required for the analysis of peak-rich spectra are sensitive to the quality of the input data. Therefore, an objective and quantitative indicator, insensitive to the conditions of the experiment, is currently in high demand for the automated treatment of mass spectrometric data. In this work, we demonstrate the utility of the Shapley value as an indicator of the quality of the individual mass spectrum in the classification task for human brain tumor tissue discrimination. The Shapley values are calculated on the training set of glioblastoma and nontumor pathological tissues spectra and used as feedback to create a random forest regression model to estimate the contributions for all spectra of each specimen. As a result, it is shown that the implementation of Shapley values significantly accelerates the data analysis of negative mode mass spectrometry data alongside simultaneous improving the regression models’ accuracy.

Fecal Metabolites as Biomarkers for Predicting Food Intake by Healthy Adults

BACKGROUND

The fecal metabolome is affected by diet and includes metabolites generated by human and microbial metabolism. Advances in -omics technologies and analytic approaches have allowed researchers to identify metabolites and better utilize large data sets to generate usable information. One promising aspect of these advancements is the ability to determine objective biomarkers of food intake.

OBJECTIVES

We aimed to utilize a multivariate, machine learning approach to identify metabolite biomarkers that accurately predict food intake.

METHODS

Data were aggregated from 5 controlled feeding studies in adults that tested the impact of specific foods (almonds, avocados, broccoli, walnuts, barley, and oats) on the gastrointestinal microbiota. Fecal samples underwent GC-MS metabolomic analysis; 344 metabolites were detected in preintervention samples, whereas 307 metabolites were detected postintervention. After removing metabolites that were only detected in either pre- or postintervention and those undetectable in ≥80% of samples in all study groups, changes in 96 metabolites relative concentrations (treatment postintervention minus preintervention) were utilized in random forest models to 1) examine the relation between food consumption and fecal metabolome changes and 2) rank the fecal metabolites by their predictive power (i.e., feature importance score).

RESULTS

Using the change in relative concentration of 96 fecal metabolites, 6 single-food random forest models for almond, avocado, broccoli, walnuts, whole-grain barley, and whole-grain oats revealed prediction accuracies between 47% and 89%. When comparing foods with one another, almond intake was differentiated from walnut intake with 91% classification accuracy.

CONCLUSIONS

Our findings reveal promise in utilizing fecal metabolites as objective complements to certain self-reported food intake estimates. Future research on other foods at different doses and dietary patterns is needed to identify biomarkers that can be applied in feeding study compliance and clinical settings.

Untargeted Metabolomics by Liquid Chromatography-Mass Spectrometry in Biomedical Research

Metabolomics, alone or in combination with other omics sciences, has shown great relevance in a large number of investigations in different branches of biomedicine, often providing novel discoveries and helping to expand the knowledge. Metabolomics analyses are carried out using different techniques, but in this chapter, we focus on liquid chromatography coupled to high-resolution mass spectrometry. The designated methodology consists of an untargeted approach for the analysis of plasma samples. The use of this method, with a reverse-phase column and electrospray ionization in positive mode, covers the detection of a broad range of metabolites, mainly of nonpolar and of intermediate polarity. This chapter also reviews the mass fragmentation spectra for the identification of bile acids, acylcarnitines, and glycerophospholipids.

Data Processing and Analysis in Liquid Chromatography-Mass Spectrometry-Based Targeted Metabolomics

Mass spectrometry (MS)-based metabolomics provides high-dimensional datasets; that is, the data include various metabolite features. Data analysis begins by converting the raw data obtained from the MS to produce a data matrix (metabolite × concentrations). This is followed by several steps, such as peak integration, alignment of multiple data, metabolite identification, and calculation of metabolite concentrations. Each step yields the analytical results and the accompanying information used for the quality assessment of the anterior steps. Thus, the measurement quality can be analyzed through data processing. Here, we introduce a typical data processing procedure and describe a method to utilize the intermediate data as quality control. Subsequently, commonly used data analysis methods for metabolomics data, such as statistical analyses, are also introduced.

Machine Learning and COVID-19: Lessons from SARS-CoV-2

Currently, methods in machine learning have opened a significant number of applications to construct classifiers with capacities to recognize, identify, and interpret patterns hidden in massive amounts of data. This technology has been used to solve a variety of social and health issues against coronavirus disease 2019 (COVID-19). In this chapter, we present some supervised and unsupervised machine learning techniques that have contributed in three aspects to supplying information to health authorities and diminishing the deadly effects of the current worldwide outbreak on the population. First is the identification and construction of powerful classifiers capable of predicting severe, moderate, or asymptomatic responses in COVID-19 patients starting from clinical or high-throughput technologies. Second is the identification of groups of patients with similar physiological responses to improve the triage classification and inform treatments. The final aspect is the combination of machine learning methods and schemes from systems biology to link associative studies with mechanistic frameworks. This chapter aims to discuss some practical applications in the use of machine learning techniques to handle data coming from social behavior and high-throughput technologies, associated with COVID-19 evolution.

Perspective: Multiomics and Machine Learning Help Unleash the Alternative Food Potential of Microalgae

Food security has become a pressing issue in the modern world. The ever-increasing world population, ongoing COVID-19 pandemic, and political conflicts together with climate change issues make the problem very challenging. Therefore, fundamental changes to the current food system and new sources of alternative food are required. Recently, the exploration of alternative food sources has been supported by numerous governmental and research organizations, as well as by small and large commercial ventures. Microalgae are gaining momentum as an effective source of alternative laboratory-based nutritional proteins as they are easy to grow under variable environmental conditions, with the added advantage of absorbing carbon dioxide. Despite their attractiveness, the utilization of microalgae faces several practical limitations. Here, we discuss both the potential and challenges of microalgae in food sustainability and their possible long-term contribution to the circular economy of converting food waste into feed via modern methods. We also argue that systems biology and artificial intelligence can play a role in overcoming some of the challenges and limitations; through data-guided metabolic flux optimization, and by systematically increasing the growth of the microalgae strains without negative outcomes, such as toxicity. This requires microalgae databases rich in omics data and further developments on its mining and analytics methods.