DaDIA: Hybridizing Data-Dependent and Data-Independent Acquisition Modes for Generating High-Quality Metabolomic Data

LC-MS/DIA-based strategy for comprehensive flavonoid profiling: an Ocotea spp. applicability case

We introduce a liquid chromatography - mass spectrometry with data-independent acquisition (LC-MS/DIA)-based strategy, specifically tailored to achieve comprehensive and reliable glycosylated flavonoid profiling. This approach facilitates in-depth and simultaneous exploration of all detected precursors and fragments during data processing, employing the widely-used open-source MZmine 3 software. It was applied to a dataset of six Ocotea plant species. This framework suggested 49 flavonoids potentially newly described for these plant species, alongside 45 known features within the genus. Flavonols kaempferol and quercetin, both exhibiting O-glycosylation patterns, were particularly prevalent. Gas-phase fragmentation reactions further supported these findings. For the first time, the apigenin flavone backbone was also annotated in most of the examined Ocotea species. Apigenin derivatives were found mainly in the C-glycoside form, with O. porosa displaying the highest flavone : flavonol ratio. The approach also allowed an unprecedented detection of kaempferol and quercetin in O. porosa species, and it has underscored the untapped potential of LC-MS/DIA data for broad and reliable flavonoid profiling. Our study annotated more than 50 flavonoid backbones in each species, surpassing the current literature.

Reproducibility in nontarget screening (NTS) of environmental emerging contaminants: Assessing different HLB SPE cartridges and instruments

Non-targeted screening (NTS) methods are integral in environmental research for detecting emerging contaminants. However, their efficacy can be influenced by variations in hydrophilic-lipophilic balance (HLB) solid phase extraction (SPE) cartridges and high-resolution mass spectrometry (HRMS) instruments across different laboratories. In this study, we scrutinized the influence of five HLB SPE cartridges (Nano, Weiqi, CNW, Waters, and J&K) and four LC-HRMS platforms (Agilent, Waters, Thermo, and AB SCIEX) on the identification of emerging environmental contaminants. Our results demonstrate that 87.6 % of the target compounds and over 59.6 % of the non-target features were consistently detected across all tested HLB cartridges, with an overall 71.2 % universally identified across the four LC-HRMS systems. Discrepancies in detection rates were primarily attributable to variations in retention time stability, mass stability of precursors and fragments, system cleanliness affecting fold change and p-values, and fragment response. These findings confirm the necessity of refining parameter criteria for NTS. Moreover, our study confirms the efficacy of the PyHRMS tool in analyzing and processing data from multiple instrumental platforms, reinforcing its utility for multi-platform NTS. Overall, our findings underscore the reliability and robustness of NTS methods in identifying potential water contaminants, while also highlighting factors that may influence these outcomes.

Windows Scanning Multiomics: Integrated Metabolomics and Proteomics

Metabolomics and proteomics offer significant advantages in understanding biological mechanisms at two hierarchical levels. However, conventional single omics analysis faces challenges due to the high demand for specimens and the complexity of intrinsic associations. To obtain comprehensive and accurate system biological information, we developed a multiomics analytical method called Windows Scanning Multiomics (WSM). In this method, we performed simultaneous extraction of metabolites and proteins from the same sample, resulting in a 10% increase in the coverage of the identified biomolecules. Both metabolomics and proteomics analyses were conducted by using ultrahigh-performance liquid chromatography mass spectrometry (UPLC-MS), eliminating the need for instrument conversions. Additionally, we designed an R-based program (WSM.R) to integrate mathematical and biological correlations between metabolites and proteins into a correlation network. The network created from simultaneously extracted biomolecules was more focused and comprehensive compared to those from separate extractions. Notably, we excluded six pairs of false-positive relationships between metabolites and proteins in the network established using simultaneously extracted biomolecules. In conclusion, this study introduces a novel approach for multiomics analysis and data processing that greatly aids in bioinformation mining from multiomics results. This method is poised to play an indispensable role in systems biology research.

Optimization of bronchoalveolar lavage fluid volume for untargeted lipidomic method and application in influenza A virus infection

Bronchoalveolar lavage (BAL) has been widely applied for the diagnosis of pulmonary diseases in clinical as it was recognized as a minimally invasive, well-tolerated and easily performed procedure. Lipid analysis of BAL fluid is a comprehensive strategy to observe lipid phenotypes, explore potential biomarkers, and elucidate the biological mechanisms of respiratory diseases. However, the highly diverse concentration of lipids in BAL fluid due to the deviation between the retrieved and injected aliquot volumes during lavage raised a challenge in obtaining high-quality lipidomic data. Here, this study aims to investigate what volume of BAL fluid is suitable for lipidomic analysis. Specifically, the BAL fluid harvested from H1N1 infected mice and controls was concentrated to varying degrees by freeze-drying technique before preparation for lipidomic analysis. The optimal concentration multiple of BAL fluid was approved by comparing the coverage and quality of identified lipids, as well as the number of differentially expressed lipids in the H1N1 infection model. Sixty-two differential lipids were identified respectively in the positive and negative modes when the BAL fluid was condensed five times, and they were classified into glycerolipids, phospholipids and fatty acids. This study focuses on the alterations of phospholipids, since they are the main constituents of pulmonary surfactants. Several phospholipids significantly accumulated in the BAL fluid of H1N1-infected mice, while most of them contained omega-3 polyunsaturated fatty acids, indicating disrupted inflammatory homeostasis in lungs. This study recommends freeze-drying/reconstitution prior to lipid extraction from BAL fluid for lipidomic analysis, as this procedure increased the richness and abundance of lipids.

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.

Identification and Prioritization of Environmental Organic Pollutants: From an Analytical and Toxicological Perspective

Exposure to environmental organic pollutants has triggered significant ecological impacts and adverse health outcomes, which have been received substantial and increasing attention. The contribution of unidentified chemical components is considered as the most significant knowledge gap in understanding the combined effects of pollutant mixtures. To address this issue, remarkable analytical breakthroughs have recently been made. In this review, the basic principles on recognition of environmental organic pollutants are overviewed. Complementary analytical methodologies (i.e., quantitative structure-activity relationship prediction, mass spectrometric nontarget screening, and effect-directed analysis) and experimental platforms are briefly described. The stages of technique development and/or essential parts of the analytical workflow for each of the methodologies are then reviewed. Finally, plausible technique paths and applications of the future nontarget screening methods, interdisciplinary techniques for achieving toxicant identification, and burgeoning strategies on risk assessment of chemical cocktails are discussed.

Improving the Data Quality of Untargeted Metabolomics through a Targeted Data-Dependent Acquisition Based on an Inclusion List of Differential and Preidentified Ions

Metabolomics based on high-resolution mass spectrometry has become a powerful technique in biomedical research. The development of various analytical tools and online libraries has promoted the identification of biomarkers. However, how to make mass spectrometry collect more data information is an important but underestimated research topic. Herein, we combined full-scan and data-dependent acquisition (DDA) modes to develop a new targeted DDA based on the inclusion list of differential and preidentified ions (dpDDA). In this workflow, the MS1 datasets for statistical analysis and metabolite preidentification were first obtained using full-scan, and then, the MS/MS datasets for metabolite identification were obtained using targeted DDA of quality control samples based on the inclusion list. Compared with the current methods (DDA, data-independent acquisition, targeted DDA with time-staggered precursor ion list, and iterative exclusion DDA), dpDDA showed better stability, higher characteristic ion coverage, higher differential metabolites' MS/MS coverage, and higher quality MS/MS spectra. Moreover, the same trend was verified in the analysis of large-scale clinical samples. More surprisingly, dpDDA can distinguish patients with different severities of coronary heart disease (CHD) based on the Canadian Cardiovascular Society angina classification, which we cannot distinguish through conventional metabolomics data collection. Finally, dpDDA was employed to differentiate CHD from healthy control, and targeted metabolomics confirmed that dpDDA could identify a more complete metabolic pathway network. At the same time, four unreported potential CHD biomarkers were identified, and the area under the receiver operating characteristic curve was greater than 0.85. These results showed that dpDDA would expand the discovery of biomarkers based on metabolomics, more comprehensively explore the key metabolites and their association with diseases, and promote the development of precision medicine.

New insights into the bioaccessibility and metabolic fates of short-chain bioactive peptides in goat milk using the INFOGEST static digestion model and an improved data acquisition strategy

The metabolic fates of potentially bioactive short-chain peptides (SCPs; amino acid numbers between 2 and 4) in gastrointestinal digestion have received little attention due to their low concentration and broad suppression during high resolution mass spectrometry (HRMS) analysis. A tailored workflow integrating mesoporous magnetic solid phase extraction and a novel ion transmission strategy (data-dependent acquisition combined with both an inclusion list and an exclusion list followed by a data-independent acquisition) was used to profile the composition of SCPs during in vitro simulated digestion (LOQ 0.02 to 0.1 μg L^-1). A total of 47 dipeptides, 59 tripeptides, and 21 tetrapeptides were identified and quantified from 0.01 to 27.84 mg L^-1 (RSD ≤ 9.1%) based on parallel reaction monitoring and an internal standard method. The structural properties of stable SCPs resistant to intestinal digestion were determined by analysis of variance (p < 0.05), with a Pro residue at the C-terminal or penultimate position, a slightly greater negative charge at pH 7.0, and fewer C-terminal aliphatic and polar amino acids. SCPs' metabolic fates varied during digestion, but the overall trend of content change for either total or individual SCP increased as the digestion proceeded, and they were further assessed by a database-driven bioactivity search, which matched a wide variety of bioactivities with the predominance of dipeptidyl peptidase (DPP) IV and angiotensin-converting enzyme (ACE) inhibitors. This study facilitated the understanding of bioaccessibility of the food-derived SCPs and provided essential guidelines for the properties of conserved structure in vivo.

Comparison of Compound Identification Tools Using Data Dependent and Data Independent High-Resolution Mass Spectrometry Spectra

Liquid chromatography combined with high-resolution mass spectrometry (LC-HRMS) is a frequently applied technique for suspect screening (SS) and non-target screening (NTS) in metabolomics and environmental toxicology. However, correctly identifying compounds based on SS or NTS approaches remains challenging, especially when using data-independent acquisition (DIA). This study assessed the performance of four HRMS-spectra identification tools to annotate in-house generated data-dependent acquisition (DDA) and DIA HRMS spectra of 32 pesticides, veterinary drugs, and their metabolites. The identification tools were challenged with a diversity of compounds, including isomeric compounds. The identification power was evaluated in solvent standards and spiked feed extract. In DDA spectra, the mass spectral library mzCloud provided the highest success rate, with 84% and 88% of the compounds correctly identified in the top three in solvent standard and spiked feed extract, respectively. The in silico tools MSfinder, CFM-ID, and Chemdistiller also performed well in DDA data, with identification success rates above 75% for both solvent standard and spiked feed extract. MSfinder provided the highest identification success rates using DIA spectra with 72% and 75% (solvent standard and spiked feed extract, respectively), and CFM-ID performed almost similarly in solvent standard and slightly less in spiked feed extract (72% and 63%). The identification success rates for Chemdistiller (66% and 38%) and mzCloud (66% and 31%) were lower, especially in spiked feed extract. The difference in success rates between DDA and DIA is most likely caused by the higher complexity of the DIA spectra, making direct spectral matching more complex. However, this study demonstrates that DIA spectra can be used for compound annotation in certain software tools, although the success rate is lower than for DDA spectra.

Analysis of Phospholipids in Digestion Using Hybrid IDA and SWATH Acquisition: An Example for Krill Oil

The composition and digestion of phospholipid-rich foods have important effects on the health of the body. Herein, a model-assisted liquid chromatography coupling mass spectrometry (LC-MS) method was established to analyze the phosphatidylcholine (PC) and lyso-phosphatidylcholine (LPC) species in krill oil before and after digestion. According to the confirmed PC and LPC species in the IDA (information dependent acquisition) results, three categories of mathematical models were set up, involving the retention time (RT), carbon number and unsaturation degree of the fatty acyl chain. All of the regression coefficient values (R²) were greater than 0.90, showing satisfactory fitting results. On this basis, using the computationally created precursor ion mass of PC and LPC species, 12 extra PC species and 4 LPC species were found in the SWATH (sequential windowed acquisition of all theoretical fragment ions) results. The PC and LPC compositions in the final digestive products had obvious differences among the different krill oils with different phospholipid content. Furthermore, more than half of the LPC species in the final digestive products were newly generated, indicating that LPC was one of basic constituents in the digestive products of krill oil. In conclusion, model-assisted hybrid IDA and SWATH acquisition has excellent detection performance, contributing to deep studies of the formations and functions of phospholipids.

Regions of Interest Multivariate Curve Resolution Liquid Chromatography with Data-Independent Acquisition Tandem Mass Spectrometry

New data-independent acquisition (DIA) modes coupled to chromatographic separations are opening new perspectives in the processing of massive mass spectrometric (MS) data using chemometric methods. In this work, the application of the regions of interest multivariate curve resolution (ROIMCR) method is shown for the simultaneous analysis of MS1 and MS2 DIA raw data obtained by liquid chromatography coupled to quadrupole-time-of-flight MS analysis. The ROIMCR method proposed in this work relies on the intrinsic bilinear structure of the MS1 and MS2 experimental data which allows us for the fast direct resolution of the elution and spectral profiles of all sample constituents giving measurable MS signals, without needing any further data pretreatment such as peak matching, alignment, or modeling. Compound annotation and identification can be achieved directly by the comparison of the ROIMCR-resolved MS1 and MS2 spectra with those from standards or from mass spectral libraries. ROIMCR elution profiles of the resolved components can be used to build calibration curves for the prediction of their concentrations in complex unknown samples. The application of the proposed procedure is shown for the analysis of mixtures of per- and polyfluoroalkyl substances in standard mixtures, spiked hen eggs, and gull egg samples, where these compounds tend to accumulate.

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.

Simulated-to-real benchmarking of acquisition methods in untargeted metabolomics

Data-Dependent and Data-Independent Acquisition modes (DDA and DIA, respectively) are both widely used to acquire MS2 spectra in untargeted liquid chromatography tandem mass spectrometry (LC-MS/MS) metabolomics analyses. Despite their wide use, little work has been attempted to systematically compare their MS/MS spectral annotation performance in untargeted settings due to the lack of ground truth and the costs involved in running a large number of acquisitions. Here, we present a systematic in silico comparison of these two acquisition methods in untargeted metabolomics by extending our Virtual Metabolomics Mass Spectrometer (ViMMS) framework with a DIA module. Our results show that the performance of these methods varies with the average number of co-eluting ions as the most important factor. At low numbers, DIA outperforms DDA, but at higher numbers, DDA has an advantage as DIA can no longer deal with the large amount of overlapping ion chromatograms. Results from simulation were further validated on an actual mass spectrometer, demonstrating that using ViMMS we can draw conclusions from simulation that translate well into the real world. The versatility of the Virtual Metabolomics Mass Spectrometer (ViMMS) framework in simulating different parameters of both Data-Dependent and Data-Independent Acquisition (DDA and DIA) modes is a key advantage of this work. Researchers can easily explore and compare the performance of different acquisition methods within the ViMMS framework, without the need for expensive and time-consuming experiments with real experimental data. By identifying the strengths and limitations of each acquisition method, researchers can optimize their choice and obtain more accurate and robust results. Furthermore, the ability to simulate and validate results using the ViMMS framework can save significant time and resources, as it eliminates the need for numerous experiments. This work not only provides valuable insights into the performance of DDA and DIA, but it also opens the door for further advancements in LC-MS/MS data acquisition methods.

Evaluating LC-HRMS metabolomics data processing software using FAIR principles for research software

BACKGROUND

Liquid chromatography-high resolution mass spectrometry (LC-HRMS) is a popular approach for metabolomics data acquisition and requires many data processing software tools. The FAIR Principles - Findability, Accessibility, Interoperability, and Reusability - were proposed to promote open science and reusable data management, and to maximize the benefit obtained from contemporary and formal scholarly digital publishing. More recently, the FAIR principles were extended to include Research Software (FAIR4RS).

AIM OF REVIEW

This study facilitates open science in metabolomics by providing an implementation solution for adopting FAIR4RS in the LC-HRMS metabolomics data processing software. We believe our evaluation guidelines and results can help improve the FAIRness of research software.

KEY SCIENTIFIC CONCEPTS OF REVIEW

We evaluated 124 LC-HRMS metabolomics data processing software obtained from a systematic review and selected 61 software for detailed evaluation using FAIR4RS-related criteria, which were extracted from the literature along with internal discussions. We assigned each criterion one or more FAIR4RS categories through discussion. The minimum, median, and maximum percentages of criteria fulfillment of software were 21.6%, 47.7%, and 71.8%. Statistical analysis revealed no significant improvement in FAIRness over time. We identified four criteria covering multiple FAIR4RS categories but had a low %fulfillment: (1) No software had semantic annotation of key information; (2) only 6.3% of evaluated software were registered to Zenodo and received DOIs; (3) only 14.5% of selected software had official software containerization or virtual machine; (4) only 16.7% of evaluated software had a fully documented functions in code. According to the results, we discussed improvement strategies and future directions.

Pseudotargeted Metabolomics Approach Enabling the Classification-Induced Ginsenoside Characterization and Differentiation of Ginseng and Its Compound Formulation Products

The use of diversified ginseng extracts in health-promoting foods is difficult to differentiate, as they share bioactive ginsenosides among different Panax species (e.g., P. ginseng, P. quinquefolius, P. notoginseng, and P. japonicus) and different parts (e.g., root, leaf, and flower). This work was designed to develop a pseudo-targeted metabolomics approach to discover ginsenoside markers facilitating the precise authentication of ginseng and its use in compound formulation products (CFPs). Versatile mass spectrometry experiments on the QTrap mass spectrometer achieved classified characterization of the neutral, malonyl, and oleanolic acid-type ginsenosides, with 567 components characterized. A pseudo-targeted metabolomics approach by multiple reaction monitoring (MRM) of 262 ion pairs could assist to establish key identification points for 12 ginseng species. The simultaneous detection of 14 markers enabled the identification of ginseng from 15 ginseng-containing CFPs. The pseudo-targeted metabolomics strategy enabled better performance in differentiating among multiple ginseng, compared with the full-scan high-resolution mass spectrometry approach.

Multicomponent Characterization of the Flower Bud of Panax notoginseng and Its Metabolites in Rat Plasma by Ultra-High Performance Liquid Chromatography/Ion Mobility Quadrupole Time-of-Flight Mass Spectrometry

The flower bud of Panax notoginseng (PNF) consumed as a tonic shows potential in the prevention and treatment of cardiovascular diseases. To identify the contained multi-components and, in particular, to clarify which components can be absorbed and what metabolites are transformed, unveiling the effective substances of PNF is of vital significance. A unique ultrahigh-performance liquid chromatography/ion mobility quadrupole time-of-flight mass spectrometry (UHPLC/IM-QTOF-MS) profiling approach and efficient data processing by the UNIFI^TM bioinformatics platform were employed to comprehensively identify the multi-components of PNF and the related metabolites in the plasma of rats after oral administration (at a dose of 3.6 g/kg). Two MS² data acquisition modes operating in the negative electrospray ionization mode, involving high-definition MS^E (HDMS^E) and data-dependent acquisition (DDA), were utilized aimed to extend the coverage and simultaneously ensure the quality of the MS² spectra. As a result, 219 components from PNF were identified or tentatively characterized, and 40 thereof could be absorbed. Moreover, 11 metabolites were characterized from the rat plasma. The metabolic pathways mainly included the phase I (deglycosylation and oxidation). To the best of our knowledge, this is the first report that systematically studies the in vivo metabolites of PNF, which can assist in better understanding its tonifying effects and benefit its further development.

Enhanced Identification of Ginsenosides Simultaneously from Seven Panax Herbal Extracts by Data-Dependent Acquisition Including a Preferred Precursor Ions List Derived from an In-House Programmed Virtual Library

Data-dependent acquisition (DDA) is widely utilized for metabolite identification in natural product research and food science, which, however, can suffer from low coverage. A potential solution to improve DDA coverage is to include the precursor ions list (PIL). Here, we aimed to construct a PIL-containing DDA strategy based on an in-house library of ginsenosides (VLG) and identify ginsenosides simultaneously from seven Panax herbal extracts. VLG, combined with mass defect filtering, could efficiently screen the ginsenoside precursors and elaborate the separate PIL involved in DDA for each ginseng extract. Consequently, we could characterize 500 ginsenosides, including 176 ones with unknown masses. Using the Panax ginseng extract, the superiority of this strategy was embodied in targeting more known ginsenoside masses and newly acquiring the MS² spectra of 13 components. Conclusively, knowledge-based large-scale molecular prediction and PIL-DDA can represent a powerful targeted/untargeted strategy beneficial to novel natural compound discovery.

Integrating Enhanced Profiling and Chemometrics to Unveil the Potential Markers for Differentiating among the Leaves of Panax ginseng, P. quinquefolius, and P. notoginseng by Ultra-High Performance Liquid Chromatography/Ion Mobility-Quadrupole Time-of-Flight Mass Spectrometry

The leaves of Panax species (e.g., Panax ginseng-PGL, P. quinquefolius-PQL, and P. notoginseng-PNL) can serve as a source for healthcare products. Comprehensive characterization and unveiling of the metabolomic difference among PGL, PQL, and PNL are critical to ensure their correct use. For this purpose, enhanced profiling and chemometrics were integrated to probe into the ginsenoside markers for PGL/PQL/PNL by ultra-high performance liquid chromatography/ion mobility-quadrupole time-of-flight mass spectrometry (UHPLC/IM-QTOF-MS). A hybrid scan approach (HDMSE-HDDDA) was established achieving the dimension-enhanced metabolic profiling, with 342 saponins identified or tentatively characterized from PGL/PQL/PNL. Multivariate statistical analysis (33 batches of leaf samples) could unveil 42 marker saponins, and the characteristic ginsenosides diagnostic for differentiating among PGL/PQL/PNL were primarily established. Compared with the single DDA or DIA, the HDMSE-HDDDA hybrid scan approach could balance between the metabolome coverage and spectral reliability, leading to high-definition MS spectra and the additional collision-cross section (CCS) useful to differentiate isomers.

Addressing big data challenges in mass spectrometry-based metabolomics

Advancements in computer science and software engineering have greatly facilitated mass spectrometry (MS)-based untargeted metabolomics. Nowadays, gigabytes of metabolomics data are routinely generated from MS platforms, containing condensed structural and quantitative information from thousands of metabolites. Manual data processing is almost impossible due to the large data size. Therefore, in the "omics" era, we are faced with new challenges, the big data challenges of how to accurately and efficiently process the raw data, extract the biological information, and visualize the results from the gigantic amount of collected data. Although important, proposing solutions to address these big data challenges requires broad interdisciplinary knowledge, which can be challenging for many metabolomics practitioners. Our laboratory in the Department of Chemistry at the University of British Columbia is committed to combining analytical chemistry, computer science, and statistics to develop bioinformatics tools that address these big data challenges. In this Feature Article, we elaborate on the major big data challenges in metabolomics, including data acquisition, feature extraction, quantitative measurements, statistical analysis, and metabolite annotation. We also introduce our recently developed bioinformatics solutions for these challenges. Notably, all of the bioinformatics tools and source codes are freely available on GitHub (https://www.github.com/HuanLab), along with revised and regularly updated content.

Prioritize biologically relevant ions for data-independent acquisition (BRI-DIA) in LC-MS/MS-based lipidomics analysis

INTRODUCTION

Data-dependent acquisition (DDA) is the most commonly used MS/MS scan method for lipidomics analysis on orbitrap-based instrument. However, MS instrument associated software decide the top N precursors for fragmentation, resulting in stochasticity of precursor selection and compromised consistency and reproducibility. We introduce a novel workflow using biologically relevant lipids to construct inclusion list for data-independent acquisition (DIA), named as BRI-DIA workflow.

OBJECTIVES

To ensure consistent coverage of biologically relevant lipids in LC-MS/MS-based lipidomics analysis.

METHODS

Biologically relevant ion list was constructed based on LIPID MAPS and lipidome atlas in MS-DIAL 4. Lipids were extracted from mouse tissues and used to assess different MS/MS scan workflow (DDA, BRI-DIA, and hybrid mode) on LC-Orbitrap Exploris 480 mass spectrometer.

RESULTS

DDA resulted in more MS/MS events, but the total number of unique lipids identified by three methods (DDA, BRI-DIA, and hybrid MS/MS scan mode) is comparable (580 unique lipids across 44 lipid subclasses in mouse liver). Major cardiolipin molecular species were identified by data generated using BRI-DIA and hybrid methods and allowed calculation of cardiolipin compositions, while identification of the most abundant cardiolipin CL72:8 was missing in data generated using DDA method, leading to wrong calculation of cardiolipin composition.

CONCLUSION

The method of using inclusion list comprised of biologically relevant lipids in DIA MS/MS scan is as efficient as traditional DDA method in profiling lipids, but offers better consistency of lipid identification, compared to DDA method. This study was performed using Orbitrap Exploris 480, and we will further evaluate this workflow on other platforms, and if verified by future work, this biologically relevant ion fragmentation workflow could be routinely used in many studies to improve MS/MS identification capacities.

Caenorhabditis elegans deep lipidome profiling by using integrative mass spectrometry acquisitions reveals significantly altered lipid networks

Lipidomics coverage improvement is essential for functional lipid and pathway construction. A powerful approach to discovering organism lipidome is to combine various data acquisitions, such as full scan mass spectrometry (full MS), data-dependent acquisition (DDA), and data-independent acquisition (DIA). Caenorhabditis elegans (C. elegans) is a useful model for discovering toxic-induced metabolism, high-throughput drug screening, and a variety of human disease pathways. To determine the lipidome of C. elegans and investigate lipid disruption from the molecular level to the system biology level, we used integrative data acquisition. The methyl-tert-butyl ether method was used to extract L4 stage C. elegans after exposure to triclosan (TCS), perfluorooctanoic acid, and nanopolystyrene (nPS). Full MS, DDA, and DIA integrations were performed to comprehensively profile the C. elegans lipidome by Q-Exactive Plus MS. All annotated lipids were then analyzed using lipid ontology and pathway analysis. We annotated up to 940 lipids from 20 lipid classes involved in various functions and pathways. The biological investigations revealed that when C. elegans were exposed to nPS, lipid droplets were disrupted, whereas plasma membrane-functionalized lipids were likely to be changed in the TCS treatment group. The nPS treatment caused a significant disruption in lipid storage. Triacylglycerol, glycerophospholipid, and ether class lipids were those primarily hindered by toxicants. Finally, toxicant exposure frequently involved numerous lipid-related pathways, including the phosphoinositide 3-kinase/protein kinase B pathway. In conclusion, an integrative data acquisition strategy was used to characterize the C. elegans lipidome, providing valuable biological insights into hypothesis generation and validation.

•

Multiple data acquisitions were used to profile the lipidome of C. elegans.

•

940 detected lipids of 20 main classes involved in various pathways.

•

Relevant hypotheses were generated using high-coverable lipidomics and pathways analysis.

Application of Large-Scale Molecular Prediction for Creating the Preferred Precursor Ions List to Enhance the Identification of Ginsenosides from the Flower Buds of Panax ginseng

This work was designed to evaluate the coverage of data-dependent acquisition (DDA) extensively utilized in the untargeted metabolite/component identification in the food sciences and pharmaceutical analysis. Using saponins from the flower buds of Panax ginseng (PGF) as an example, precursor ions list (PIL)-including DDA on a Q-Orbitrap mass spectrometer could enable higher coverage than the other four MS² acquisition approaches in characterizing PGF ginsenosides. A "Virtual Library of Ginsenoside" containing 13,536 ginsenoside molecules was established by C-language-programmed large-scale molecular prediction, which in combination with mass defect filtering could create a new PIL involving 1859 PGF saponin precursors. We could newly obtain the MS² spectra of at least 17 components and characterize 36 ginsenosides with unknown masses, among the 164 compounds identified from PGF. Conclusively, a molecular-prediction-oriented PIL in DDA can assist to discover more potentially novel molecules benefiting to the development of functional foods and new drugs.

A novel hybrid scan approach enabling the ion-mobility separation and the alternate data-dependent and data-independent acquisitions (HDDIDDA): Its combination with off-line two-dimensional liquid chromatography for comprehensively characterizing the multicomponents from Compound Danshen Dripping Pill

Data-dependent acquisition (DDA) and data-independent acquisition (DIA)-based MSⁿ strategies are extensively applied in metabolites characterization. DDA gives accurate MSⁿ information, but receives low coverage, while DIA covers the entire mass range, but the precursor-product ions matching often yields false positives. Currently available MS scan approaches rarely integrate DIA and DDA within a duty circle. Utilizing a Vion™ IM-QTOF (ion mobility-quadrupole time-of-flight) mass spectrometer, we report a novel hybrid scan approach, namely HDDIDDA, which involves three scan events: 1) IM-enabled full scan (MS¹), 2) high-definition MS^E (HDMS^E) of all precursor ions (MS²); and 3) high-definition DDA (HDDDA) of top N precursors (MS²). As a proof-of-concept, the HDDIDDA approach combined with off-line two-dimensional liquid chromatography (2D-LC) was applied to characterize the multiple ingredients from a reputable Chinese patent medicine, Compound Danshen Dripping Pill (CDDP) used for treating the cardiovascular diseases. An off-line 2D-LC system by configuring an XBridge Amide column and an HSS T3 column showed a measurable orthogonality of 0.92 and enhanced the separation of co-eluting components. A fit-for-purpose HDDIDDA methodology was developed in the negative mode to characterize saponins and salvianolic acids, while tanshinones in the positive mode. Computational workflows to efficiently process the acquired HDMS^E and HDDDA data were established, and the searching of an in-house CDDP library (recording 712 compounds) eventually characterized 403 components from CDDP, indicating approximate 12-fold improvement compared with the previous report. The HDDIDDA approach can measure collision cross section of each component, and merges the merits of DIA and DDA in MS² data acquisition.

EVA: Evaluation of Metabolic Feature Fidelity Using a Deep Learning Model Trained With Over 25000 Extracted Ion Chromatograms

Extracting metabolic features from liquid chromatography-mass spectrometry (LC-MS) data relies on the recognition of extracted ion chromatogram (EIC) peak shapes using peak picking algorithms. Unfortunately, all peak picking algorithms present a significant drawback of generating a problematic number of false positives. In this work, we take advantage of deep learning technology to develop a convolutional neural network (CNN)-based program that can automatically recognize metabolic features with poor EIC shapes, which are of low feature fidelity and more likely to be false. Our CNN model was trained using 25095 EIC plots collected from 22 LC-MS-based metabolomics projects of various sample types, LC and MS conditions. Notably, we manually inspected all the EIC plots to assign good or poor EIC quality for accurate model training. The trained CNN model is embedded into a C#-based program, named EVA (short for evaluation). The EVA Windows Application is a versatile platform that can process metabolic features generated by LC-MS systems of various vendors and processed using various data processing software. Our comprehensive evaluation of EVA indicates that it achieves over 90% classification accuracy. EVA can be readily used in LC-MS-based metabolomics projects and is freely available on the Microsoft Store by searching "EVA Metabolomics".

CRB-SWATH: A Method for Enhancing Untargeted Precursor Ion Extraction and Automatically Constructing Their Tandem Mass Spectra from SWATH Datasets by Chromatographic Retention Behaviors

Sequential window acquisition of all theoretical spectra (SWATH) as a typical data-independent acquisition (DIA) strategy is favorable for untargeted metabolomics. It could theoretically acquire product ions of all precursor ions, including precursor ions showing chromatographic peaks of rather poor qualities. However, existing data processing methods present limited capabilities in capturing poor-quality peaks of precursor ions. Thus, although their product ions could be acquired, their precursor ions are absent. Here, we present a new strategy, chromatographic retention behavior-SWATH (CRB-SWATH), that could unbiasedly capture poor-quality peaks and provide high resolutions of multiplexed mass spectroscopy (MS/MS) spectra in SWATH datasets. CRB-SWATH monitors CRBs of SWATH-MS signals under a series of altered elution gradients. As signals of compounds differ from noise by showing CRBs, both the precursor and fragment ions are captured, while ignoring their peak qualities. Moreover, CRB-SWATH offers good chances to resolve highly multiplexed MS/MS spectra in SWATH datasets because precursor ions coeluted in a single elution gradient often present different CRBs. In the untargeted metabolic analysis of Hela cell extracts, CRB-SWATH showed the advantage in exclusively capturing 2645 ions of poor-quality peaks (i.e., tiny peaks, discontinuous ion traces, tailing peaks, zigzag peaks, etc.), accounting for 34.4% of all the untargeted precursor ions extracted. Therein, it is noteworthy that among 2116 negative ions detected in hydrophilic interaction liquid chromatography (HILIC) mode, 1284 poor-quality ion peaks (>60%) were exclusively captured by CRB-SWATH. As CRB-SWATH automatically captures a large sum of true ion peaks of poor qualities, extracts MS/MS spectra of high purities, and provides chromatographic retention behaviors of untargeted metabolites for identification and classification, it could be a useful metabolomics tool for understanding biological phenomena better.

ISFrag: De Novo Recognition of In-Source Fragments for Liquid Chromatography-Mass Spectrometry Data

In-source fragmentation (ISF) is a naturally occurring phenomenon during electrospray ionization (ESI) in liquid chromatography-mass spectrometry (LC-MS) analysis. ISF leads to false metabolite annotation in untargeted metabolomics, prompting misinterpretation of the underlying biological mechanisms. Conventional metabolomic data cleaning mainly focuses on the annotation of adducts and isotopes, and the recognition of ISF features is mainly based on common neutral losses and the LC coelution pattern. In this work, we recognized three increasingly important patterns of ISF features, including (1) coeluting with their precursor ions, (2) being in the tandem MS (MS²) spectra of their precursor ions, and (3) sharing similar MS² fragmentation patterns with their precursor ions. Based on these patterns, we developed an R package, ISFrag, to comprehensively recognize all possible ISF features from LC-MS data generated from full-scan, data-dependent acquisition, and data-independent acquisition modes without the assistance of common neutral loss information or MS² spectral library. Tested using metabolite standards, we achieved a 100% correct recognition of level 1 ISF features and over 80% correct recognition for level 2 ISF features. Further application of ISFrag on untargeted metabolomics data allows us to identify ISF features that can potentially cause false metabolite annotation at an omics-scale. With the help of ISFrag, we performed a systematic investigation of how ISF features are influenced by different MS parameters, including capillary voltage, end plate offset, ion energy, and "collision energy". Our results show that while increasing energies can increase the number of real metabolic features and ISF features, the percentage of ISF features might not necessarily increase. Finally, using ISFrag, we created an ISF pathway to visualize the relationships between multiple ISF features that belong to the same precursor ion. ISFrag is freely available on GitHub (https://github.com/HuanLab/ISFrag).

High-Coverage Screening of Sulfonamide Metabolites in Goat Milk by Magnetic Doped S Graphene Combined with Ultrahigh-Performance Liquid Chromatography-High-Resolution Mass Spectrometry

Currently, there are more than 1000 varieties of synthetic sulfonamides universally used as antibiotics causing severe results of potential carcinogenicity and drug resistance for human health due to excessive residue of animal-derived food. A facile and novel approach for untargeted screening of sulfonamides (SAs) and metabolites was proposed based on magnetic solid-phase extraction-ultrahigh-performance liquid chromatography-tandem high-resolution mass spectrometry (MSPE-UHPLC-HRMS). Compared with QuEChERS without the clean-up procedure and SPE in terms of matrix effect and absolute recovery, magnetic doped S graphene (S-doping level: 2.82%) synthesized via a solid-state microwave approach and the aggregation wrap mechanism was used as a novel adsorbent for nonspecific extraction of desired analytes by the noncovalent interaction between electron-deficient thiophene sulfur and electron donors such as amino or amide as well as π-π stacking interactions. Combined with variable data-independent acquisition, characteristic fragment-ion filtering (m/z 156.01138 or m/z 108.04439) and assignment of extracted-ion chromatograms of marked fragment ions were successfully utilized to screen the desired analytes and subsequently confirmed with the availability of reference standards. The optimized and validated approach for spiked 26 SAs and 9 metabolites in control goat milk demonstrated satisfactory accuracy (80.1-112.6%) and precision (RSDs < 6.4%) for matrix-matched standard addition. After applying suspect goat milk samples, untargeted SA analytes including sulfanilamide or an N⁴-acetylsulfamethazine metabolite with concentration ranging from 66.3 to 398.5 ng L^-1 were determined in 5 of 45 goat milk samples.

Novel Strategy for Untargeted Chiral Metabolomics using Liquid Chromatography-High Resolution Tandem Mass Spectrometry

Stereospecific recognition of metabolites plays a significant role in the detection of potential disease biomarkers thereby providing new insights in diagnosis and prognosis. D-Hdroxy/amino acids are recognized as potential biomarkers in several metabolic disorders. Despite continuous advances in metabolomics technologies, the simultaneous measurement of different classes of enantiomeric metabolites in a single analytical run remains challenging. Here, we develop a novel strategy for untargeted chiral metabolomics of hydroxy/amine groups (-OH/-NH₂) containing metabolites, including all hydroxy acids (HAs) and amino acids (AAs), by chiral derivatization coupled with liquid chromatography-high resolution tandem mass spectrometry (LC-HR-MS/MS). Diacetyl-tartaric anhydride (DATAN) was used for the simultaneous derivatization of-OH/-NH₂ containing metabolites as well as the resulting diastereomers, and all the derivatized metabolites were resolved in a single analytical run. Data independent MS/MS acquisition (DIA) was applied to positively identify DATAN-labeled metabolites based on reagent specific diagnostic fragment ions. We discriminated chiral from achiral metabolites based on the reversal of elution order of D and L isomers derivatized with the enantiomeric pair (±) of DATAN in an untargeted manner. Using the developed strategy, a library of 301 standards that consisted of 214 chiral and 87 achiral metabolites were separated and detected in a single analytical run. This approach was then applied to investigate the enantioselective metabolic profile of the bone marrow (BM) and peripheral blood (PB) plasma samples from patients with acute myeloid leukemia (AML) at diagnosis and following completion of the induction phase of chemotherapeutic treatment. The sensitivity and selectivity of the developed method enabled the detection of trace levels of the D-enantiomer of HAs and AAs in primary plasma patient samples. Several of these metabolites were significantly altered in response to chemotherapy. The developed LC-HR-MS method entails a valuable step forward in chiral metabolomics.