Metabolites of interest were annotated using different approaches, in accordance with the guidelines of the Metabolomics Standards Initiative. For level 1 identification, data from QC samples were compared to an in-house library of authentic standards (see section 2.1), using retention time and spectral data (MS1 and MS/MS2) analyzed with the Data Analysis software. Level 2 annotation employed two complementary tools. The first was the Data Analysis software, used to compare MS/MS2 fragmentation patterns against those from public spectral libraries, such as MassBank, Human Metabolome Database (HMDB), and in-house spectral collections. The second tool was the Global Natural Products Social (GNPS) platform. Raw data (.d) files containing MS/MS2 data were converted to .mzML format using Data Analysis software. Positive and negative ionization modes were processed separately to generate two peak lists for metabolite identification. Classical Molecular Networking was applied on GNPS, with a precursor ion mass tolerance of 5 ppm and fragment ion mass tolerances also set to 5 ppm. Spectra were clustered into nodes, which were connected only if the cosine similarity score exceeded 0.7 and at least three fragment peaks matched. The resulting molecular networks were then searched against GNPS spectral libraries using the same parameters. Level 3 annotation was conducted using SIRIUS v5.6.3, a Java-based software that performs computational predictions of molecular formulas and proposes candidate structures based on MS/MS2 fragmentation patterns. This approach relies on in silico fragmentation and scoring algorithms, without validation by reference standards or spectral libraries.
"],"repository":["MetaboLights"],"study_status":["Public"],"ptm_modification":[""],"instrument_platform":["Liquid Chromatography MS - negative - reverse-phase","Liquid Chromatography MS - positive - reverse-phase"],"chromatography_protocol":["Metabolomic experiments were carried out using an ultra-high performance liquid chromatography system (UHPLC, Nexera X2 Shimadzu), connected to a mass spectrometer (MS) equipped with Quadrupole-Time-of Flight (QTOF) analyzer. An electrospray (ESI) (Impact II Bruker, Bruker Corporation, Massachusetts, EUA) was employed as an ionization source. Samples were randomized for the run order and analyzed both in the negative and positive electrospray modes. The separation in the UHPLC was performed with an Acquity ULPC column T3 (50 x 2.1 mm, 1.8 µm particle size), at the flow rate of 0.5 mL min-1 and column temperature of 40 ºC. The analysis was performed with an injection volume of 5 μL per sample. The mobile phase consisted of solvents A (water acidified with 0.1% formic acid) and B (acetonitrile acidified with 0.1 % formic acid). The gradient of mobile phases was the following: 0 % B held for 0.5 min, a linear increase to 100 % B over 8 min, 100 % B held for 1 min, followed by a return to 0 % B for 1 additional min. The injection volume of samples was 5 µL. Quality control (QC) samples were repeatedly injected to stabilize the instrument until consistent base peak chromatograms were achieved. Additionally, QC samples were analyzed every six-sample injection (approximately once per hour) to assess intra-run variability.
"],"publication":["Untargeted UHPLC-MS metabolomics reveals candidate regional markers in commercial wines from Southern Brazil."],"submitter_name":["Gustavo Henz"],"submitter_affiliation":["Federal University of Rio Grande do Sul"],"organism_part":["blank","Grape","Quality Control"],"technology_type":["mass spectrometry assay"],"disease":[""],"extraction_protocol":["As the samples consisted of wine, no extraction step was performed.
"],"organism":["blank","Vitis vinifera","Quality Control"],"full_dataset_link":["https://www.ebi.ac.uk/metabolights/MTBLS14574"],"author":["Giovana Domeneghini Mercali.","Gustavo Henz. Universidade Federal do Rio Grande do Sul. ghenz1@ucs.br.","Vitor Manfroi.","Eliseu Rodrigues."],"data_transformation_protocol":["Data obtained from UHPLC-QTOF-MS analysis, originally stored in Bruker (.d) format, were converted to .mzML files using DataAnalysis software. The converted files were processed in MZMine 4.7.29 (Heuckeroth et al., 2024), an open-source platform for mass spectrometry data analysis for peak deconvolution. This process deconvolutes LC-MS spectral features based on expected isotope patterns, adducts, and charge states, and aligns them across all samples. The workflow included mass detection using wavelets (Automated Data Analysis Pipeline, ADAP), encompassing feature detection, isotope grouping, alignment, gap filling, and filtering. The workflow and detailed parameter settings are provided in the Supplementary Material (Tables S3-S6), and the feature tables obtained after pre-processing are presented in Tables S7-S10. The workflow and the detailed parameter settings are provided in Supplementary Material, Tables S1 and S2. Data processing was carried out using R software (version 4.4.0). Metabolites that exhibited less than a tenfold increase in quality control (QC) samples compared to blanks were removed from the dataset. Only those features consistently detected across all QC samples and showing a coefficient of variation below 30% were retained. Missing values were estimated as half of the smallest detected peak area. To consolidate ions originating from the same metabolite, feature clustering was performed using the notame package. Before statistical analyses, data were transformed to a log2 scale. Principal Component Analysis (PCA) was applied for dimensionality reduction and visualization of sample groupings, with graphical outputs generated via the ggplot2 packag. Features differing significantly between groups (Student’s t-test, p ≤ 0.05, adjusted using the Benjamini–Hochberg false discovery rate correction) were illustrated through heatmaps constructed with hierarchical clustering based on Euclidean distance and Ward’s linkage method. Differences in the relative abundance of identified metabolites were evaluated by one-way Student’s t-test ANOVA (α = 0.05p ≤ 0.05, adjusted using the Benjamini–Hochberg false discovery rate correction) followed by Tukey’s post hoc test, with results displayed as boxplots. Unpaired t-tests and ANOVAs with FDR adjustment were conducted using the rstatix package.
"],"study_factor":["Region"],"submitter_email":["ghenz1@ucs.br"],"sample_collection_protocol":["A total of 23 Chardonnay and 33 Merlot wine samples produced in the state of Rio Grande do Sul were collected between June 17 and August 31, 2024. The grapes used for these wines originated from Geographical Indications, including Vale dos Vinhedos, Pinto Bandeira, Monte Belo, Altos Montes, and Campanha Gaúcha. Additionally, some samples were produced from grapes grown in other locations within the Serra Gaúcha region, such as Santa Tereza, Cotiporã, and Muitos Capões. The wines represented vintages from 2017 to 2024 and were collected in different containers, all of them completely filled, as detailed in. After collecting, the wine samples were aliquoted into Eppendorf tubes and stored in an ultra-freezer until analysis.
"],"omics_type":["Metabolomics"],"study_design":["Wine","pooled quality control sample","Metabolomics","blank","Waters ACQUITY UPLC H-Class System","untargeted analysis","solvent blank","Chardonnay grape cultivar","UHPLC instrument","Shimadzu Nexera X2 UHPLC system (binary pump, DGU-20A degasser, and CBM-20A controller)","Merlot grape cultivar","Vitis vinifera","experimental sample","untargeted metabolite profiling","Grape","Bruker impact II UHR-TOF","Thermo Scientific Q Exactive HF","Quality Control"],"curator_keywords":["Wine","Metabolomics","pooled quality control sample","Waters ACQUITY UPLC H-Class System","blank","untargeted analysis","solvent blank","Chardonnay grape cultivar","UHPLC instrument","Shimadzu Nexera X2 UHPLC system (binary pump, DGU-20A degasser, and CBM-20A controller)","Merlot grape cultivar","Vitis vinifera","experimental sample","untargeted metabolite profiling","Grape","Bruker impact II UHR-TOF","Thermo Scientific Q Exactive HF","Quality Control"],"mass_spectrometry_protocol":["The QTOF-MS parameters were set as follows: the scan range from m/z 50 to 1500 Da, capillary 3500 V (positive mode)/+3500 V (negative mode), drying gas temperature of 310 ºC, and nebulizer gas pressure of 4 bar. MS/MS2 spectra were acquired in data-dependent acquisition (DDA) mode with a threshold of 1500, performed exclusively for the QC sample. Instrument control and raw data processing were carried out using Esquire Control and DataAnalysis 4.3 software, respectively (Bruker Daltonics, Billerica, MA, USA). Raw data were calibrated internally using sodium formate clusters.
"],"additional_accession":[]},"is_claimable":false,"name":"Untargeted UHPLC-MS metabolomics reveals candidate regional markers in commercial wines from Southern Brazil","description":"Rio Grande do Sul is Brazil’s leading wine-producing state and hosts the majority of the country’s Geographical Indications. This study hypothesizes that commercial Merlot and Chardonnay wines from different Geographical Indications and municipalities of Rio Grande do Sul exhibit distinct chemical fingerprints identifiable by untargeted UHPLC–MS/MS metabolomic profiling. To test this hypothesis, a strategy based on ultra-high-performance liquid chromatography coupled with high-resolution mass spectrometry (UHPLC–MS/MS) was employed. The integration of metabolomic profiling with multivariate statistical analyses, alongside the evaluation of reducing and antioxidant capacity, enabled a differentiation of wines from the two main mesoregions, Campanha and Serra Gaúcha, despite the high variability of the dataset. Metabolite annotation revealed that several some discriminant compounds were associated with agronomic practices, oenological processes, and climatic conditions, with particular emphasis on variations in ultraviolet radiation exposure. Nitrogen-containing metabolites, such as amino acids, peptides, and choline, as well as flavonols and anthocyanins, consistently played a central role in regional discrimination. Notably, the recurrent detection of quercetin as a distinguishing feature of Serra Gaúcha wines underscores its potential utility as a chemical marker linked to terroir. Collectively, these findings demonstrate that metabolomic profiling constitutes a robust and reliable approach for investigating geographical origin, typicity, and authenticity in Brazilian wines.","dates":{"publication":"2026-06-15","submission":"2026-05-25"},"accession":"MTBLS14574","cross_references":{}}