Variable Importance In Projection Values Research Articles

Untargeted and targeted metabolomics strategy for the classification of strong aroma-type baijiu (liquor) according to geographical origin using comprehensive two-dimensional gas chromatography-time-of-flight mass spectrometry

A metabolomics strategy was developed to differentiate strong aroma-type baijiu (SAB) (distilled liquor) from the Sichuan basin (SCB) and Yangtze-Huaihe River Basin (YHRB) through liquid–liquid extraction coupled with GC×GC-TOFMS. PCA effectively separated the samples from these two regions. The PLS-DA training model was excellent, with explained variation and predictive capability values of 0.988 and 0.982, respectively. As a result, the model demonstrated its ability to perfectly differentiate all the unknown SAB samples. Twenty-nine potential markers were located by variable importance in projection values, and twenty-four of them were identified and quantitated. Discrimination ability is closely correlated to the characteristic flavor compounds, such as acid, esters, furans, alcohols, sulfides and pyrazine. Most of the marker compounds were less abundant in the SCB samples than in the YHRB samples. The quantitated markers were further processed using hierarchical cluster analysis for targeted analysis, indicating that the markers had great discrimination power to differentiate the SAB samples.

Exploration of Floral Volatile Organic Compounds in Six Typical Lycoris taxa by GC-MS.

Lycoris, which is known as the ‘Chinese tulip,’ has diverse flower colors and shapes, and some species have a delicate fragrance. However, limited studies have reported the volatile organic compounds (VOCs) of Lycoris. In this study, headspace solid-phase microextraction combined with gas chromatography-mass spectrometry was used to analyze the floral VOCs of six typical Lycoris taxa. Thirty-two VOCs were identified, including terpenoids, alcohols, esters, aldehydes, ketones, and phenols. The aldehyde and terpenoid contents in Lycoris aurea were higher than in the other taxa, and the ester and alcohol contents in L. sprengeri were the highest compared to all taxa tested. Compared with other species and cultivars, L. longituba and L. longituba var. flava were the two most scented taxa and the VOCs were dominated by terpenoids and esters. L. radiate and L. chinensis were two unscented taxa and, accordingly, the VOC content was weak. A partial least squares discriminate analysis of the floral VOCs among the six Lycoris taxa showed that the six taxa could be successfully separated. Moreover, the VOCs of L. longituba and L. longituba var. flava clustered together. β-Ocimene was verified as the most important aroma compound, as determined via the calculation of the variable importance in projection values and significance analysis. β-Ocimene and its trans isomer, trans-β-ocimene, had a high relative content in L. longituba, L. longituba var. flava, L. aurea, and L. chinensis but were not detected in L. sprengeri and L. radiata. These results indicate that floral VOCs might be selected during the evolutional processes of Lycoris, and β-ocimene could be the most typical VOC among the different Lycoris taxa.

Quality Evaluation of Mahonia bealei (Fort.) Carr. Using Supercritical Fluid Chromatography with Chemical Pattern Recognition.

Mahonia bealei (Fort.) Carr. (M. bealei) plays an important role in the treatment of many diseases. In the present study, a comprehensive method combining supercritical fluid chromatography (SFC) fingerprints and chemical pattern recognition (CPR) for quality evaluation of M. bealei was developed. Similarity analysis, hierarchical cluster analysis (HCA), principal component analysis (PCA) were applied to classify and evaluate the samples of wild M. bealei, cultivated M. bealei and its substitutes according to the peak area of 11 components but an accurate classification could not be achieved. PLS-DA was then adopted to select the characteristic variables based on variable importance in projection (VIP) values that responsible for accurate classification. Six characteristics peaks with higher VIP values (≥1) were selected for building the CPR model. Based on the six variables, three types of samples were accurately classified into three related clusters. The model was further validated by a testing set samples and predication set samples. The results indicated the model was successfully established and predictive ability was also verified satisfactory. The established model demonstrated that the developed SFC coupled with PLS-DA method showed a great potential application for quality assessment of M. bealei.

Characterization and discrimination of selected chicken eggs in China's retail market based on multi-element and lipidomics analysis.

The present study focuses on the differentiation of Deqingyuan (n = 90), Taihe (n = 90) and crossbred (n = 90) black-boned silky fowl eggs based on chemometrics analysis of multi-element and lipidomics data. A total of 53 chemical elements were simultaneously quantified in the eggs using inductively coupled plasma mass spectrometry. Multivariate statistical analysis of element data screened 30, 13 and 12 candidate discriminants for whole egg, egg yolk and egg white samples, respectively. Moreover, we utilized an LC-MS/MS-based lipidomics approach to profile lipids in the three types of egg yolks. In total, 1633 lipid species (including 43.78% glycerophospholipids, 25.66% glycerolipids, 16.66% fatty acyls, 6.86% sphingolipids, 4.10% sterol lipids, 1.53% polyketides, 1.10% prenol lipids and 0.31% saccharolipids) were consistently detected in all samples. A partial least squares discriminant analysis model was established based on the lipidomics data, which enables a 100% correct classification of the egg yolks. Using a cutoff of variable importance in projection value >1 and p value <.05, a panel of 22 potential lipid markers was discovered for the discrimination of Deqingyuan, Taihe and crossbred egg yolks. These results could facilitate a better understanding of the nutritional qualities of the evaluated eggs. Our method of combining chemical analysis with chemometrics offers a convenient and powerful tool to distinguish commercially-available eggs.

Which intensities, types, and patterns of movement behaviors are most strongly associated with cardiometabolic risk factors among children?

PurposeThe purpose of this study was to determine which intensities, patterns, and types of 24-h movement behaviors are most strongly associated with cardiometabolic risk factors among children. MethodsA total of 369 children aged 10–13 years were studied. Participants wore an Actical accelerometer and a Garmin Forerunner 220 Global Positioning System logger and completed an activity and sleep log for 7 days. Data from these instruments were combined to estimate average minute per day spent in 14 intensities, 11 types, and 14 patterns of movement. Body mass index, resting heart rate, and systolic blood pressure values were combined to create a cardiometabolic risk factor score. Partial least squares regression analysis was used to examine associations between the 39 movement behavior characteristics and the cardiometabolic risk factor score. The variable importance in projection (VIP) values were used to determine and rank important movement behavior characteristics. There was evidence of interaction by biological maturity, and the analyses were conducted separately in the 50% least mature and 50% most mature participants. ResultsFor the least biologically mature participants, fifteen of the 39 movement behavior characteristics had important VIP value scores; eight of these reflected movement intensities (particularly moderate and vigorous intensities), six reflected movement patterns, and one reflected a movement type. For the most mature participants, thirteen of the 39 movement behavior characteristics had important VIP value scores, with five reflecting intensities (particularly moderate and vigorous intensities), five reflecting patterns, and three reflecting types of movement. ConclusionMore than 12 movement behavior characteristics were associated with cardiometabolic risk factors within both the most and least mature participants. Movement intensities within the moderate and vigorous intensity ranges were the most consistent correlates of these risk factors.

Modeling growth and fermentation inhibition during bioethanol production using component profiles obtained by performing comprehensive targeted and non-targeted analyses

Corn cob and corn stover hydrolysates are forms of lignocellulosic biomass that can be used in second generation bioethanol production and biorefinery processes. Growth and fermentation inhibitors generated during physicochemical and enzymatic hydrolysis decrease ethanol and biomaterial production during the subsequent biological processes. Here, estimates of growth and fermentation inhibition during bioethanol fermentation were made using component profiles of corn cobs and corn stover at different degrees of hydrolysis. The component profiles were acquired by non-targeted gas chromatography mass spectrometry and targeted high-performance liquid chromatography. Correlations between the comprehensive analysis results and yeast growth and ethanol production were modeled very accurately by partial-least-squares regression analysis. Acetate, apocynin, butyrovanillone, furfural, furyl hydroxymethyl ketone, m-methoxyacetophenone, palmitic acid, syringaldehyde, and xylose, were compounds with very variable importance in projection values and had negative correlation coefficients in the model. In fact, methoxyacetophenone, apocynin, and syringaldehyde inhibited fermentation more than furfural in equivalent concentration.

Characterization and discrimination of selected China's domestic pork using an LC-MS-based lipidomics approach

A lipidomics study using liquid chromatography-tandem mass spectrometry and multivariate statistics was conducted in this work to discriminate raw pork meat. A total of 1180 lipid species were identified in the studied pork samples. Four, three and eight lipids were determined as potential discriminatory markers for the five cuts (shoulder, rump, loin, shank and belly) of Tibetan, Jilin and Sanmenxia black pigs, respectively. Distinct lipidomic fingerprints of Tibetan, Jilin and Sanmenxia pork were obtained and they were clearly separated into three clusters by partial least squares discriminant analysis (PLS-DA). The developed PLS-DA model (R2X = 0.603, R2Y = 0.861 and Q2 = 0.752) enables a 91.1% correct classification of pork samples. One-hundred variables, including 61 glycerolipids, 17 glycerophospholipids, 4 sterol lipids, 2 sphingolipids, 3 polyketides, 7 fatty acyls and 6 prenol lipids, were found to have high potential (variable importance in projection value > 1, p-value<0.05) to differentiate Tibetan, Jilin and Sanmenxia pork meat. The current data set will facilitate a better understanding of the nutritional values of the investigated pork and can be expanded to a larger sample size for lipid marker validation. Our findings demonstrate that lipidomic analysis together with multivariate statistics is a promising approach for the differentiation of China's domestic pork.

An Outbreak of Crayfish-Related Haff Disease in the Jiangsu Province of China: A Focus on the Metabolites

Background: Haff disease is a rare syndrome of unexplained myalgia and rhabdomyolysis occurring within 24 h of consumption of cooked fish or crustaceans, and the etiology remains unknown. Herein, we implemented a metabolomics strategy to investigate Haff disease-induced metabolite disorders. Methods: This Single, tertiary, academic medical center and Laboratory investigation included 73 crayfish-induced Haff disease-diagnosed case-patients and 73 non-crayfish diners. The serum metabolites of Haff disease patients (n = 60) were detected and analyzed with gas chromatography-mass spectrometry. The spectrum was deconvoluted, aligned, and identified using MS DIAL, equipped with FiehnLib. The data matrix was processed via univariate and multivariate analyses for the selection of metabolites with variable importance in projection values >1 and the t-test (p-value < 0.05). Results: In the case-patients, the concentrations of 20 metabolites were significantly increased and those of 6 metabolites decreased. The metabolic pathway analysis was analyzed with the MetaMapp program and visualized on CytoScape. Intriguingly, there was an enormous increase in the serum concentration of myristoleic acid in patients with Haff disease, indicating that myristoleic acid might be a key metabolite of Haff disease and a potential biomarker for crayfish or other seafood-related Haff disease. A further understanding of this disease would allow for the improved management of future disease outbreaks. Conclusions: The metabolites in the serum of the Haff disease patients was significantly disturbed, and some of which may be benefit for the diagnosis of this rare disease. Funding Statement: This work has been supported by the National Natural Science Foundation of China (81470150), the program for key diseases by the Jiangsu Province Science and Technology Department (BL2014088), and the National Natural Science Foundation of Jiangsu Province (BK20171490) Declaration of Interests: The authors declare that they have no competing interests. Ethics Approval Statement: This study was approved by the (Local) Ethics Committee at the Institute of the First Affiliated Hospital of Nanjing Medical University and the investigation was performed according to the ethical guidelines in Document No. 2018-SR-151.

Exploration of predictive metabolic factors for gestational diabetes mellitus in Japanese women using metabolomic analysis.

Aims/IntroductionWe aimed to explore novel predictive markers for gestational diabetes mellitus using metabolomic analysis in pregnant Japanese women.Materials and MethodsWe carried out a case–control study with a cohort of participants enrolled during the first or early second trimester in the Center of Chiba Unit of the Japan Environment and Children's Study. Participants were classified as either gestational diabetes mellitus cases or matched controls based on age, body mass index and parity. Metabolite levels of their serum and urine obtained randomly before the diagnosis of gestational diabetes mellitus were analyzed using hydrophilic interaction chromatography tandem mass spectrometry. Orthogonal projections to latent structures discriminant analysis was carried out to investigate metabolome profiles for the different groups. Metabolites with a variable importance in projection value of >1.5 were identified as potential markers.ResultsIn total, 242 participants were enrolled in the study, of which 121 were cases. The R2X, R2Y and Q2 parameters for the discrimination ability of the resulting models were 0.388, 0.492 and 0.45 for serum, and 0.454, 0.674 and 0.483 for urine, respectively. We finally identified three metabolites in serum and 20 in urine as potential biomarkers. Glutamine in serum and ethanolamine and 1,3‐diphosphoglycerate in urine showed >0.8 area under the receiver operating characteristic curves.ConclusionsThe present study identified serum and urine metabolites that are possible predictive markers of subsequent gestational diabetes mellitus in Japanese women. Further studies are required to elucidate their efficacy.

Comparative metabolic profiling of four transgenic maize lines and two non-transgenic maize lines using high-performance liquid chromatography mass spectrometry

We compared metabolic changes in transgenic maize lines (differing in genetic backgrounds and exogenous genes insertion) with non-transgenic counterparts. At the corn seedling stage, entirely expanding leaves were obtained from maize lines 1 (transgenic maize SK12-5; overexpressing cry1Ab/cry2Aj; parental maize ZD958), 2 (transgenic maize IE034; overexpressing cry1Ie; parental maize ZD958), 3 (transgenic maize Bt799; overexpressing cry1Ac-M; parental maize Z58), 4 (transgenic maize Bt799; overexpressing cry1Ac-M; parental maize ZD958), 5 (non-transgenic maize Z58), and 6 (non-transgenic maize ZD958). For each line, six biological replicates were prepared; each contained an entirely expanded leaf at the top of the plant. Metabolites were identified with high-performance liquid chromatography mass spectrometry in three comparison groups (4 transgenic maize lines vs. their corresponding controls, line 3 vs. 4, and transgenic vs. non-transgenic maize). Top 200 cationic and top 200 anionic metabolites with higher variable importance in projection value were chosen for principal component analysis, followed by pathway enrichment analysis. Annotation was possible for 227 of 400 metabolites. Twelve up- and 26 down-regulated overlapping metabolites were identified in line 1 vs. 6, line 2 vs. 6, and line 4 vs. 6 comparison groups. Altered metabolites significantly enriched in purine and glutathione metabolism pathways, but no pathways were enriched in line 3 vs. 4. Comparing transgenic and non-transgenic maize lines revealed 59 up- and 37 down-regulated metabolites, which were also significantly enriched in purine and glutathione metabolism pathways. Therefore, transgenic Bt maize differed metabolically from non-transgenic maize. Additionally, purine and glutathione metabolism pathways may be important in the transgenic maize metabolism.

Quantification winter wheat LAI with HJ-1CCD image features over multiple growing seasons

Remote sensing images are widely used to map leaf area index (LAI) continuously over landscape. The objective of this study is to explore the ideal image features from Chinese HJ-1 A/B CCD images for estimating winter wheat LAI in Beijing. Image features were extracted from such images over four seasons of winter wheat growth, including five vegetation indices (VIs), principal components (PC), tasseled cap transformations (TCT) and texture parameters. The LAI was significantly correlated with the near-infrared reflectance band, five VIs [normalized difference vegetation index, enhanced vegetation index (EVI), modified nonlinear vegetation index (MNLI), optimization of soil-adjusted vegetation index, and ratio vegetation index], the first principal component (PC1) and the second TCT component (TCT2). However, these image features cannot significantly improve the estimation accuracy of winter wheat LAI in conjunction with eight texture measures. To determine the few ideal features with the best estimation accuracy, partial least squares regression (PLSR) and variable importance in projection (VIP) were applied to predict LAI values. Four remote sensing features (TCT2, PC1, MNLI and EVI) were chosen based on VIP values. The result of leave-one-out cross-validation demonstrated that the PLSR model based on these four features produced better result than the ten features’ model, throughout the whole growing season. The results of this study suggest that selecting a few ideal image features is sufficient for LAI estimation.

Prediction of placental barrier permeability: a model based on partial least squares variable selection procedure.

Assessing the human placental barrier permeability of drugs is very important to guarantee drug safety during pregnancy. Quantitative structure–activity relationship (QSAR) method was used as an effective assessing tool for the placental transfer study of drugs, while in vitro human placental perfusion is the most widely used method. In this study, the partial least squares (PLS) variable selection and modeling procedure was used to pick out optimal descriptors from a pool of 620 descriptors of 65 compounds and to simultaneously develop a QSAR model between the descriptors and the placental barrier permeability expressed by the clearance indices (CI). The model was subjected to internal validation by cross-validation and y-randomization and to external validation by predicting CI values of 19 compounds. It was shown that the model developed is robust and has a good predictive potential (r2 = 0.9064, RMSE = 0.09, q2 = 0.7323, rp2 = 0.7656, RMSP = 0.14). The mechanistic interpretation of the final model was given by the high variable importance in projection values of descriptors. Using PLS procedure, we can rapidly and effectively select optimal descriptors and thus construct a model with good stability and predictability. This analysis can provide an effective tool for the high-throughput screening of the placental barrier permeability of drugs.

Exploring the Best Hyperspectral Features for LAI Estimation Using Partial Least Squares Regression

The use of spectral features to estimate leaf area index (LAI) is generally considered a challenging task for hyperspectral data. In this study, the hyperspectral reflectance of winter wheat was selected to optimize the selection of spectral features and to evaluate their performance in modeling LAI at various growth stages during 2008 and 2009. We extracted hyperspectral features using different techniques, including reflectance spectra and first derivative spectra, absorption and reflectance position and vegetation indices. In order to find the best subset of features with the best predictive accuracy, partial least squares regression (PLSR) and variable importance in projection (VIP) were applied to estimated LAI values. The results indicated that the red edge–NIR spectral region (680 nm–1300 nm) was the most sensitive to LAI. Most features in this region exhibited a high correlation with LAI and had higher VIP values, especially the first derivative waveband at 750 nm (r = 0.900, VIP = 1.144). Adding a large number of features would not significantly improve the accuracy of the PLSR model. The PLSR model based on the fourteen features with the highest VIP values predicted LAI with a mean bootstrapped R2 value of 0.880 and a mean RMSE of 0.943 on the validation dataset and produced an estimated LAI result better than that, including the entire 54-feature dataset with a mean R2 of 0.875 and a mean RMSE of 0.965. The results of this study thus suggest that the use of only a few of the best features by VIP values is sufficient for LAI estimation.

Global metabolite profiling of synovial fluid for the specific diagnosis of rheumatoid arthritis from other inflammatory arthritis.

Currently, reliable biomarkers that can be used to distinguish rheumatoid arthritis (RA) from other inflammatory diseases are unavailable. To find possible distinctive metabolic patterns and biomarker candidates for RA, we performed global metabolite profiling of synovial fluid samples. Synovial fluid samples from 38 patients with RA, ankylosing spondylitis, Behçet's disease, and gout were analyzed by gas chromatography/time-of-flight mass spectrometry (GC/TOF MS). Orthogonal partial least-squares discriminant and hierarchical clustering analyses were performed for the discrimination of RA and non-RA groups. Variable importance for projection values were determined, and the Wilcoxon-Mann-Whitney test and the breakdown and one-way analysis of variance were conducted to identify potential biomarkers for RA. A total of 105 metabolites were identified from synovial fluid samples. The score plot of orthogonal partial least squares discriminant analysis showed significant discrimination between the RA and non-RA groups. The 20 metabolites, including citrulline, succinate, glutamine, octadecanol, isopalmitic acid, and glycerol, were identified as potential biomarkers for RA. These metabolites were found to be associated with the urea and TCA cycles as well as fatty acid and amino acid metabolism. The metabolomic analysis results demonstrated that global metabolite profiling by GC/TOF MS might be a useful tool for the effective diagnosis and further understanding of RA.

FRI0511 Global metabolite profiling of synovial fluids from different forms of inflammatory arthritis for the identification of putative biomarker in rheumatoid arthritis

BackgroundMetabolomics is the study of unique chemical fingerprints of specific cellular processes. Metabolite profiling is recently applied in identifying biomarkers in medical research including rheumatologic diseases.ObjectivesThe aim is to assess...

FRI0515 Discriminative metabolite profiling of synovial fluid in rheumatoid arthritis compared to osteoarthritis

BackgroundMetabolomics is the study of unique chemical imprints that represent specific cellular processes in a cell, tissue, organ or organism. Synovial fluid in pathologic conditions reflects the diseased process and...

Assessing feature relevance in NPLS models by VIP

Multilinear PLS (NPLS) and its discriminant version (NPLS-DA) are very diffuse tools to model multi-way data arrays. Analysis of NPLS weights and NPLS regression coefficients allows data patterns, feature correlation and covariance structure to be depicted. In this study we propose an extension of the Variable Importance in Projection (VIP) parameter to multi-way arrays in order to highlight the most relevant features to predict the studied dependent properties either for interpretative purposes or to operate feature selection. The VIPs are implemented for each mode of the data array and in the case of multivariate dependent responses considering both the cases of expressing VIP with respect to each single y-variable and of taking into account all y-variables altogether.Three different applications to real data are presented: i) NPLS has been used to model the properties of bread loaves from near infrared spectra of dough, acquired at different leavening times, and corresponding to different flour formulations. VIP values were used to assess the spectral regions mainly involved in determining flour performance; ii) assessing the authenticity of extra virgin olive oils by NPLS-DA elaboration of gas chromatography/mass spectrometry data (GC–MS). VIP values were used to assess both GC and MS discriminant features; iii) NPLS analysis of a fMRI-BOLD experiment based on a pain paradigm of acute prolonged pain in healthy volunteers, in order to reproduce efficiently the corresponding psychophysical pain profiles. VIP values were used to identify the brain regions mainly involved in determining the pain intensity profile.

PGI chicory (Cichorium intybus L.) traceability by means of HRMAS-NMR spectroscopy: a preliminary study.

Analytical traceability of PGI and PDO foods (Protected Geographical Indication and Protected Denomination Origin respectively) is one of the most challenging tasks of current applied research. Here we proposed a metabolomic approach based on the combination of (1)H high-resolution magic angle spinning-nuclear magnetic resonance (HRMAS-NMR) spectroscopy with multivariate analysis, i.e. PLS-DA, as a reliable tool for the traceability of Italian PGI chicories (Cichorium intybus L.), i.e. Radicchio Rosso di Treviso and Radicchio Variegato di Castelfranco, also known as red and red-spotted, respectively. The metabolic profile was gained by means of HRMAS-NMR, and multivariate data analysis allowed us to build statistical models capable of providing clear discrimination among the two varieties and classification according to the geographical origin. Based on Variable Importance in Projection values, the molecular markers for classifying the different types of red chicories analysed were found accounting for both the cultivar and the place of origin.

HRMAS-NMR spectroscopy and multivariate analysis meat characterisation

1H-high resolution magic angle spinning-nuclear magnetic resonance spectroscopy was employed to gain the metabolic profile of longissimus dorsi and semitendinosus muscles of four different breeds: Chianina, Holstein Friesian, Maremmana and Buffalo.Principal component analysis, partial least squares projection to latent structure – discriminant analysis and orthogonal partial least squares projection to latent structure – discriminant analysis were used to build models capable of discriminating the muscle type according to the breed. Data analysis led to an excellent classification for Buffalo and Chianina, while for Holstein Friesian the separation was lower. In the case of Maremmana the use of intelligent bucketing was necessary due to some resonances shifting allowed improvement of the discrimination ability. Finally, by using the Variable Importance in Projection values the metabolites relevant for the classification were identified.

Prediction of the ethanol yield of dry-grind maize grain using near infrared spectroscopy

The primary objective was to investigate the potential of using dry-grind maize spectral data for predicting ethanol yield, as well as to identity important wavelengths related to ethanol yield production. Spectral data from 222 ground maize grain samples were obtained over the spectral region of 400–2498 nm in 2 nm intervals. Thirty replicate runs were conducted, and for each run, 166 out of 222 samples were randomly selected for calibration, and the remaining 56 samples were used for validation. Partial least square regressions (PLSRs) were conducted on the complete spectra and on the wavelengths selected by bootstrapping based on the computed variable importance for projection values (Boot_VIP). Models with the wavelengths selected by the Boot_VIP and models with the complete spectra had similar prediction capabilities in the independent validation, with the average root mean square error of prediction (RMSEP) of 0.56%, which is comparable to the standard deviation of the dry fermentation reference method. The Boot_VIP procedure selected wavelengths consistently in 30 replicate runs, averaging 109 wavelengths selected at each run, and there were 101 wavelengths selected for all 30 runs. The selected wavelengths concentrated on 400–550 nm of the visible region and 2300–2360 nm of the NIR region. The first two factors from the Boot_VIP models were almost identical to the respective first two factors from models with the complete spectra.