Metabolomics Combined with Correlation Analysis Revealed the Differences in Antioxidant Activities of Lotus Seeds with Varied Cultivars.

Functional foods have potential health benefits for humans. Lotus seeds (LS) as functional foods have excellent antioxidant activities. However, the differences in chemical composition of different LS cultivars may affect their antioxidant activities. This study comprehensively analyzed the differences among five LS cultivars based on metabolomics and further revealed the effects of metabolites on antioxidant activities by correlation analysis. A total of 125 metabolites were identified in LS using UPLC-Q/TOF-MS. Then, 15 metabolites were screened as differential metabolites of different LS cultivars by chemometrics. The antioxidant activities of LS were evaluated by DPPH•, FRAP, and ABTS•+ assays. The antioxidant activities varied among different LS cultivars, with the cultivar Taikong 66 showing the highest antioxidant activities. The correlation analysis among metabolites and antioxidant activities highlighted the important contribution of phenolics and alkaloids to the antioxidant activities of LS. Particularly, 11 metabolites such as p-coumaric acid showed significant positive correlation with antioxidant activities. Notably, 6 differential metabolites screened in different LS cultivars showed significant effects on antioxidant activities. These results revealed the important effects of phytochemicals on the antioxidant activities of different LS cultivars. This study provided evidence for the health benefits of different LS cultivars.

NIR Spectrometric Approach for Geographical Origin Identification and Taste Related Compounds Content Prediction of Lushan Yunwu Tea.

Lushan Yunwu Tea is one of a unique Chinese tea series, and total polyphenols (TP), free amino acids (FAA), and polyphenols-to-amino acids ratio models (TP/FAA) represent its most important taste-related indicators. In this work, a feasibility study was proposed to simultaneously predict the authenticity identification and taste-related indicators of Lushan Yunwu tea, using near-infrared spectroscopy combined with multivariate analysis. Different waveband selections and spectral pre-processing methods were compared during the discriminant analysis (DA) and partial least squares (PLS) model-building process. The DA model achieved optimal performance in distinguishing Lushan Yunwu tea from other non-Lushan Yunwu teas, with a correct classification rate of up to 100%. The synergy interval partial least squares (siPLS) and backward interval partial least squares (biPLS) algorithms showed considerable advantages in improving the prediction performance of TP, FAA, and TP/FAA. The siPLS algorithms achieved the best prediction results for TP (RP = 0.9407, RPD = 3.00), FAA (RP = 0.9110, RPD = 2.21) and TP/FAA (RP = 0.9377, RPD = 2.90). These results indicated that NIR spectroscopy was a useful and low-cost tool by which to offer definitive quantitative and qualitative analysis for Lushan Yunwu tea.

Fast quantification of total volatile basic nitrogen (TVB-N) content in beef and pork by near-infrared spectroscopy: Comparison of SVR and PLS model.

With application of PLS regression and SVR, quantitation models of near infrared diffuse reflectance spectroscopy were established for the first time to predict the content of volatile basic nitrogen (TVB-N) content in beef and pork. Results indicated that the best PLS model based on the raw spectra showed an excellent prediction performance with a high value of correlation coefficient at 0.9366 and a low root-mean-square error of prediction value of 3.15, and none of those pretreatment methods could improve the prediction performance of the PLS model. Moreover, comparatively the model obtained by SVR showed inferior quantitative predictive ability (R = 0.8314, RMSEP = 4.61). Analysis on VIP selected wavelengths inferred amino bond containing compounds and lipid may play important roles in the development of PLS models for TVB-N. Results from this study demonstrated the potential of using NIR spectroscopy and PLS for the prediction of TVB-N in beef and pork while more efforts are required to improve the performance of SVR models.

Comparison of Furans Formation and Volatile Aldehydes Profiles of Four Different Vegetable Oils During Thermal Oxidation.

The evolution of volatile aldehydes and the conversion of oxygenated ityß-unsaturated aldehydes (OαßUAs) into furans were compared in four vegetable oils (soybean oil, olive oil [OVO], peanut oil [PO], and perilla oil [PAO]) thermally oxidized at temperatures of 150, 180, and 210 °C for 10 hr/day over a 3-day period. Results showed that 2 alkyl furans and 23 volatile aldehydes including 4 toxic OdßUAs were detected by GC-MS. The original fatty acid compositions of the oils played a key role in the type and concentration of those volatile compounds. 4-Hydroxy-2-hexenal (HHE) and ethyl furan were only detected in PAO with a high content of linolenic acid, while the greatest level of pentyl furan was detected in PO with abundant linoleic acid. Greater amounts of 4-hydroxy-(E)-2-nonenal (HNE) and 4-oxo-(E)-2-nonenal (ONE) were formed in the OVO with abundant oleic acid. The close relativity of HHE and ethyl furan was also demonstrated. With principal component analysis, these vegetable oils could be discriminated based on their fatty acids and volatile compounds. The loading plot confirmed that HHE and ethyl furan were derived from the linolenic acid oxidation and degradation. PRACTICAL APPLICATION: The chemometric results showed that the formation of the volatile components during heating in different vegetable oils has close correlation with the original fatty acids composition of vegetable oils. Our research has also confirmed the presence of toxic OɑßUAs in oils after heating. Considering that they are proven to generate lots of degenerative diseases, further studies are needed to establish the risk level of using certain oils in frying and seek effective methods to inhibit their formation.

Prediction of fatty acid composition in camellia oil by 1H NMR combined with PLS regression.

A rapid method for the determination of fatty acid (FA) composition in camellia oils was developed based on the 1H NMR technique combined with partial least squares (PLS) method. Outliers detection, LVs optimization and data pre-processing selection were explored during the model building process. The results showed the optimal models for predicting the content of C18:1, C18:2, C18:3, saturated, unsaturated, monounsaturated and polyunsaturated FA were achieved by Pareto scaling (Par) pretreatment, with correlation coefficient (R2) above 0.99, the root mean square error of estimation and prediction (RMSEE, RMSEP) lower than 0.954 and 0.947, respectively. Mean-centering (Ctr) was more suitable for the model of C16:0 and C18:0 with the best performance indicators (R2 ≥ 0.945, RMSEE ≤ 0.377, RMSEP ≤ 0.212). This study indicated that 1H NMR has the potential to be applied as a rapid and routine method for the analysis of FA composition in camellia oils.

Variational AutoEncoder For Regression: Application to Brain Aging Analysis.

While unsupervised variational autoencoders (VAE) have become a powerful tool in neuroimage analysis, their application to supervised learning is under-explored. We aim to close this gap by proposing a unified probabilistic model for learning the latent space of imaging data and performing supervised regression. Based on recent advances in learning disentangled representations, the novel generative process explicitly models the conditional distribution of latent representations with respect to the regression target variable. Performing a variational inference procedure on this model leads to joint regularization between the VAE and a neural-network regressor. In predicting the age of 245 subjects from their structural Magnetic Resonance (MR) images, our model is more accurate than state-of-the-art methods when applied to either region-of-interest (ROI) measurements or raw 3D volume images. More importantly, unlike simple feed-forward neural-networks, disentanglement of age in latent representations allows for intuitive interpretation of the structural developmental patterns of the human brain.

Data Augmentation Based on Substituting Regional MRIs Volume Scores.

Due to difficulties in collecting sufficient training data, recent advances in neural-network-based methods have not been fully explored in the analysis of brain Magnetic Resonance Imaging (MRI). A possible solution to the limited-data issue is to augment the training set with synthetically generated data. In this paper, we propose a data augmentation strategy based on regional feature substitution. We demonstrate the advantages of this strategy with respect to training a simple neural-network-based classifier in predicting when individual youth transition from no-to-low to medium-to-heavy alcohol drinkers solely based on their volumetric MRI measurements. Based on 20-fold cross-validation, we generate more than one million synthetic samples from less than 500 subjects for each training run. The classifier achieves an accuracy of 74.1% in correctly distinguishing non-drinkers from drinkers at baseline and a 43.2% weighted accuracy in predicting the transition over a three year period (5-group classification task). Both accuracy scores are significantly better than training the classifier on the original dataset.