scholarly journals Applying Discriminant and Cluster Analyses to Separate Allergenic from Non-allergenic Proteins

2019 ◽  
Vol 17 (1) ◽  
pp. 401-407 ◽  
Author(s):  
L. Naneva ◽  
M. Nedyalkova ◽  
S. Madurga ◽  
F. Mas ◽  
V. Simeonov

AbstractAs a result of increased healthcare requirements and the introduction of genetically modified foods, the problem of allergies is becoming a growing health problem. The concept of allergies has prompted the use of new methods such as genomics and proteomics to uncover the nature of allergies. In the present study, a selection of 1400 food proteins was analysed by PLS-DA (Partial Least Square-based Discriminant Analysis) after suitable transformation of structural parameters into uniform vectors. Then, the resulting strings of different length were converted into vectors with equal length by Auto and Cross-Covariance (ACC) analysis. Hierarchical and non-hierarchical (K-means) Cluster Analysis (CA) was also performed in order to reach a certain level of separation within a small training set of plant proteins (16 allergenic and 16 non-allergenic) using a new three-dimensional descriptor based on surface protein properties in combination with amino acid hydrophobicity scales. The novelty of the approach in protein differentiation into allergenic and non-allergenic classes is described in the article.The general goal of the present study was to show the effectiveness of a traditional chemometric method for classification (PLS–DA) and the options of Cluster Analysis (CA) to separate by multivariate statistical methods allergenic from non-allergenic proteins.

Author(s):  
Bartlomiej Blachowski

The present study deals with a comprehensive approach for damage identification of spatial truss structures. The novelty of the proposed approach consists of a three-level analysis. First, sensitivity of assumed modal characteristics is calculated. Second, natural frequency sensitivity is used to determine hardly identifiable structural parameters and mode shape sensitivity is applied to select damage-sensitive locations of sensors. Third, two sparsity constrained optimization algorithms are tested towards efficient identification of applied damage scenarios. These two algorithms are based on ℓ1-norm minimization and non-negative least square (NNLS) solution.Performances of both proposed algorithms have been compared in two realistic case studies: the first one concerned a three-dimensional truss girder with 61 structural parameters and the second one was devoted to an upper-deck arch bridge composed of 416 steel members.


2019 ◽  
Vol 14 (1) ◽  
Author(s):  
Zhaoyan Zhang ◽  
Liang Yang ◽  
Xiaoyan Huang ◽  
Yue Gao

Abstract Background The side effects caused by Polygoni Multiflori Radix (PMR) and Polygoni Multiflori Radix Praeparata (PMRP) have often appeared globally. There is no research on the changes of endogenous metabolites among PMR- and PMRP-treated rats. The aim of this study was to evaluate the varying metabolomic effects between PMR- and PMRP-treated rats. We tried to discover relevant differences in biomarkers and endogenous metabolic pathways. Methods Hematoxylin and eosin staining and immunohistochemistry staining were performed to find pathological changes. Biochemical indicators were also measured, one-way analysis of variance with Dunnett’s multiple comparison test was used for biochemical indicators comparison among various groups. Metabolomics analysis based on ultra-high performance liquid chromatography-quadrupole time of flight mass spectrometry (UPLC-Q/TOF-MS) was performed to find the changes in metabolic biomarkers. Multivariate statistical approaches such as principal component analysis (PCA) and orthogonal partial least square-discriminant analysis (OPLS-DA) were applied to reveal group clustering trend, evaluate and maximize the discrimination between the two groups. MetaboAnalyst 4.0 was performed to find and confirm the pathways. Results PMR extracts exhibited slight hepatotoxic effects on the liver by increasing aspartate and alanine aminotransferase levels. Twenty-nine metabolites were identified as biomarkers, belonging to five pathways, including alpha-linolenic acid metabolism, taurine and hypotaurine metabolism, glycerophospholipid metabolism, arginine and proline metabolism, and primary bile acid biosynthesis. Conclusion This study provided a comprehensive description of metabolomic changes between PMR- and PMRP-treated rats. The underlying mechanisms require further research.


IAWA Journal ◽  
2019 ◽  
Vol 40 (1) ◽  
pp. 58-74 ◽  
Author(s):  
Maomao Zhang ◽  
Guang Jie Zhao ◽  
Bo Liu ◽  
Tuo He ◽  
Juan Guo ◽  
...  

ABSTRACT Pterocarpus santalinus, listed in CITES Appendix II, is an endangered timber species as a result of illegal harvesting due to its high value and commercial demand. The growing demand for P. santalinus and timbers with the morphologically similar Pterocarpus tinctorius has resulted in confusion as well as identification problems. Therefore, it is of vital importance to explore reliable ways to accurately discriminate between P. santalinus and P. tinctorius. In this study, the method of direct analysis in real time and fourier transform ion cyclotron resonance mass spectrometry (DART-FTICR-MS), combined with multivariate statistical analysis, was used to extract chemical information from xylarium wood specimens and to explore the feasibility of distinguishing these two species. Significant differences were observed in their DART-FTICR-MS spectra. Orthogonal partial least square-discriminant analysis (OPLS-DA) showed the highest prediction, with an accuracy of 100%. These findings demonstrate the feasibility of authenticating wood types using DART-FTICR-MS coupled with multivariate statistical analysis.


2012 ◽  
Vol 457-458 ◽  
pp. 46-49
Author(s):  
Xue Jun Zhu ◽  
Zhi Wen Zhu

In this paper, a kind of giant magnetostrictive material (GMM) model based on hysteretic nonlinear theory was developed. Van de Pol nonlinear difference item was introduced to interpret the hysteresis phenomenon of the strain-magnetic field intensity (MFI) curve of GMM. The coupling relationship between MFI and frequency was obtained in partial least-square regression method to describe the driftage phenomenon of the strain-MFI curves of GMM in different frequencies. Based on above, the final relationship among strain, MFI and frequency was set up. The result of significance test shows that the effects of all of the items in the final model are remarkable, and that of forecast test shows that the model can describe the characteristics of GMA in different frequencies well. The new GMM model broadens the region of controlling frequency, and is easy to be analyzed in theory, which is helpful to vibration control.


2019 ◽  
Vol 20 (1) ◽  
pp. 77
Author(s):  
Kukuh Winarso ◽  
Moh. Jufriyanto

PT. X Provides paid internet services known as Indihome. The problem that occurs in Indihome Internet service is that there are 42.36% complaints from consumers to the service process provided by PT. X and there is a 7.46% complaint about the amount of the bill. The purpose of this study is to model and know what variables affect the quality of the Indihome Internet service. The method used is Partial Least Square (PLS). Partial Least Square (PLS) is one of the multivariate statistical analysis techniques that can handle multiple response variables and an explanatory variable. The variables used are tangible, empathy, assurance, reliability and responsiveness. The results of the research are the increase of R square value after the implementation of the A5 indicator that does not meet the model PLS. All five variables affect the quality of indihome Internet service is 76,1%. Based on the hypothesis test results on the outer model It is said that the variable indicator is valid with p-value = 0 < 0,05. In testing the inner model obtained that the variable tangible and responsiveness has no significant effect on the quality of Internet service Indihome, while the variable empathy, assurance, reliability significantly affect the quality of service Indihome Internet.


Author(s):  
Serge Nabeneza ◽  
Vincent Porphyre ◽  
Fabrice Davrieux

Le miel est traditionnellement consommé dans les îles de l’océan Indien et utilisé également pour ses propriétés cosmétiques et thérapeutiques. Ce produit, issu du nectar de fleurs ou de miellat d’insectes, est collecté et transformé par les abeilles produisant ainsi un miel unique caractéristique de la flore de chaque île. A l’issue de la récolte, l’apiculteur doit nor­malement veiller à ce que le miel soit conforme et respecte des caractéristiques physico-chimiques spécifiques selon les recom­mandations du Codex alimentarius. De plus, il doit indiquer des informations, comme l’origine botanique et géographique, sur chaque pot de miel vendu. En pratique il est toutefois très diffi­cile de vérifier l’exactitude de ces informations, d’autant qu’elles sont mises en avant comme argument de vente.La spectrométrie dans le proche infrarouge (SPIR), basée sur les propriétés physiques d’absorption de la lumière par les com­posés organiques, peut être un outil de contrôle et de traçabi­lité des miels mis sur le marché (1, 3). L’objectif principal de cette étude a été d’évaluer la faisabilité d’un contrôle qualité (authentification de l’origine botanique et/ou géographique et détection d’adultération) des miels commerciaux produits dans l’océan Indien. Les perspectives de cette méthode de laboratoire sont importantes dans le cadre du développement des filières apicoles, notamment lorsqu’elles sont orientées vers les marchés d’exportation et vers la promotion de leur qualité auprès de consommateurs désireux de profiter de produits d’excellence et fortement attachés à des territoires insulaires uniques.Au total, 625 miels des pays de l’océan Indien (tableau I) ayant diverses origines botaniques (tableau II) ont été mesurés sans ajustement à 30 °C et après ajustement à 70 °Brix avec de l’eau distillée (3). La mesure a été réalisée en réflexion diffuse à l’aide d’une cellule ronde avec un fond réflecteur doré (trajet optique 0,1 mm) sur un spectromètre FOSS NIRSystem 5000 (1 100–2 500 nm, 2 nm). En parallèle, les mesures du degré Brix, de l’humidité et de la conductivité des miels bruts ont été réalisées au laboratoire. Après l’analyse en composantes prin­cipales (ACP) de la base spectrale des 625 miels et le calcul des distances H de Mahalanobis, 83 miels ont été identifiés comme spectralement atypiques et retirés du jeu de données (tableau I). Sur la base des distances spectrales (H, Mahalanobis), 64 miels représentatifs des 542 miels retenus ont été sélectionnés, ajustés à 70 °Brix, puis adultérés avec du sirop de sucre de canne du commerce (Mascarin®) à 25 et à 10 p. 100 par pesée.Les modèles prédictifs des teneurs en humidité, Brix et conducti­vité ont été établis en utilisant la régression modified partial least square (M-PLS) du logiciel WinISI 3 (Infrasoft, Port Mathilda, PA, Etats-Unis). Les modèles PLS ont été réalisés sur la base des dérivées premières des spectres normalisés (SNV) et corri­gés pour la ligne de base (Detrend). Pour expliquer les critères géographiques et botaniques, nous avons utilisé une méthode non supervisée de discrimination (Cluster Analysis, Unscrumbler 10.3, CAMO, Oslo, Norvège), qui constitue des groupements naturels, sur la base des distances spectrales. Les spectres des miels non adultérés ont servi de base pour définir un espace multidimensionnel (ACP centrée, non réduite sur dérivée seconde des spectres bruts). La projection des spectres (dérivée seconde des spectres bruts) des miels adultérés sur cet espace a permis d’identifier un plan factoriel (CP1–CP7) expliquant 65 p. 100 de la variance totale, sur lequel les miels adultérés étaient nettement séparés des miels non adultérés. Les modèles d’étalonnage obtenus par régression PLS pour le degré Brix, l’humidité et la conductivité ont été performants avec des R² de l’ordre de 0,92 et des erreurs de validation croisée acceptables (tableau III) (figure 1).Les essais de groupement des individus sur la base de leur spectre (recherche de clusters), de même que l’utilisation d’autres méthodes mathématiques, comme l’analyse discrimi­nante linéaire (LDA) et la classification support vecteur machine (SVM) à partir des origines géographiques et botaniques n’ont pas donné de résultats satisfaisants (tableau IV). Cette impossi­bilité de discriminer les miels de l’océan Indien serait peut-être due à une identification botanique imprécise des miels. Sans analyses polliniques des échantillons de miels, les origines bota­niques ne peuvent être certifiées, en particulier les miels dits monofloraux. Plusieurs hypothèses peuvent expliquer l’impos­sibilité de discriminer les miels de l’océan Indien en fonction de leur origine géographique et/ou botanique. Parmi elles, les différences chimiques caractéristiques de l’origine géographique et/ou botanique correspondent à des classes de composés (aro­matiques par exemple) qui n’impactent pas suffisamment les empreintes spectrales, ne permettant pas de trouver des fac­teurs discriminants. La projection des miels adultérés à 25 p. 100 (figure 2), sur la base de miels non adultérés montre une sépara­tion nette des populations. En revanche, pour un niveau d’adul­tération à 10 p. 100, la séparation n’est pas évidente (figure 3).Les résultats obtenus montrent que la spectrométrie dans le proche infrarouge peut être utilisée pour prédire certains para­mètres physico-chimiques des miels. Ces premiers résultats montrent que la constitution d’une base spectrale de miels adultérés à différents niveaux au laboratoire, associée à des méthodes de régression ou de discrimination devrait permettre l’identification de miels non-conformes, comme montré précé­demment par Rios-Corripio et coll. (2).Ces résultats ne permettent pas, dans l’état actuel, de différencier l’origine géographique et botanique des miels collectés dans le commerce. La limite de détection des miels adultérés par du sucre de canne devra être déterminée. Elle se trouve entre 10 et 25 p. 100. Les fraudeurs adultérant fréquemment les miels avec du glucose, il serait important de poursuivre les essais avec ce produit. Il est donc possible de détecter, par la technique SPIR, des altérations importantes du miel par ajout de sucre de canne.


2018 ◽  
Vol 1 (1) ◽  
pp. 10-19 ◽  
Author(s):  
Ana Hadiana

Users’ psychological emotion plays important role in designing an interface of software including application of information system. This research attempted to implement Kansei Engineering Type I (KEPack) as a method to analyze kinds of emotional factor related to user interface for mobile Parental Information System. This research used Kansei Words to explore users’ requirements based on psychological factors. Eighteen words were used for Kansei Words that have relationship with Parental Information System. Ten samples of mobile information system were selected as specimens considered suitable for designing interface of Parental Information System. Data questionnaires collected from thirty respondents were processed using multivariate statistical analysis such as Factor Analysis (FA) and Partial Least Square (PLS). This research found that the two important emotional factors i.e funny and informative have to be considered for designing user interface for mobile Parental Information System.


2012 ◽  
Vol 554-556 ◽  
pp. 1289-1292
Author(s):  
Neng Sheng Ye ◽  
Ya Li Xie ◽  
Chang Liu ◽  
Jian Li

A micellar electrokinetic electrophoresis-laser induced fluorescence detection (MEKC-LIF) method was applied for the profiling of native fluorescence in three different kinds of Chinese green teas. The running buffer was 100 mM sodium borate (pH9.8) containing 20 mM SDS, and tea infusion was injected at the pressure of 0.5 psi for 5 seconds. The applied voltage was set at 25 kV, and the excited wavelength was 488 nm, and the detected wavelength was 520 nm. Ten tea samples were collected and analyzed by the developed method. Eighteen common peaks in the MEKC-LIF profiling of green teas were selected and these samples were classified into 3 groups by cluster analysis and partial least square discriminant analysis (PLS-DA). Our primary results indicated that the developed method could be applied for the classification of three different Chinese green teas


2011 ◽  
Vol 356-360 ◽  
pp. 89-94
Author(s):  
Kun Xie ◽  
Shu Qiao ◽  
Chuan Fu ◽  
Jun Sheng Qi

Polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) are a group of important persistent organic pollutants. In the present study, three-dimensional holographic vector of atomic interaction field (3D-HoVAIF) is used to describe the chemical structures of PCDD/Fs. After variable screening by stepwise multiple regression (SMR) technique, the liner relationships between solid vapor pressure (logPS) of PCDD/Fs and 3D-HoVAIF descriptors by partial least square (PLS) regression.The result shows that the 3D-HoVAIF descriptors can be well used to express the quantitative structure-property relationships of PCDD/Fs. Predictive capability of the models has also been demonstrated by leave-one-out cross-validation. Moreover, the estimated values have been presented for those PCDD/Fs which are lack of experimentally data by the optimum model.


Author(s):  
Mahadzirah Mohamad ◽  
Asyraf Afthanorhan ◽  
Zainudin Awang ◽  
Morliyati Mohammad

Structural Equation Modelling (SEM) is a powerful multivariate statistical analysis technique which combines both factor analysis and multiple regression analysis. It is capable of analysing the inter-relationships among latent constructs simultaneously in a model. These latent constructs are measured using certain number of items in a questionnaire. Covariance-based SEM (CB-SEM) or full SEM have become the choice for many researchers in a variety of disciplines because of their ability to evaluate complex relationships using parametric statistical approach. Researchers could also opt for Variance-based SEM (VB-SEM) or Partial Least Square-SEM (PLS-SEM) when their data failed the parametric assumptions such as multivariate normality distribution and minimum sample size. However, the approach of VB-SEM or PLS-SEM is a non-parametric instead of a parametric approach in CB-SEM. This article compared the performance of both SEM approaches using the same dataset to validate the Measurement Model for Maqasid Syariah Quality of Life (MSQoL). The findings of both analyses suggested that CB-SEM or full SEM is more appropriate to validate and confirm the MSQoL measurement model.


Sign in / Sign up

Export Citation Format

Share Document