The role of permutation tests in exploratory multivariate data analysis

In this paper, we propose tools for studying biophysical characteristics of a person by processing polygraph charts. We present the results produced by two methods of multivariate data analysis: the principal component analysis (PCA) and projection on latent structures (PLS). Data of biophysical state of a person during the polygraph testing - galvanic skin response (GSP), photoplethysmogram, blood pressure, and other biophysical characteristics – are processed and analyzed.It is shown that load diagrams are the preferable method to identify the relationship between the biophysical characteristics of a person. The principal component analysis provides an opportunity to assess minor changes in biophysical characteristics of a person and to take into account situational significant identifiers. They can attest to the fact that a possibly significant physical-chemical process runs in a human body.It is demonstrated that the projection on latent structures should be used to develop regression models that can constantly be used in polygraph testing.It is concluded that proposed methods of multivariate data analysis can help in an investigation of biophysical characteristics of a person using polygraph data. Also, the proposed methods are able to detect possible malfunctions in the responsible physical-chemical processes. The obtained results enable solving practical problems of information security matters, for example, the identification of an insider.DOI 10.14258/izvasu(2018)1-05

Multivariate data analysis for Raman imaging of a model pharmaceutical tablet

The performance of neural networks in classifying mass spectral data is evaluated and compared to methods of multivariate data analysis and pattern recognition. Back propagation networks are matched with linear discriminant analysis, Kohonen feature maps are compared to the knearest neighbour clustering algorithm. Eight classifiers were trained, in order to discriminate mass spectra of steroids from eight distinct classes of chemical compounds. The results obtained show slightly better performance of Kohonen networks compared to k-nearest neighbour clustering and equal performance of multi-layer perceptrons and discriminant analysis.

The purpose of the research was to identify the method of preparation and the composition of aceclofenac based nanoparticles using various multivariate data analysis methods. Top down and bottom up approaches were used for the preparation of the nanoparticles and polymers like hydroxylpropyl methylcellulose and polyvinyl pyrrolidone, surfactants like sodium lauryl sulphate and Tween 80 were screened for the purpose of formulation. Different combinations of these factors were used for designing of the experiments and the obtained products were evaluated for particle size, poly dispersibility index, zeta potential, drug content and entrapment efficiency. Principal component analysis, agglomerative hierarchical clustering, composite desirability index aided to identify the best possible combination. Nanoprecipitation along with sonication as process and poly vinyl pyrrolidone as stabilizer gave the highest values of desirability index. The formulation composition of nanosuspension was oscillated around the settings for the zeroing in on optimum quantitative formulation. The nanosuspension was gelled using Carbopol. The gel was studied for in vitro and ex vivo permeation characteristics. The optimized formulation was found to be stable and had acceptable levels of irritation as studied in wistar rats. Thus various data analysis methods could help identify the right combination of process and formulation composition.

Kleiner and Hartigan (1981) introduced trees and castles (graphical techniques for representing multidimensional data) in which the variables are assigned to components of the display on the basis of hierarchical clustering. An experiment was performed to assess the efficacy of trees and castles in discrimination. The graphs were compared to two types of histograms: one with variables assigned randomly, the other with variables assigned according to hierarchical clustering. Trees tended to give the best results.

The goal of this work is the comparison of differences in the lipidomic compositions of human cell lines derived from normal and cancerous breast tissues, and tumor vs. normal tissues obtained after the surgery of breast cancer patients. Hydrophilic interaction liquid chromatography/electrospray ionization mass spectrometry (HILIC/ESI-MS) using the single internal standard approach and response factors is used for the determination of relative abundances of individual lipid species from five lipid classes in total lipid extracts of cell lines and tissues. The supplementary information on the fatty acyl composition is obtained by gas chromatography/mass spectrometry (GC/MS) of fatty acid methyl esters. Multivariate data analysis (MDA) methods, such as nonsupervised principal component analysis (PCA), hierarchical clustering analysis (HCA) and supervised orthogonal partial least-squares discriminant analysis (OPLS-DA), are used for the visualization of differences between normal and tumor samples and the correlation of similarity between cell lines and tissues either for tumor or normal samples. MDA methods are used for differentiation of sample groups and also for identification of the most up- and downregulated lipids in tumor samples in comparison to normal samples. Observed changes are subsequently generalized and correlated with data from tumor and normal tissues of breast cancer patients. In total, 123 lipid species are identified based on their retention behavior in HILIC and observed ions in ESI mass spectra, and relative abundances are determined. MDA methods are applied for a clear differentiation between tumor and normal samples both for cell lines and tissues. The most upregulated lipids are phospholipids (PL) with a low degree of unsaturation (e.g., 32:1 and 34:1) and also some highly polyunsaturated PL (e.g., 40:6), while the most downregulated lipids are PL containing polyunsaturated fatty acyls (e.g., 20:4), plasmalogens and ether lipids. Copyright © 2016 John Wiley & Sons, Ltd.

Some Thoughts About Data Type, Distribution, and Statistical Significance

Understanding complexity in the response of ornamental rosemary to different substrates: A multivariate analysis

Lipids and volatile compounds of shea butter samples from different African countries (Burkina Faso, Chad, Mali, and Uganda) and different production technologies have been analyzed using an approach based on MALDI-TOF-MS and SPME-GC-MS combined with exploratory multivariate data analysis. By applying MALDI-MS analysis total DAGs were found in the range of 7.6- 22.6% and TAGs in the range of 77.7-92.4% of the shea butter samples. Thereby, major DAGs were found PO (7.3-21.9%), PL (0.0-26.3%), SO (16.3-42.1%), OO (15.1-57.4%), LO (9.1-17.9) and major TAGs were identified POO (2.7-7.8%), POS (4.9-8.3%), SOS (13.8-45.3%), SOO (18.9-39.5%), OOO (1.0-24.3%), AOO (1.4-2.5%) and AOS (1.4-3.7%), respectively. TAG/DAG ratios were found to be dependent on production conditions indicating for inactivation of the lipases by boiling the shea kernels before fat extraction. Volatiles pattern of shea samples detected by headspace-SPME-GC-MS contained more than 100 different compounds, among them many aldehydes, ketones and organic acids. PCA of MALDI data showed a dominating influence of geographic origin on TAG composition characterized mostly by a different saturation degree of the fatty acyl residues declining from West to East Africa. Similar geographic relationships could be visualised by exploratory PLS-DA of the volatiles data. Mainly lipid peroxidation products (in particular unsaturated aldehydes) were identified to be responsible for this sample distribution. Volatile patterns are also influenced by producing methods. For example, methylketones - probably of microbial origin - have been found as marker compounds for "heaping" of the shea kernels before fat extraction. Keywords: Butyrospermum parkii (C.F. Gaertn.), MALDI-TOF-MS, Partial least squares discriminant analysis (PLS-DA), Principal component analysis (PCA), Shea butter, Triacylglycerols, Vitellaria paradoxa (C.F. Gaertn.), Volatiles.

Variation patterns of nitric oxide in Catalonia during the period from 2001 to 2006 using multivariate data analysis methods

Multivariate data analysis methods, although widely diffused since the 1960’s, are really much older. Hotelling founded both the Principal Component Analysis and the Canonical Analysis in the 30’s, by further developing previous studies made by Spearman and Pearson at the beginning of the century. These methods were subsequently expanded and enriched, but remained beyond the reach of the common practitioner because of the massive calculations they required. With the birth, and subsequent dramatic development of computers, multivariate data analysis methods have finally become practicable (Bouroche and Saporta, 1983).KeywordsFactor ScoremUltivariate MethodMultivariate Data AnalysisAmes TestHydrazine SulfateThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

A novel method for calculating biomagnification factors is presented and demonstrated using contaminant concentration data from the Swedish national monitoring program regarding organochlorine contaminants (OCs) in herring (Clupea harengus) muscle and guillemot (Uria aalge) egg, sampled from 1996 to 1999 from the Baltic Sea. With this randomly sampled ratios (RSR) method, biomagnification factors (BMF(RSR)) were generated and denoted with standard deviation (SD) as a measure of the variation. The BMFRsR were calculated by randomly selecting one guillemot egg out of a total of 29 and one herring out of a total of 74, and the ratio was determined between the concentration of a given OC in that egg and the concentration of the same OC in that herring. With the resampling technique, this was performed 50 000 times for any given OC, and from this new distribution of ratios, BMF(RSR) for each OC were calculated and given as geometric mean (GM) with GM standard deviation (GMSD) range, arithmetic mean (AM) with AMSD range, and minimum (BMF(MIN)) as well as maximum (BMF(MAX)) biomagnification factors. The 14 analyzed OCs were p,p'DDT and its metabolites p,p'DDE and p,p'DDD, polychlorinated biphenyls (PCB congeners: CB28, CB52, CB101, CB118, CB138, CB153, and CB180), hexachlorocyclohexane isomers (alpha-, beta-, and gammaHCH), and hexachlorobenzene (HCB). Multivariate data analysis (MVDA) methods, including principal components analysis (PCA), partial least squares regression (PLS), and PLS discriminant analyses (PLS-DA), were first used to extract information from the complex biological and chemical data generated from each individual animal. MVDA were used to model similarities/dissimilarities regarding species (PCA, PLS-DA), sample years (PLS), and sample location (PLS-DA) to give a deeper understanding of the data that the BMF modeling was based upon. Contaminants that biomagnify, that had BMF(RSR) significantly higher than one, were p,p'DDE, CB118, HCB, CB138, CB180, CB153, ,betaHCH, and CB28. The contaminants that did not biomagnifywere p,p'DDT, p,p'DDD, alphaHCH, CB101, and CB52. Eventual biomagnification for gammaHCH could not be determined. The BMF(RSR) for OCs present in herring muscle and guillemot egg showed a broad span with large variations for each contaminant. To be able to make reliable calculations of BMFs for different contaminants, we emphasize the importance of using data based upon large numbers of, as well as well-defined, individuals.

Five multivariate data analysis methods, namely principal components analysis, non-linear mapping, hierarchical clustering, minimal spanning tree and mode analysis, were used to analyse the natural structure of red wines from Tarragona (Catalonia, Spain). Seventeen parameters were analysed for each sample to determine the multi-dimensional spectrum in which the wines are studied. In spite of the homogeneity of the wine samples which resulted from the similarity of the edaphic, varietal factors as well as the cultural conditions of the producing areas, two different clusters can be distinguished on the basis of all the methods used. The groups of wine samples obtained coincide with two viticultural zones that differ with regard to climate and geographical conditions. Therefore, a natural distinction can bei made between the wines produced in each adjacent zone, forming the basis for future differentiation according to brand name.

Lipidomic differentiation between human kidney tumors and surrounding normal tissues using HILIC-HPLC/ESI-MS and multivariate data analysis.

Multivariate data analysis of key pollutants in sewage samples: a case study