Classification and identification of bacteria by Fourier-transform infrared spectroscopy.

This study describes a computer-based technique for classifying and identifying bacterial samples using Fourier-transform infrared spectroscopy (FT-IR) patterns. Classification schemes were tested for selected series of bacterial strains and species from a variety of different genera. Dissimilarities between bacterial IR spectra were calculated using modified correlation coefficients. Dissimilarity matrices were used for cluster analysis, which yielded dendrograms broadly equated with conventional taxonomic classification schemes. Analyses were performed with selected strains of the taxa Staphylococcus, Streptococcus, Clostridium, Legionella and Escherichia coli in particular, and with a database containing 139 bacterial reference spectra. The latter covered a wide range of Gram-negative and Gram-positive bacteria. Unknown specimens could be identified when included in an established cluster analysis. Thirty-six clinical isolates of Staphylococcus aureus and 24 of Streptococcus faecalis were tested and all were assigned to the correct species cluster. It is concluded that: (1) FT-IR patterns can be used to type bacteria; (2) FT-IR provides data which can be treated such that classifications are similar and/or complementary to conventional classification schemes; and (3) FT-IR can be used as an easy and safe method for the rapid identification of clinical isolates.     

Evaluation of the bony repair in rat cranial defect using near infrared reflectance spectroscopy and discriminant analysis

We set out to determine whether near infrared reflectance spectroscopy (NIRS) combined with principal component analysis–linear discriminant analysis (LDA) or, variable selection techniques employing successive projection algorithm or genetic algorithm (GA) could evaluate the bone repair in cranial critical‐size (5 mm) defect after stimulation with collagen sponge scaffold and/or infrared low‐level laser therapy directly on the local. Forty‐five Winstar rats were divided into nine groups of five each, namely: group H – healthy, n = 5 (without treatment and without cranial critical‐size defect), (GI positive control – n = 5, 21 days or n = 5, 30 days) without treatment and with cranial critical‐size defect; (GII‐n = 5, 21 days or n = 5, 30 days) cranial critical‐size defect filled with collagen sponge scaffold; (GIII–n = 5, 21 days or n = 5, 30 days) cranial critical‐size defect submitted to low‐level laser therapy; (GIV–n = 5, 21 days or n = 5, 30 days) cranial critical‐size defect submitted to combined collagen sponge scaffold + low‐level laser therapy treatment. In relation to the histological analysis, the collagen sponge scaffold + low‐level laser therapy treatment group (GIV) 30 days showed the best result with the presence of secondary bone, immature bone (osteoid) and newly formed connective tissue (periosteum). GA–LDA model also successfully classified control class of the others classes. Thus, the results provided by the good‐quality classification model revealed the feasibility of NIRS for application to evaluation of the wound healing in rat cranial defect, thanks to the short analysis time of a few seconds and nondestructive advantages of NIRS as an alternative approach for bone repair purposes. © 2017 American Institute of Chemical Engineers Biotechnol. Prog., 33:1160–1168, 2017     

Classification and misclassification in sexing the Black femur by discriminant function analysis

Stepwise discriminant function analysis for sex assessment was applied to 130 North American Black femora. The measurements included femoral length and three midshaft dimensions likely to be preserved in archaeologically derived and forensic remains. The method correctly assigned sex for 76.4% of the sample (range 70.8–81.5%). This compares favorably with results achieved with other skeletal parts; it also compares favorably with results using the femur in sexing other racial groups. Among our other conclusions are: (1) a “general size factor” is one of major significance in correct classification and in misclassification of sex, and most misclassified individuals are anomalous for this factor; (2) the inconsistency in the relation between circumference and femoral length, which characterizes the remaining misclassified individuals, suggests that anomalous functional demands of body weight/musculature are at fault, and affect circumference more than length; and (3) discriminant function analysis of the same variables in Whites produced similar results, suggesting that sex overrides race in sex assessment; this was confirmed by cross-validating the predictive accuracy of Black discriminant function coefficients on White data, and vice versa.     

Regularized linear discriminant analysis and its application in microarrays

In this paper, we introduce a modified version of linear discriminant analysis, called the "shrunken centroids regularized discriminant analysis" (SCRDA). This method generalizes the idea of the "nearest shrunken centroids" (NSC) (Tibshirani and others, 2003) into the classical discriminant analysis. The SCRDA method is specially designed for classification problems in high dimension low sample size situations, for example, microarray data. Through both simulated data and real life data, it is shown that this method performs very well in multivariate classification problems, often outperforms the PAM method (using the NSC algorithm) and can be as competitive as the support vector machines classifiers. It is also suitable for feature elimination purpose and can be used as gene selection method. The open source R package for this method (named "rda") is available on CRAN (http://www.r-project.org) for download and testing.     

Classification of root canal microorganisms using electronic-nose and discriminant analysis

Background

The finite volume solver Fluent (Lebanon, NH, USA) is a computational fluid dynamics software employed to analyse biological mass-transport in the vasculature. A principal consideration for computational modelling of blood-side mass-transport is convection-diffusion discretisation scheme selection. Due to numerous discretisation schemes available when developing a mass-transport numerical model, the results obtained should either be validated against benchmark theoretical solutions or experimentally obtained results.

Methods

An idealised aneurysm model was selected for the experimental and computational mass-transport analysis of species concentration due to its well-defined recirculation region within the aneurysmal sac, allowing species concentration to vary slowly with time. The experimental results were obtained from fluid samples extracted from a glass aneurysm model, using the direct spectrophometric concentration measurement technique. The computational analysis was conducted using the four convection-diffusion discretisation schemes available to the Fluent user, including the First-Order Upwind, the Power Law, the Second-Order Upwind and the Quadratic Upstream Interpolation for Convective Kinetics (QUICK) schemes. The fluid has a diffusivity of 3.125 × 10^-10 m²/s in water, resulting in a Peclet number of 2,560,000, indicating strongly convection-dominated flow.

Results

The discretisation scheme applied to the solution of the convection-diffusion equation, for blood-side mass-transport within the vasculature, has a significant influence on the resultant species concentration field. The First-Order Upwind and the Power Law schemes produce similar results. The Second-Order Upwind and QUICK schemes also correlate well but differ considerably from the concentration contour plots of the First-Order Upwind and Power Law schemes. The computational results were then compared to the experimental findings. An average error of 140% and 116% was demonstrated between the experimental results and those obtained from the First-Order Upwind and Power Law schemes, respectively. However, both the Second-Order upwind and QUICK schemes accurately predict species concentration under high Peclet number, convection-dominated flow conditions.

Conclusion

Convection-diffusion discretisation scheme selection has a strong influence on resultant species concentration fields, as determined by CFD. Furthermore, either the Second-Order or QUICK discretisation schemes should be implemented when numerically modelling convection-dominated mass-transport conditions. Finally, care should be taken not to utilize computationally inexpensive discretisation schemes at the cost of accuracy in resultant species concentration.     

Unbiased bootstrap error estimation for linear discriminant analysis

Convex bootstrap error estimation is a popular tool for classifier error estimation in gene expression studies. A basic question is how to determine the weight for the convex combination between the basic bootstrap estimator and the resubstitution estimator such that the resulting estimator is unbiased at finite sample sizes. The well-known 0.632 bootstrap error estimator uses asymptotic arguments to propose a fixed 0.632 weight, whereas the more recent 0.632+ bootstrap error estimator attempts to set the weight adaptively. In this paper, we study the finite sample problem in the case of linear discriminant analysis under Gaussian populations. We derive exact expressions for the weight that guarantee unbiasedness of the convex bootstrap error estimator in the univariate and multivariate cases, without making asymptotic simplifications. Using exact computation in the univariate case and an accurate approximation in the multivariate case, we obtain the required weight and show that it can deviate significantly from the constant 0.632 weight, depending on the sample size and Bayes error for the problem. The methodology is illustrated by application on data from a well-known cancer classification study.     

Classification of breast cancer versus normal samples from mass spectrometry profiles using linear discriminant analysis of important features selected by random forest

We present our approach to classifying the processed proteomic data that were made available to the participants of the classification competition. Although classification of the spectra was the goal of the competition we feel that proteomic applications to cancer biomarker studies make certain additional demands. For example, one such requirement should be identification of certain features which collectively could differentiate the two groups of samples. Also ideally, the size of the feature set should be small. To that end we propose a linear discriminant classifier based on nine m/z intensity values. Construction and performance of this classifier are discussed.     

Quantitative analysis of cellulose nitrates by Fourier transform infrared spectroscopy

Quantitative express analysis of nitrogen content in cellulose nitrates by Fourier transform infrared spectroscopy has been developed. The slope of the dependence of the ratio of the band intensity (and area) to sample weight in a tablet, on the nitrogen content in a sample was used to find the reduced extinction coefficients for quantitative analysis of nitrogen content in cellulose nitrate samples by IR spectroscopy. The results were compared with the nitrogen content values in the same samples determined by the ferrosulfate method.     

In vivo analysis of the skin by Fourier-transformed infrared spectroscopy

Advantages of Fourier transform infrared spectroscopy using reflexion techniques are used for skin analyses. Several examples of spectra obtained with a skin analyser are given and the major absorptions in the range 3,500 to 1,150 cm-1 are assigned to fundamental vibrations. Some of them can be used to determine the hydratation level of the skin.     

Distinguishing and grading human gliomas by IR spectroscopy

As a molecular probe of tissue composition, IR spectroscopy can potentially serve as an adjunct to histopathology in detecting and diagnosing disease. This study demonstrates that cancerous brain tissue (astrocytoma, glioblastoma) is distinguishable from control tissue on the basis of the IR spectra of thin tissue sections. It is further shown that the IR spectra of astrocytoma and glioblastoma affected tissue can be discriminated from one another, thus providing insight into the malignancy grade of the tissue. Both the spectra and the methods employed for their classification reveal characteristic differences in tissue composition. In particular, the nature and relative amounts of brain lipids, including both the gangliosides and phospholipids, appear to be altered in cancerous compared to control tissue. Using a genetic classification approach, classification success rates of up to 89% accuracy were obtained, depending on the number of regions included in the model. The diagnostic potential and practical applications of IR spectroscopy in brain tumor diagnosis are discussed.     

Lipid-protein ratios by infrared spectroscopy

A method for measuring the ratio of diacyl phospholipid to protein in lipid-protein mixtures and membranes by infrared spectroscopy is described. Samples made of diacyl phospholipid and proteins mixed in known ratios were analyzed for lipid-protein ratio by the infrared (ir) method. Results had a standard deviation of less than +/- 4% over the lipid-protein molar ratio range of 9:1 to 320:1. Calculations of the ratio of total lipid to protein require that the diacyl phospholipid-to-protein ratio be divided by the mole fraction of diacyl phospholipid in the total lipid. Phospholipid-protein ratios for various sarcoplasmic reticulum membrane preparations (R1-washed, octylglucoside purified, deoxycholate treated) were determined by the ir method and compared to literature values. Also, phospholipid-protein ratios were determined for R1-washed sarcoplasmic reticulum by three chemical analyses using different protein assays and were compared with ratios obtained by the infrared method. The infrared results were closest to those of a chemical method designed specifically for membrane proteins.     

The quantitative analysis of uronic acid polymers by infrared spectroscopy.

I.r. absorption bands associated with the functional groups of carboxylic acid derivatives are useful for the analysis of alginates and pectins. The ester, amide, and uronate contents of pectins and the uronate content of alginates were determined, respectively, from the ester-carbonyl stretching band (1740 cm- minus 1), the amide I band (1650 cm- minus 1), and the carboxylate antisymmetric stretching band (1607 cm- minus 1) obtained from the spectra of solutions in D2O-phosphate buffer. The results are accurate to within plus or minus 2-4%, are self consistent, and agree well with the few reliable results that are available. The method should be applicable for the determination of carboxylic acid derivatives in other polysaccharides.     

Predictive value of discriminant analysis of monocyte ultrastructure in malignant lymphoma

Discriminant analysis was applied to morphometric data obtained from ultrastructural studies of blood monocytes from 20 normal subjects, 23 patients with Hodgkin's disease and 12 patients with non-Hodgkin's lymphoma. The aim was to assess the efficiency of predicting subject groups from such data. The analysis, performed on a microcomputer using a standard statistical package, considered nuclear volume, nuclear surface area, nucleolar volume, nucleolar surface area, nucleolar volume fraction, number of nucleoli per section, cell surface area, mitochondrial surface area and subject age. The overall agreement between predicted and actual subject groups was 64%; considering only normality and disease, the agreement was 80%. While the predictive value of such data from circulating monocytes would appear insufficient for diagnostic purposes, discriminant analysis as used here might be of value in indicating the state of host defense in malignancy.     

Taxonomic discrimination of flowering plants by multivariate analysis of Fourier transform infrared spectroscopy data

Fourier transform infrared spectroscopy (FTIR) provides biochemical profiles containing overlapping signals from a majority of the compounds that are present when whole cells are analyzed. Leaf samples of seven higher plant species and varieties were subjected to FTIR to determine whether plants can be discriminated phylogenetically on the basis of biochemical profiles. A hierarchical dendrogram based on principal component analysis (PCA) of FTIR data showed relationships between plants that were in agreement with known plant taxonomy. Genetic programming (GP) analysis determined the top three to five biomarkers from FTIR data that discriminated plants at each hierarchical level of the dendrogram. Most biomarkers determined by GP analysis at each hierarchical level were specific to the carbohydrate fingerprint region (1,200–800 cm^–1) of the FTIR spectrum. Our results indicate that differences in cell-wall composition and structure can provide the basis for chemotaxonomy of flowering plants.Abbreviations FTIR Fourier transform infrared spectroscopy - GP Genetic programming - PCA Principal component analysis - PyMS Pyrolysis mass spectrometry     

Morphometry of macrophage development in malignant lymphoma. Predictive value of discriminant analysis

Discriminant analysis of morphometric data on the ultrastructure of developing macrophages has been used to classify 62 individual subjects into one of the 3 groups of origin, namely normal, Hodgkin's disease and non-Hodgkin's lymphoma, each finding being compared with the known diagnosis. The data had been obtained from blood monocytes grown in suspension culture over a period of 6 days, and related to whole cell, nucleus, nucleoli and mitochondria. Over 80% of subjects were correctly classified as between the 3 groups and over 90% as to their normality or otherwise. Although the non-specific nature of changes in defence cells makes it unlikely that morphometric studies of macrophages will find a place in the diagnosis of specific malignancies, the present work indicates it could be useful in assessing host response and hence prognosis and response to treatment. Discriminant analysis of quantitative differences in cell structure could have wide clinico-pathological application.