Biomedical subjects

Jon G Wilkes

Publications and source records attributed to Jon G Wilkes.

10 recordsLinked to original sources

PubMed2006

Improved cell typing by charge-state deconvolution of matrix-assisted laser desorption/ionization mass spectra.

Robust, specific, and rapid identification of toxic strains of bacteria and viruses, to guide the mitigation of their adverse health effects and optimum implementation of other response actions, remains a major analytical challenge. This need has driven the development of methods for classification of microorganisms using mass spectrometry, particularly matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS), that allows high-throughput analyses with minimum sample preparation. We describe a novel approach to cell typing based on pattern recognition of MALDI mass spectra, which involves charge-state deconvolution in conjunction with a new correlation analysis procedure. The method is applicable to both prokaryotic and eukaryotic cells. Charge-state deconvolution improves the quantitative reproducibility of spectra because multiply charged ions resulting from the same biomarker attaching a different number of protons are recognized and their abundances are combined. This allows a clearer distinction of bacterial strains or of cancerous and normal liver cells. Improved class distinction provided by charge-state deconvolution was demonstrated by cluster spacing on canonical variate score charts and by correlation analyses. Deconvolution may enhance detection of early disease state or therapy progress markers in various tissues analyzed by MALDI-MS.

Algorithms↗

PubMed2006

Pyrolysis mass spectrometry for distinguishing potential hoax materials from bioterror agents.

Pyrolysis mass spectrometry (PyMS) was investigated as a rapid tool to distinguish potential bioterror hoax materials from samples containing pathogenic bacteria. A pyrolysis time-of-flight (TOF) mass spectrometer equipped with an alternative ionization technique, metastable atom bombardment (MAB), was used to produce sample spectra. These spectra were analyzed by principal component and discriminant analysis for pattern recognition. Materials investigated were two strains of Vibrio parahaemolyticus, one of which produced the tdh toxin, two Salmonella enterica serotypes, a biological mosquito control product containing spores of Bacillus thuringiensis, and several white to off-white powders (which could be used as hoax materials), such as flour, corn starch, methyl cellulose, and xanthan gum. PyMS distinguished bacterial samples from hoax materials. Furthermore, pattern analysis differentiated Vibrios from Salmonellae, Salmonella enterica Anatum from S. enterica Heidelberg, and the two V. parahaemolyticus strains from each other. The B. thuringiensis mixture was distinguished from other bacteria and powders, suggesting that PyMS with pattern recognition may differentiate samples containing pathogens, including Bacillus spp., from nonbiological agents and that it can be a rapid method for detection of bacteria. MS data acquisition took only 7 min for each sample.

Bacillus thuringiensis↗

PubMed2005

Rapid phenotypic characterization of Vibrio isolates by pyrolysis metastable atom bombardment mass spectrometry.

Pyrolysis mass spectrometry was investigated for rapid characterization of food-borne bacterial pathogens. Nine isolates of Vibrio parahaemolyticus and one isolate each of Vibrio fluvialis, Vibrio hollisae, and Vibrio vulnificus were analyzed. Pyrolysis mass spectra, generated via an alternative ionization method, metastable atom bombardment, were subject to principal component-discriminant analysis. The spectral patterns were used to distinguish Vibrio isolates differing in species, serotype and expression of the thermostable direct hemolysin gene. The patterns of similarity and dissimilarity amongst spectra in the Vibrio test set generally reflected those associated with species, serotype or hemolysin-producing genes, though the combined influence of these and other variables in the multi-dimensional data did not produce a simple clustering with respect to any one of these characteristics. These results suggested that with enough examples to model the most common combinations, the method should be able to characterize Vibrio isolates according to their phenotypic characteristics. Pyrolysis-mass spectrometry with metastable atom bombardment and pattern recognition appeared suitable for rapid infraspecific comparison of Vibrio isolates. This integrated analytical, pattern-recognition system should be examined further for potential utility in clinical and public health diagnostic contexts.

Animals↗

PubMed2005

Rapid phenotypic characterization of Salmonella enterica strains by pyrolysis metastable atom bombardment mass spectrometry with multivariate statistical and artificial neural network pattern recognition.

Pyrolysis mass spectrometry was investigated for rapid characterization of bacteria. Spectra of Salmonella were compared to their serovars, pulsed-field gel electrophoresis (PFGE) patterns, antibiotic resistance profiles, and MIC values. Pyrolysis mass spectra generated via metastable atom bombardment were analyzed by multivariate principal component-discriminant analysis and artificial neural networks (ANNs). Spectral patterns developed by discriminant analysis and tested with Leave-One-Out (LOO) cross-validation distinguished Salmonella strains by serovar (97% correct) and by PFGE groups (49%). An ANN model of the same PFGE groups was cross-validated, using the LOO rule, with 92% agreement. Using an ANN, thirty previously unseen spectra were correctly classified by serotype (97%) and at the PFGE level (67%). Attempts by ANN to model spectra grouped by resistance profile-but ignoring PFGE or serotype-failed (10% correct), but ANNs differentiating ten samples of the same serotype/PFGE class were more successful. To assess the information content of PyMS data serendipitously associated with or directly related to resistance character, the ten isolates were grouped into four, three, or two categories. The four categories corresponded to four resistance profiles. The four class and three class ANNs showed much improved but insufficient modeling power. The two-class ANN and a corresponding multivariate model maximized inferential power for a coarse antibiotic-resistance-related distinction. They each cross-validated by LOO at 90%. This is the first direct correlation of pyrolysis metastable atom bombardment mass spectrometry with immunological (e.g. serology) or molecular biology (e.g. PFGE) based techniques.

Bacterial Typing Techniques↗

PubMed2004

Predicting toxic equivalence factors from 13C nuclear magnetic resonance spectra for dioxins, furans, and polychlorinated biphenyls using linear and nonlinear pattern recognition methods.

Two quantitative spectrometric data-activity relationships (QSDAR) models have been developed relating 29 dioxin or dioxin-like molecules to their toxic equivalence factors (TEFs). These models were based on patterns in simulated 13C nuclear magnetic resonance (NMR) data with the patterns defined by comparative spectral analysis (CoSA). Two versions of CoSA multiple linear regression (MLR) models using 7 or 10 spectral bins had, respectively, explained variances (r2) of 0.88 and 0.95, and leave-one-out (LOO) cross-validated variances (q2) of 0.78 and 0.88. A third, artificial neural network model--using a feed forward, back propagating, three-layer neural network--produced an r2 of 0.99, a LOO q2 of 0.82, and a leave-three-out q2 of 0.81. A postulated reason that the results of these QSDAR models are better than traditional quantitative structure-activity relationship (QSAR) models is based on the difference in descriptors rather than on any differences in pattern recognition approach. Results suggest that the 13C NMR spectral data contain molecular quantum mechanical information more reflective of each molecule's biochemical properties than do the calculated electrostatic potentials and molecular alignment assumptions used in developing QSAR models. The QSDAR models provide a rapid, simple way to model the toxicity of dioxin and dioxin-like compounds.

Animals↗

PubMed2003

The use of carbon thirteen nuclear magnetic resonance spectra to predict dioxin and furan binding affinities to the aryl hydrocarbon receptor.

Four spectroscopic data-activity relationship (SDAR) models for polychlorinated dibenzofurans (PCDFs) and dibenzodioxins (PCDDs) binding to the aryl hydrocarbon receptor (AhR) have been developed based on simulated 13C nuclear magnetic resonance (NMR) data. Models were developed using discriminant function analysis of the compounds' spectral data. An SDAR model with two classifications for 26 PCDF compounds had a leave-one-out (LOO) cross-validation accuracy of 89%. A two-classification SDAR model for 14 PCDD compounds had LOO cross-validation accuracy of 95%. A two-classification SDAR model combining 14 PCDD and 26 PCDF compounds had LOO cross-validation accuracy of 88%, while a four-classification SDAR model based on the same 14 PCDD and 26 PCDF compounds had LOO cross-validation accuracy of 92%. We used each appropriate SDAR model to classify 41 PCDD and/or 121 PCDF compounds with unknown binding affinities to the AhR. The SDAR models provide a rapid, simple, and valid way to model the PCDF and PCDD binding activity in relation to the AhR.

Benzofurans↗

PubMed2002

Defining and using microbial spectral databases.

This work shows how fingerprints of mass spectral patterns from microbial isolates are affected by variations in instrumental condition, by sample environment, and by sample handling factors. It describes a novel method by which pattern distortions can be mathematically corrected for variations in factors not amenable to experimental control. One uncontrollable variable is "between-batch" differences in culture media. Another, relevant for determination of noncultured extracts, is differences between the cells' environmental experience (e.g., starved environmental extracts versus cultured standards). The method suggests that, after a single growth cycle on a solid medium (perhaps, a selective one), pyrolysis MS spectra of microbial isolates can be algorithmically compensated and an unknown isolate identified using a spectral database defined by culture on a different (perhaps, nonselective) medium. This reduces identification time to as few as 24 h from sample collection. The concept also proposes a possible way to compensate certain noncultured, nonisolated samples (e.g., cells concentrated from urine or impacted from aerosol or semi-selectively extracted by immunoaffinity methods from heavily contaminated matrices) for identification within half an hour. Using the method, microbial mass spectra from different labs can be assembled into coherent databases similar to those routinely used to identify pure compounds. This type of data treatment is applicable for rapid detection in biowarfare and bioterror events as well as in forensic, research, and clinical laboratory contexts.

Algorithms↗

PubMed2002

Combining NMR spectral and structural data to form models of polychlorinated dibenzodioxins, dibenzofurans, and biphenyls binding to the AhR.

A three-dimensional quantitative spectrometric data-activity relationship (3D-QSDAR) modeling technique which uses NMR spectral and structural information that is combined in a 3D-connectivity matrix has been developed. A 3D-connectivity matrix was built by displaying all possible assigned carbon NMR chemical shifts, carbon-to-carbon connections, and distances between the carbons. Two-dimensional 13C-13C COSY and 2D slices from the distance dimension of the 3D-connectivity matrix were used to produce a relationship among the 2D spectral patterns for polychlorinated dibenzofurans, dibenzodioxins, and biphenyls (PCDFs, PCDDs, and PCBs respectively) binding to the aryl hydrocarbon receptor (AhR). We refer to this technique as comparative structural connectivity spectral analysis (CoSCoSA) modeling. All CoSCoSA models were developed using forward multiple linear regression analysis of the predicted 13C NMR structure-connectivity spectral bins. A CoSCoSA model for 26 PCDFs had an explained variance (r2) of 0.93 and an average leave-four-out cross-validated variance (q(2)4) of 0.89. A CoSCoSA model for 14 PCDDs produced an r2 of 0.90 and an average leave-two-out cross-validated variance (q(2)2) of 0.79. One CoSCoSA model for 12 PCBs gave an r2 of 0.91 and an average q(2)2 of 0.80. Another CoSCoSA model for all 52 compounds had an r2 of 0.85 and an average q(2)2 of 0.52. Major benefits of CoSCoSA modeling include ease of development since the technique does not use molecular docking routines.

Algorithms↗

PubMedDate not supplied

Comparative structural connectivity spectra analysis (CoSCoSA) models of steroids binding to the aromatase enzyme.

A method that combines NMR spectral and structural information into a constructed three-dimensional (3D)-connectivity matrix is developed for modeling biological binding activity of small molecules. The 3D-connectivity matrix for a molecule is defined by associating the distances between all possible carbon-to-carbon connections with their assigned carbon NMR chemical shifts. In this project we selected from the total 3D-connectivity matrix a subset, the two-dimensional (2D) (13)C-(13)C COSY and a theoretical long range 2D (13)C-(13)C distance connectivity spectral plane. Patterns of (13)C chemical shifts observed at these two relative distances for 50 steroids were used to produce a mathematical relationship for the steroids' relative binding affinity (pK(i)) to the aromatase enzyme. We call this technique comparative structural connectivity spectra analysis (CoSCoSA) modeling. Using combinations of the 2D COSY and 2D long-range distance spectra as modeling parameters, we built four CoSCoSA models. One model was made from the 2D COSY spectra alone and another was developed using only the 2D long-range distance spectra. Then the COSY and long-distance spectra were combined in two different ways: starting with the combined principal components (PCs) from the separately calculated COSY and distance spectra or using the combined raw spectra (3D). The best CoSCoSA model was based on the combined PCs from COSY and distance spectra. This model had an r(2) of 0.96 and a leave-one-out cross-validation (q(2)) of 0.92. In general CoSCoSA modeling combines the quantum mechanical information inherent in NMR chemical shifts with internal molecular atom-to-atom distances to give a reliable and straightforward basis for predictive modeling. The technique has the flexibility and accuracy to outperform not only the cross-validated variance q(2) of previously published quantitative structure-activity relationships (QSAR) but also those obtained by related quantitative spectral data-activity relationships (QSDARs) lacking connectivity dimensions.

Aromatase↗

PubMedDate not supplied

Comparative structural connectivity spectra analysis (CoSCoSA) models of steroid binding to the corticosteroid binding globulin.

A three-dimensional quantitative spectrometric data-activity relationship (3D-QSDAR) model was developed that is built by combining NMR spectral information with structural information in a 3D-connectivity matrix. The 3D-connectivity matrix is built by displaying all possible carbon-to-carbon connections with their assigned carbon NMR chemical shifts and distances between the carbons. Selected 2D (13)C-(13)C COrrelation SpectroscopY (COSY) (through-bond nearest neighbors) and selected theoretical 2D (13)C-(13)C distance connectivity spectral slices from the 3D-connectivity matrix to produce a relationship among the spectral patterns for 30 steroids binding to corticosteroid binding globulin. We call this technique a comparative structural connectivity spectra analysis (CoSCoSA) modeling. A CoSCoSA principal component linear regression model based on the combination of (13)C-(13)C COSY and (13)C-(13)C distance spectra principal components (PCs) had an r(2) of 0.96 and a leave-one-out (LOO) cross-validation q(2) of 0.92. A CoSCoSA parallel distributed artificial neural network (PD-ANN) model based on the combination of (13)C-(13)C COSY and (13)C-(13)C distance spectra had an r(2) of 0.96, a leave-three-out q(3)(2) of 0.78, and a leave-ten-out q(10)(2) of 0.73. CoSCoSA modeling attempts to uniquely combine the quantum mechanics information from the NMR chemical shifts with internal molecular atom-to-atom distances into an accurate modeling technique. The CoSCoSA modeling technique has the flexibility and accuracy to outperform the cross-validated variance q(2) of previously published quantitative structure-activity relationship (QSAR), quantitative spectral data-activity relationship (QSDAR), self-organizing map (SOM), and electrotopological state (E-state) models.

Binding Sites↗