A robust regression approach for spectrophotometric signal analysis

The effectiveness of a regression method strongly depends on the characteristics of the considered regression problem. As a consequence, this makes it difficult to choose a priori the most appropriate algorithm for a given dataset. This issue is faced in this work through a novel regression approach based on the fusion of an ensemble of different regressors. In order to implement the proposed robust multiple system (RMS), four different fusion strategies are explored. In this context, we propose a novel fusion strategy named selection‐based strategy (SBS) that provides as output the estimate obtained by the regression algorithm (included in the ensemble) characterized by the highest expected accuracy in the region of the feature space associated with the considered model. The SBS is based not on a direct combination of the estimates yielded by all the regressors but on a selection mechanism that identifies the expected best available estimate. For such purpose, it exploits the accuracies of the regressors included in the ensemble in different portions of the input feature space. The experimental assessment of the RMS was carried out on three different datasets: a wine, an orange juice, and an apple datasets. The obtained experimental results suggest that, in general, the fusion of an ensemble of different regression algorithms leads to a regression process that is more robust and sometimes also more accurate than traditional regression methods. In particular, the proposed SBS method represents an effective solution to carry out the fusion process. Copyright © 2012 John Wiley & Sons, Ltd.     

Reduced‐order model for monitoring spectroscopic and chromatographic polymer properties

A reduced‐order mechanistic polymerization model and its application in the design of a low‐dimension multi‐rate state estimator (soft sensor) for monitoring spectroscopic and chromatographic polymer properties are presented. A model reduction approach is used to simplify a method‐of‐moments mechanistic model. Using this approach, the order (number of the state variables) of the model is reduced from 20 to 7. The soft sensor estimates spectroscopic and chromatographic polymer properties from (a) frequent measurements of the reactor temperature and the flow rates of monomer (n‐butyl acrylate), initiator (t‐butyl peroxy acetate) solution and solvent (xylene) feed streams, and (b) infrequent and delayed measurements of polymer number‐ and weight‐average molecular weights and the concentrations of terminal solvent groups, terminal double bonds and short chain branches. The benefits of using the infrequent measurements in the estimation are shown. The soft sensor is implemented in real‐time, and the calculated continuous estimates of polymer number‐ and weight‐average molecular weights and the concentrations of solvent, terminal double bonds and short chain branches are compared to the corresponding chromatographic and spectroscopic measurements. Copyright © 2007 John Wiley & Sons, Ltd.     

Waste-recycling Monte Carlo with optimal estimates: application to free energy calculations in alloys

The estimator proposed recently by Delmas and Jourdain for waste-recycling Monte Carlo achieves variance reduction optimally with respect to a control variate that is evaluated directly using the simulation data. Here, the performance of this estimator is assessed numerically for free energy calculations in generic binary alloys and is compared to those of other estimators taken from the literature. A systematic investigation with varying simulation parameters of a simplified system, the anti-ferromagnetic Ising model, is first carried out in the transmutation ensemble using path-sampling. We observe numerically that (i) the variance of the Delmas-Jourdain estimator is indeed reduced compared to that of other estimators; and that (ii) the resulting reduction is close to the maximal possible one, despite the inaccuracy in the estimated control variate. More extensive path-sampling simulations involving an FeCr alloy system described by a many-body potential additionally show that (iii) gradual transmutations accommodate the atomic frustrations; thus, alleviating the numerical ergodicity issue present in numerous alloy systems and eventually enabling the determination of phase coexistence conditions.     

Estimating equilibrium ensemble averages using multiple time slices from driven nonequilibrium processes: theory and application to free energies, moments, and thermodynamic length in single-molecule pulling experiments

Recently discovered identities in statistical mechanics have enabled the calculation of equilibrium ensemble averages from realizations of driven nonequilibrium processes, including single-molecule pulling experiments and analogous computer simulations. Challenges in collecting large data sets motivate the pursuit of efficient statistical estimators that maximize use of available information. Along these lines, Hummer and Szabo developed an estimator that combines data from multiple time slices along a driven nonequilibrium process to compute the potential of mean force. Here, we generalize their approach, pooling information from multiple time slices to estimate arbitrary equilibrium expectations. Our expression may be combined with estimators of path-ensemble averages, including existing optimal estimators that use data collected by unidirectional and bidirectional protocols. We demonstrate the estimator by calculating free energies, moments of the polymer extension, the thermodynamic metric tensor, and the thermodynamic length in a model single-molecule pulling experiment. Compared to estimators that only use individual time slices, our multiple time-slice estimators yield substantially smoother estimates and achieve lower variance for higher-order moments.     

A categorical structure–activity relationship analysis of GPR119 ligands

The categorical structure–activity relationship (cat-SAR) expert system has been successfully used in the analysis of chemical compounds that cause toxicity. Herein we describe the use of this fragment-based approach to model ligands for the G protein-coupled receptor 119 (GPR119). Using compounds that are known GPR119 agonists and compounds that we have confirmed experimentally that are not GPR119 agonists, four distinct cat-SAR models were developed. Using a leave-one-out validation routine, the best GPR119 model had an overall concordance of 99%, a sensitivity of 99%, and a specificity of 100%. Our findings from the in-depth fragment analysis of several known GPR119 agonists were consistent with previously reported GPR119 structure–activity relationship (SAR) analyses. Overall, while our results indicate that we have developed a highly predictive cat-SAR model that can be potentially used to rapidly screen for prospective GPR119 ligands, the applicability domain must be taken into consideration. Moreover, our study demonstrates for the first time that the cat-SAR expert system can be used to model G protein-coupled receptor ligands, many of which are important therapeutic agents.     

Optimization of a novel headspace–solid‐phase microextraction–gas chromatographic method by means of a Doehlert uniform shell design for the analysis of trace level ethylene oxide residuals in sterilized medical devices

Medical devices sterilized by ethylene oxide (EtO) retain trace quantities of EtO residuals, which may irritate patients' tissue. Reliably quantifying trace level EtO residuals in small medical devices requires an extremely sensitive analytical method. In this research, a Doehlert uniform shell design was utilized in obtaining a response surface to optimize a novel headspace–solid‐phase microextraction–gas chromatographic (HS‐SPME‐GC) method developed for analyzing trace levels of EtO residuals in sterilized medical devices, by evaluating sterilized, polymer‐coated, drug‐eluting cardiovascular stents. The effects of four independent experimental variables (HS‐SPME desorption time, extraction temperature, GC inlet temperature and extraction time) on GC peak area response of EtO were investigated simultaneously and the most influential experimental variables determined were extraction temperature and GC inlet temperature, with the fitted model showing no evidence of lack‐of‐fit. The optimized HS‐SPME‐GC method demonstrated overall good linearity/linear range, accuracy, repeatability, reproducibility, absolute recovery and high sensitivity. This novel method was successfully applied to analysis of trace levels of EtO residuals in sterilized/aerated cardiovascular stents of various lengths and internal diameter, where, upon heating, trace EtO residuals fully volatilized into HS for extraction, thereby nullifying matrix effects. As an alternative, this novel HS‐SPME‐GC method can offer higher sensitivity compared with conventional headspace analyzer‐based sampling. Copyright © 2009 John Wiley & Sons, Ltd.     

AFM‐Correlated CSM‐Coupled Raman/Fluorescence Studies on Selective Interactions of Water Soluble Porphyrins with DNA

The Raman and fluorescence spectroscopic properties of water‐soluble oxo‐titanium(IV) mesotetrakis (1‐methyl pyridium‐4‐yl) porphyrin (O=Ti(TMPyP)⁴⁺) bound with calf thymus DNA and artificial DNAs such as double stranded poly[d(A‐T)₂] and poly[d(G‐C)₂] have been investigated on the single DNA molecule basis by AFM‐correlated confocal scanning microscope (CSM)‐coupled Raman and fluorescence spectroscopic techniques as well as the ensemble‐averaged spectroscopy. The ensemble‐averaged spectroscopic studies imply that the porphyrin interacts with DNA in different groove binding patterns depending on the base pairs. AFM‐images of the different DNAs bound with O=Ti(TMPyP)⁴⁺ were measured, and their morphologies are found to depend on kind of base pairs interacting with O=Ti(TMPyP)⁴⁺. Being correlated with the AFM images, the CSM‐coupled Raman and fluorescence spectral properties of the three different single O=Ti(TMPyP)⁴⁺‐DNA complexes were observed to be highly resolved and sensitive to base pair‐dependent axial ligation of Ti‐O bond as compared to the corresponding ensemble‐averaged spectral properties, which affect the groove binding and its strength of the O=Ti(TMPyP)⁴⁺ with DNA. The axial ligation was found to be accompanied by vibration structural change of the porphyrin ring, leading to keep the shape of double stranded poly[d(A‐T)₂] rigid while poly‐[d(G‐C)₂] and calf thymus DNA flexible after binding with the oxo‐titanyl porphyrin. The base pair dependence of the fluorescence decay times of the DNA‐bound porphyrins was also observed, implying that an excited‐state charge transfer takes place in the G‐C rich major groove in calf thymus DNA. These results suggest that binding of O=Ti(TMPyP)⁴⁺ is more preferential with the G‐C rich major groove than with the A‐T rich minor groove in calf thymus DNA so that the morphology of DNA is changed.     

Efficacité de l'estimateur d'erreur de Verfürt h pour la solution EF de l'équation de Helmholtz

The finite element solution of Helmholtz equation is dispersive and the existent a posteriori error estimators underestimate the pollution error implied by this phenomenon. In this paper, a new type of residual estimator for the Helmholtz operator is proposed and tested on a one-dimensional model problem. The numerical results show that the error is correctly estimated but new difficulties appear in the exact evaluation of problem dependent constants.     

Fuzzy robust estimation of central location

The problem of a new robust algorithm for estimation of central location has been described in a mathematically simpler way using the fuzzy sets theory. It was compared with ordinary mean estimator and other robust estimators - median, 5% trimmed mean and Huber-, Tukey-, Hampel-, and Andrews-type M-estimators. The performance of Fuzzy 1-means algorithm (FM) proposed is demonstrated by applying it to different data sets from published literature and has been shown to exceed that of conventional ordinary mean estimator and equals or often exceeds that of the most robust estimators.     

The degree of crystallinity of monoclinic isotactic poly(propylene)

The degree of crystallinity of a set of monoclinic (alpha) isotactic poly(propylenes), prepared by a metallocene‐type catalyst, were determined at room temperature. Three different methods were used: density, enthalpy of fusion, and wide‐angle X‐ray scattering, and the results compared. The relation between the heat of fusion and the specific volume of these poly(propylenes) was found to be nonlinear, thus precluding any linear extrapolation to obtain the heat of fusion of the pure crystal (ΔH_u). The value of ΔH_u obtained from depression of the melting temperature by diluents is used. Based on the unit cell density of monoclinic crystals formed from a low defected fraction, the density obtained crystallinity levels were found to be between 0.l5–0.25 higher than those calculated from the heat of fusion. This relatively large difference holds for the isothermally crystallized and quenched isotactic poly(propylenes), and reflects the contribution of the interphase to the density determined crystallinity, which does not contribute to the heat of fusion. Paralleling results found in other systems, the crystallinity levels obtained from wide‐angle X‐ray scattering agree with those obtained from density, indicating a significant contribution of the partially ordered phase to the total diffraction. Emphasis is given on the need to account for the large differences in the crystallinities of poly(propylene) measured by different techniques when evaluating the dependence of properties on this quantity. © 1999 John Wiley & Sons, Inc. J Polym Sci B: Polym Phys 37: 323–334, 1999     

A Methodology for Estimating Kinetic Parameters and Reactivity Ratios of Multi‐site Type Catalysts Using Polymerization,Fractionation, and Spectroscopic Techniques

This work describes a polymer reaction engineering framework for understanding how catalyst kinetic parameters affect the microstructure of polyolefins made with single‐ or multi‐site catalysts. Moreover, a methodology for deconvolution and kinetic parameters estimation is presented to estimate the reactivity ratios of multi‐site catalysts based on the combination of polymerization, fractionation, and spectroscopic techniques, namely, gel permeation chromatography‐IR and carbon‐13 nuclear magnetic resonance spectroscopy. The methodology capabilities are then demonstrated and validated using a case study simulated via a Monte Carlo model including random noise in order to better represent experimental result uncertainties. The methodology can reverse engineer experimental results and estimate all relevant reaction performance parameters.     

Wide‐angle X‐ray diffraction and differential scanning calorimetry study of the crystallization of poly(ethylene naphthalate), poly(butylene naphthalate), and their copolymers

The crystallization behavior of a series of poly(ethylene‐co‐butylene naphthalate) (PEBN) random copolymers was studied. Wide‐angle X‐ray diffraction (WAXD) patterns showed that the crystallization of these copolymers could occur over the entire range of compositions. This resulted in the formation of poly(ethylene naphthalate) or poly(butylene naphthalate) crystals, depending on the composition of the copolymers. Sharp diffraction peaks were observed, except for 50/50 PEBN. Eutectic behavior was also observed. This showed isodimorphic cocrystallization of the PEBN copolymers. The variation of the enthalpy of fusion of the copolymers with the composition was estimated. The isothermal and nonisothermal crystallization kinetics were studied. The crystallization rates were found to decrease as the comonomer unit content increased. The tensile properties were also measured and were found to decrease as the butylene naphthalate content of the copolymers increased. For initially amorphous specimens, orientation was proved by WAXD patterns after drawing, but no crystalline reflections were observed. However, the fast crystallization of drawn specimens occurred when they were heated above the glass‐transition temperature. © 2004 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 42: 843–860, 2004     

An ESR Approach to the Estimation of the Rate Constants of the Addition and Fragmentation Processes Involved in the RAFT Polymerization of Styrene

The reversible‐addition‐fragmentation chain transfer (RAFT) controlled radical polymerization of such vinylic monomers as styrene (= ethenylbenzene) has gained increasing popularity in current years. While there is a general agreement on the mechanism of RAFT polymerization, there is an ongoing debate about the values of the rate constants of its key steps, i.e., the addition of the propagating radicals to the mediator and the fragmentation of the resulting spin adducts. By carrying out an ESR spectroscopic investigation of the AIBN‐initiated polymerization of styrene (AIBN = 2,2′‐azobis[2‐methylpropanenitrile]), mediated by benzyl (diethoxyphosphoryl)dithioformate ( 5 ) as RAFT agent, we were able to detect and characterize four different radical species involved in the process. By reproducing their concentration–time profiles through a kinetic model, the addition and fragmentation rate constants at 90° of the propagating radicals to and from the mediator were estimated to be ca.10⁷ M ^?1 s^?1 and ca. 10³ s^?1, respectively. The validity of the kinetic model was supported by hybrid meta DFT calculations with the BB1K functional that predicted addition‐ and fragmentation‐rate‐constant values in good agreement with those estimated from the ESR experiments.     

Proteomics analysis of altered proteins in kidney of mice with aristolochic acid nephropathy using the fluorogenic derivatization–liquid chromatography–tandem mass spectrometry method

Aristolochic acid (AA) causes interstitial renal fibrosis, called aristolochic acid nephropathy (AAN). There is no specific indicator for diagnosing AAN, so this study aimed to investigate the biomarkers for AAN using a proteomics method. The C3H/He female mice were given ad libitum AA–distilled water (0.5 mg/kg/day) and distilled water for 56 days in the AA and normal groups, respectively. The AA‐induced proteins in the kidney were investigated using a proteomics study, including fluorogenic derivatization with 7‐chloro‐N‐[2‐(dimethylamino)ethyl]‐2,1,3‐benzoxadiazole‐4‐sulfonamide, followed by high‐performance liquid chromatography analysis and liquid chromatography tandem mass spectrometry with a MASCOT database searching system. There were two altered proteins, thrombospondin type 1 (TSP1) and G protein‐coupled receptor 87 (GPR87), in the kidney of AA‐group mice on day 56. GPR87, a tumorigenesis‐related protein, is reported for the first time in the current study. The renal interstitial fibrosis was certainly induced in the AA‐group mice under histological examination. Based on the results of histological examination and the proteomics study, this model might be applied to AAN studies in the future. TSP1 might be a novel biomarker for AAN, and the further role of GPR87 leading to AA‐induced tumorigenesis should be researched in future studies.     

Simulation of Molecular‐Weight Distribution Using a New Kinetic Model of Ethene Polymerization Catalyzed by Cp2ZrCl2‐MAO

The kinetics of ethene polymerization catalyzed by Cp₂ZrCl₂‐methylaluminoxane (MAO) is studied by applying a new kinetic model. Important kinetic parameters of polymerization were estimated. In addition a method of calculating the molecular‐weight distribution (MWD) of the resultant polyethene was established by developing this new model. The final product is expected to comprise three components, which are produced by different active‐site types, and the MWD of one of the components is less than 2. Good agreement between the estimated value and the variation of polydispersity was achieved.     

Comparative characterization of oligomeric precursors intended for injectable implants

The use of injectable materials is a simple approach for drug delivery and tissue repair, in, e.g. minimally invasive surgery applications. If these materials are used past their glass transition temperature and have a low viscosity, they will be able to flow while delivered in situ. Whether these materials are to be used as low viscosity drug carriers or further crosslinked for tissue repair, there is a need for a better understanding of their handling properties. In this study, oligo(trimethylene carbonate) (oTMC) and oligo[D,L‐lactide‐co‐(ε‐caprolactone)] (oDLLA‐co‐CL) of various molecular weights within a relevant injectability range were synthesized via ring‐opening polymerization. The materials were comparatively characterized by ¹H NMR spectroscopy, differential scanning calorimetry, gel permeation chromatography, and rheological measurements. After comparing the viscosities and molecular weights of the materials, it was concluded that oDLLA‐co‐CLs were, generally, better suited as an injectable in situ crosslinking network, whereas oTMCs were found to be better candidates as injectable drug carriers. This study provides useful data and guidelines on the use of these and other similar oligomers intended for injectable implants. Copyright © 2012 John Wiley & Sons, Ltd.     

1H NMR and UV-visible data fusion for determining Sudan dyes in culinary spices

Two data fusion strategies (variable and decision level) combined with a multivariate classification approach (Partial Least Squares-Discriminant Analysis, PLS-DA) have been applied to get benefits from the synergistic effect of the information obtained from two spectroscopic techniques: UV-visible and ¹H NMR. Variable level data fusion consists of merging the spectra obtained from each spectroscopic technique in what is called “meta-spectrum” and then applying the classification technique. Decision level data fusion combines the results of individually applying the classification technique in each spectroscopic technique. Among the possible ways of combinations, we have used the fuzzy aggregation connective operators. This procedure has been applied to determine banned dyes (Sudan III and IV) in culinary spices. The results show that data fusion is an effective strategy since the classification results are better than the individual ones: between 80 and 100% for the individual techniques and between 97 and 100% with the two fusion strategies.     

Round-robin evaluation of a solid-phase microextraction-gas chromatographic method for reliable determination of trace level ethylene oxide in sterilized medical devices

Medical devices that are sterilized with ethylene oxide (EtO) retain small quantities of EtO residuals, which may cause negative systemic and local irritating effects, and must be accurately quantified to ensure non-toxicity. The goal of this round-robin study is to investigate the capability of a novel solid-phase microextraction-gas chromatographic (SPME-GC) method for trace-level EtO residuals analysis: three independent laboratories conducted a guided experiment using this SPME-GC method, in assessing method performance, ruggedness and the feasibility of SPME fibers. These were satisfactory across the independent laboratories, at the 0.05-5.00 ppm EtO range. This method was then successfully applied to analyze EtO residuals in several sterilized/aerated medical devices of various polymeric composition, reliably detecting and quantifying the trace levels of EtO residuals present ( approximately 0.05 ppm EtO). SPME is a feasible alternative for quantifying trace-level EtO residuals in sterilized medical devices, thereby lowering the limit of quantification (LOQ) by as much as two to three orders of magnitude over the current GC methodology of direct liquid injection.     

Insect Protein Content Analysis in Handcrafted Fitness Bars by NIR Spectroscopy. Gaussian Process Regression and Data Fusion for Performance Enhancement of Miniaturized Cost-Effective Consumer-Grade Sensors

Future food supply will become increasingly dependent on edible material extracted from insects. The growing popularity of artisanal food products enhanced by insect proteins creates particular needs for establishing effective methods for quality control. This study focuses on developing rapid and efficient on-site quantitative analysis of protein content in handcrafted insect bars by miniaturized near-infrared (NIR) spectrometers. Benchtop (Büchi NIRFlex N-500) and three miniaturized (MicroNIR 1700 ES, Tellspec Enterprise Sensor and SCiO Sensor) in hyphenation to partial least squares regression (PLSR) and Gaussian process regression (GPR) calibration methods and data fusion concept were evaluated via test-set validation in performance of protein content analysis. These NIR spectrometers markedly differ by technical principles, operational characteristics and cost-effectiveness. In the non-destructive analysis of intact bars, the root mean square error of cross prediction (RMSEP) values were 0.611% (benchtop) and 0.545–0.659% (miniaturized) with PLSR, and 0.506% (benchtop) and 0.482–0.580% (miniaturized) with GPR calibration, while the analyzed total protein content was 19.3–23.0%. For milled samples, with PLSR the RMSEP values improved to 0.210% for benchtop spectrometer but remained in the inferior range of 0.525–0.571% for the miniaturized ones. GPR calibration improved the predictive performance of the miniaturized spectrometers, with RMSEP values of 0.230% (MicroNIR 1700 ES), 0.326% (Tellspec) and 0.338% (SCiO). Furthermore, Tellspec and SCiO sensors are consumer-oriented devices, and their combined use for enhanced performance remains a viable economical choice. With GPR calibration and test-set validation performed for fused (Tellspec + SCiO) data, the RMSEP values were improved to 0.517% (in the analysis of intact samples) and 0.295% (for milled samples).     

Determination of the crystallinity of polyethylene/α‐olefin copolymers by thermal analysis: Relationship of the heat of fusion of 100% polyethylene crystal and the density

The heat of fusion measured with differential scanning calorimetry (DSC) is typically divided by a constant value of the heat of fusion of 100% polyethylene (PE) crystal (ΔH) for the estimation of the fraction crystallinity of PE copolymers, regardless of the density [i.e., the short‐chain branching (SCB) concentration]. In this work, values of ΔH of about 288 J/g were determined with a combined DSC and X‐ray diffraction (XRD) method for a series of PE copolymers containing SCB from 0 to 50 Br/1000 C (density = 0.965–0.865 g/cc). There was no systematic change in ΔH observed across this density range. This result supports the suitability of determining the fraction crystallinity of PE of any density by the simple division of the observed heat of fusion determined by DSC by a constant value of ΔH. This DSC method yielded values of PE crystallinity in good agreement with corresponding values determined by XRD for a series of PE copolymers. The determination of ΔH involved a small precision error for higher density (lower SCB) PEs, but the precision error increased for lower density (i.e., higher SCB) PEs. This was due to the difficulty in measuring the heat of fusion for lower density PEs, which exhibited low values of the heat of fusion and melted only slightly above room temperature, and due to the difficulty of measuring lower values of crystallinity by XRD. The crystal thickness measured by small‐angle X‐ray scattering for this series of PE copolymers decreased exponentially from about 280 to 6 Å. © 2002 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 40: 1637–1643, 2002