Genetic algorithms as a method for variable selection in multiple linear regression and partial least squares regression, with applications to pyrolysis mass spectrometry

A veterinary cerumenomic assay for bovine laminitis identification

Bovine laminitis disorder results in animal welfare and economic concerns in dairy and beef farms worldwide. However, the affected metabolic pathways, pathophysiologic characteristics, and inflammatory mechanisms remain unclear, hampering the development of new diagnostics. Using cerumen (earwax) as a source of volatile metabolites (cerumenomic) that carry valuable biological information has interesting implications for veterinary medicine. Nonetheless, up to now, no applications of veterinary cerumenomic assays have been made to identify bovine laminitis. This work aims to develop a veterinary cerumenomic assay for bovine laminitis identification that is non-invasive, robust, accurate, and sensitive to detecting the metabolic disturbances in bovine volatile metabolome. Twenty earwax samples (10 from healthy/control calves and 10 from laminitis calves) were collected from Nellore cattle, followed by Headspace/Gas Chromatography-Mass Spectrometry (HS/GC-MS) analysis and biomarker selection in two multivariate approaches: semiquantitative (intensity data) and semiqualitative (binary data). Following the analysis, cerumen volatile metabolites were indicated as candidate biomarkers for identifying bovine laminitis by monitoring their intensity or occurrence. In the semiquantitative strategy, the p-cresol presented the highest diagnostic figures of merit (area under the curve: 0.845, sensitivity: 0.700, and specificity: 0.900). Regarding the binary approach, a panel combining eight variables/volatiles, with formamide being the most prominent one, showed an area under the curve, sensitivity, and specificity of 0.97, 0.81, and 0.90, respectively. In summary, this work describes the first veterinary cerumenomic assay for bovine laminitis that indicates new metabolites altered during the inflammatory condition, paving the way for developing laminitis early diagnosis by monitoring the cerumen metabolites.

A Monte Carlo resampling based multiple feature-spaces ensemble (MFE) strategy for consistency-enhanced spectral variable selection

BACKGROUND

Variable selection has gained significant attention as a means to enhance spectroscopic calibration performance. However, existing methods still have certain limitations. Firstly, the selection results are sensitive to the choice of training samples, indicating that the selected variables may not be truly relevant. Secondly, the number of the selected variables is still too large in some situations, and modelling with too many predictors may lead to over-fitting issues. To address these challenges, we propose and implement a novel multiple feature-spaces ensemble (MFE) strategy with the least absolute shrinkage and selection operator (LASSO) method.

RESULTS

The MFE strategy synergizes the advantages of LASSO regression and ensemble strategy, thereby facilitating a more robust identification of key variables. We demonstrated the efficacy of our approach through extensive experimentation on publicly available datasets. The results not only demonstrate enhanced consistency in variable selection but also manifest improved prediction performance compared to benchmark methods.

SIGNIFICANT

The MFE strategy provided a comprehensive framework for conducting variable importance analysis, leading to robust and consistent variable selection. Furthermore, the improved consistency in variable selection contributes to enhanced prediction performance for spectroscopic calibration, making it more robust and accurate.

Detection of terbufos in cases of intoxication by means of entomotoxicological analysis using ATR-FTIR spectroscopy combined with chemometrics

The detection of toxic substances in larvae from carcasses in an advanced stage of decomposition may help criminal expertise in elucidating the cause of death in suspected cases of poisoning. Terbufos (Counter®) or O,O-diethyl-S-[(tert-butylsulfanyl)methyl] phosphorodithioate is an insecticide and systemic nematicide, which has very high toxicity from an acute point of view (oral LD₅₀ in rodents ranging from 1.4 to 9.2 mg/kg) that has been marketed irregularly and indiscriminately in Brazil as a rodenticide, often being used to practice homicides. The present study aims to evaluate the use of attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy to detect traces of terbufos pesticide in fly larvae (Sarcophagidae). ATR-FTIR spectra of scavenger fly larvae from control (n = 31) and intoxicated (n = 80) groups were collected and submitted to chemometric analysis by means of multivariate classification using principal component analysis with quadratic discriminant analysis (PCA-QDA), successive projections algorithm with quadratic discriminant analysis (SPA-QDA) and genetic algorithm with quadratic discriminant analysis (GA-QDA) in order to distinguish between control and intoxicated groups. All discriminant models showed sensitivity and specificity above 90%, with the GA-QDA model showing the best performance with 98.9% sensitivity and specificity. The proposed methodology proved to be sensitive and promising for the detection of terbufos in scavenger fly larvae from intoxicated rat carcasses. In addition, the non-destructive nature of the ATR-FTIR technique may be useful in preserving the forensic evidence, meeting the precepts of the chain of custody and allowing for counter-proof.

Precision Medicine Approaches with Metabolomics and Artificial Intelligence

Recent technological innovations in the field of mass spectrometry have supported the use of metabolomics analysis for precision medicine. This growth has been allowed also by the application of algorithms to data analysis, including multivariate and machine learning methods, which are fundamental to managing large number of variables and samples. In the present review, we reported and discussed the application of artificial intelligence (AI) strategies for metabolomics data analysis. Particularly, we focused on widely used non-linear machine learning classifiers, such as ANN, random forest, and support vector machine (SVM) algorithms. A discussion of recent studies and research focused on disease classification, biomarker identification and early diagnosis is presented. Challenges in the implementation of metabolomics–AI systems, limitations thereof and recent tools were also discussed.

Multivariate strategy for identifying and quantifying jet fuel contaminants by MCR-ALS/PLS models coupled to combined MIR/NIR spectra

The investigation and control of jet fuel contamination for private aircrafts has gained attention due to the softer monitoring in comparison to commercial aviation. The possible contamination with kerosene solvent (KS) makes this investigation more challenging, since it has physicochemical similarities with jet fuel. To help solve this problem, a chemometric methodology was applied in this research combining multivariate curve resolution with alternating least squares (MCR-ALS) and partial least squares (PLS) models coupled to near- and mid-infrared spectroscopies (MIR/NIR) in order to detect and quantify KS in blends with JET-A1 using 23 samples (5-60% v/v). Additionally, 98 samples were stored for 60 days, and principal component analysis, genetic algorithm, and successive projections algorithm were coupled to linear discriminant analysis (PCA-LDA, GA-LDA, and SPA-LDA) in order to classify the blends according to the bands assigned to oxidation products, such as phenols and carboxylic acids. GA-LDA and SPA-LDA models were accurate and reached 100% sensitivity and specificity. Physicochemical analysis was not able to detect the presence of KS in contaminated jet fuel samples, even in high concentrations. The use of MIR-NIR combined spectra improved the quantification results, thus decreasing the experimental error from 5.22% (using only NIR) to 1.64%. PLS regression quantified the content of KS with high accuracy (RMSEP < 1.64%, R² > 0.995). The MCR-ALS model stood out for recovering the spectral profile of kerosene solvent by segregating it from jet fuel spectra. The development of models using chemometric tools contributed to a fast, low-cost, and efficient process for quality control that can be applied in the fuel industry.

A cerumenolomic approach to bovine trypanosomosis diagnosis

INTRODUCTION

Trypanosomiasis caused by Trypanosoma vivax (T. vivax, subgenus Duttonella) is a burden disease in bovines that induces losses of billions of dollars in livestock activity worldwide. To control the disease, the first step is identifying the infected animals at early stages. However, convention tools for animal infection detection by T. vivax present some challenges, facilitating the spread of the pathogenesis.

OBJECTIVES

This work aims to develop a new procedure to identify infected bovines by T. vivax using cerumen (earwax) in a volatilomic approach, here named cerumenolomic, which is performed in an easy, quick, accurate, and non-invasive manner.

METHODS

Seventy-eight earwax samples from Brazilian Curraleiro Pé-Duro calves were collected in a longitudinal study protocol during health and inoculated stages. The samples were analyzed using Headspace/Gas Chromatography-Mass Spectrometry followed by multivariate analysis approaches.

RESULTS

The cerumen analyses lead to the identification of a broad spectrum of volatile organic metabolites (VOMs), of which 20 VOMs can discriminate between healthy and infected calves (AUC = 0.991, sensitivity = 0.967, specificity = 1.000). Furthermore, 13 VOMs can indicate a pattern of discrimination between the acute and chronic phases of the T. vivax infection in the animals (AUC = 0.989, sensitivity = 0.944, specificity = 1.000).

CONCLUSION

The cerumen volatile metabolites present alterations in their occurrence during the T.vivax infection, which may lead to identifying the infection in the first weeks of inoculation and discriminating between the acute and chronic phases of the illness. These results may be a breakthrough to avoid the T. vivax outbreak and provide a faster clinical approach to the animal.

Effect of Cell Morphology on Flexural Behavior of Injection-Molded Microcellular Polycarbonate

The quantitative study of the structure and properties relationship in cellular materials is mostly limited to cell diameter, cell density, skin layer thickness, and cell size distribution. In addition, the investigation of the morphology is generally carried out in two dimensions. Therefore, the interrelation between morphological properties and mechanical characteristics of the foam structure has remained in an uncertain state. In this study, during the physical foaming process, a foam morphology is locally created by using a mold equipped with a core-back insert. The variation in morphology is obtained by modifying the mold temperature, injection flow rate, and blowing agent content in the polymer melt. X-ray microtomography (μCT) is used to acquire the 3D visualization of the cells structure. The Cell Distribution Index (CDI) is calculated to represent the polydispersity in cell size distribution. The relationship between the wide range of morphological qualities and relevant flexural properties is made explicit via a statistical model. According to the results, the morphology, particularly cell shape, characterizes the mechanism of the linear elastic deformation of the closed-cell foams. IR-thermography reveals the bending failure of cellular structures in the tensile region despite the differences in cell diameter.

Application of Metaheuristic Approaches for the Variable Selection Problem

Variable selection is an old topic from regression models. Besides many conventional approaches, some metaheuristic approaches from the realm of optimization such as GA (Genetic Algorithm) or simulated annealing have been suggested to date. These methods have a considerable advantage to deal with many problems over the classical methods, but they must control relevant fine-tuning parameters associated with cross-over or mutation, which can be difficult and time-consuming. In this paper, Jaya, one of several parameter-free approaches will be suggested and explored. Several metaheuristic methods will be compared using results from a real-world dataset and a simulated dataset. The impact of using local search will be analyzed.

Building robust models for identification of adulteration in olive oil using FT-NIR, PLS-DA and variable selection

Being a product with a high market value, olive oil undergoes adulterations. Therefore, studies that make the verification of the authenticity of olive oil more efficient are necessary. The aim of this study was to develop a robust model using FT-NIR and PLS-DA to discriminate extra virgin olive oil samples and build individual models to differentiate adulterated extra virgin olive oil samples. The best PLS-DA-OPS classification model for olive oils showed specificity (Spe) and accuracy (Acc) values higher than 99.7% and 99.9%. For the classification of adulterants, PLS-DA-OPS models presented values of Spe at 96.0% and Acc above 95.5% for varieties. For the blend, the best PLS-DA-GA model presented Acc and Spe values greater than 98.2% and 98.8%. Reliable and robust models have been built, allowing differentiation from seven adulterants to genuine extra virgin olive oils.

QSAR and molecular docking studies of lethal factor protease inhibitors against Bacillus anthracis

Bacillus anthracis is considered as a biological warfare agent because it is the causative agent of the serious infectious anthrax disease. Delay in treatment leads to lethal factor-mediated toxaemia which is very critical due to lack of therapeutic options. Consequently, attempts have been made to discover potent lethal factor (LF) protease inhibitors such as small-molecule synthetic 2-thio-1,3-thiazolidine-4-one (rhodanine) compounds. But computed descriptor-based quantitative structure-activity relationship (QSAR) and drug design studies on such aspect are poorly represented. Therefore, an attempt was made for developing QSAR models using structural descriptors for 1,3-thiazolidine-4-one compounds. The models were developed on a series of 49 LF protease inhibitors using the combination of constitutional, functional group, atom-centred fragment and molecular property descriptors. The best QSAR model included four variables, namely, C-040, nR05, GVWAI-80 and ALOGP that correlated well with the anti-LF protease activity with a good correlation coefficient (r = 0.870) of good statistical significance (F_{4, 29} = 14.09 (α = 0.001) F_{4, 29} = 6.19). This model was also validated and explained 58.1% of variances of the Bacillus anthracis inhibitory activities of the studied compounds with r²_pred = 0.710 which denotes external predictability. Finally, molecular docking was carried out to predict the mode of binding of some highly active congeneric compounds. It was shown that VAL 1403 is an important residue for phenyl ring. TYR 1456 and HIS 1418 are responsible for interaction with the rhodanine nucleus. Therefore, these residues are considered responsible for the inhibition of LF protease anthrax and can predict significant dimension of essential structural features of these inhibitors to evaluate, screen and help priorities of the synthesis of the candidates against anthrax bioterrorism.

Conditional Granger Causality and Genetic Algorithms in VAR Model Selection

Overcoming symmetry in combinatorial evolutionary algorithms is a challenge for existing niching methods. This research presents a genetic algorithm designed for the shrinkage of the coefficient matrix in vector autoregression (VAR) models, constructed on two pillars: conditional Granger causality and Lasso regression. Departing from a recent information theory proof that Granger causality and transfer entropy are equivalent, we propose a heuristic method for the identification of true structural dependencies in multivariate economic time series. Through rigorous testing, both empirically and through simulations, the present paper proves that genetic algorithms initialized with classical solutions are able to easily break the symmetry of random search and progress towards specific modeling.

A weight of evidence approach to investigate potential common mechanisms in pesticide groups to support cumulative risk assessment: A case study with dinitroaniline pesticides

In 2016, the United States Environmental Protection Agency's (EPA) Office of Pesticide Programs published guidelines for establishing candidate common mechanism groups (CMGs) for cumulative risk assessment (CRA) weight-of-evidence-based screenings. A candidate CMG is a group of chemicals that may share similar structure, apical endpoints, and/or mechanistic data that suggest the potential for a common mechanism of toxicity among them. Here, a weight-of-evidence approach is presented to establish candidacy of a CMG for a group of nine dinitroaniline pesticides. This approach involves review of available in vivo toxicity information and literature to determine mode of action, along with analyses of in vitro toxicity data and chemical structure. Despite structural similarity among some dinitroanilines and some shared target organs identified through toxicity observed in in vivo studies, there were no consistencies among groups, suggesting lack of a common mechanism when all analyses are considered together. For example, two structurally similar compounds with thyroid/liver in vivo effects were not found active in any Toxicity Forecaster (ToxCast) in vitro assays. The weight-of-evidence is insufficient to support the testable hypothesis that dinitroanilines could form a CMG, and highlights the importance of establishing a consensus among multiple lines of evidence prior to CRA.

Sex and Age Don't Matter, but Breed Type Does-Factors Influencing Eye Wrinkle Expression in Horses

Identifying valid indicators to assess animals' emotional states is a critical objective of animal welfare science. In horses, eye wrinkles above the eyeball have been shown to be affected by pain and other emotional states. From other species we know that individual characteristics, e.g., age in humans, affect facial wrinkles, but it has not yet been investigated whether eye wrinkle expression in horses is systematically affected by such characteristics. Therefore, the aim of this study was to assess how age, sex, breed type, body condition, and coat colour affect the expression and/or the assessment of eye wrinkles in horses. To this end, we adapted the eye wrinkle assessment scale from Hintze et al. (1) and assessed eye wrinkle expression in pictures taken from the left and the right eye of 181 horses in a presumably neutral situation, using five outcome measures: a qualitative first impression reflecting how worried the horse is perceived by humans, the extent to which the brow is raised, the number of wrinkles, their markedness and the angle between a line through both corners of the eye and the topmost wrinkle. All measures could be assessed highly reliable with respect to intra- and inter-observer agreement. Breed type affected the width of the angle [F (2,114) = 8.20, p < 0.001], with thoroughbreds having the narrowest angle (M = 23.80, SD = 1.60), followed by warmbloods (M = 28.00, SD = 0.60), and coldbloods (M = 31.00, SD = 0.90). None of the other characteristics affected any of the outcome measures, and eye wrinkle expression did not differ between the left and the right eye area (all p-values > 0.05). In conclusion, horses' eye wrinkle expression and its assessment in neutral situations was not systematically affected by the investigated characteristics, except for "breed type", which accounted for some variation in "angle"; how much eye wrinkle expression is affected by emotion or perhaps mood needs further investigation and validation.

Stacked ensemble extreme learning machine coupled with Partial Least Squares-based weighting strategy for nonlinear multivariate calibration

With its simple theory and strong implementation, extreme learning machine (ELM) becomes a competitive single hidden layer feed forward networks for nonlinear multivariate calibration in chemometrics. To improve the generalization and robustness of ELM further, stacked generalization is introduced into ELM to construct a modified ELM model called stacked ensemble ELM (SE-ELM). The SE-ELM is to create a set of sub-models by applying ELM repeatedly to different sub-regions of the spectra and then combine the predictions of those sub-models according to a weighting strategy. Three different weighting strategies are explored to implement the proposed SE-ELM, such as the Winner-takes-all (WTA) weighting strategy, the constraint non-negative least squares (CNNLS) weighing strategy and the partial least squares (PLS) weighting strategy. Furthermore, PLS is suggested to be selected as the optimal weighting method that can handle the multi-colinearity among the predictions yielded by all the sub-models. The experimental assessment of the three SE-ELM models with different weighting strategies is carried out on six real spectroscopic datasets and compared with ELM, back-propagation neural network (BPNN) and Radial basis function neural network (RBFNN), statistically tested by the Wilcoxon signed rank test. The obtained experimental results suggest that, in general, all the SE-ELM models are more robust and more accurate than traditional ELM. In particular, the proposed PLS-based weighting strategy is at least statistically not worse than, and frequently better than the other two weighting strategies, BPNN, and RBFNN.

Spectral interval combination optimization (ICO) on rapid quality assessment of Solanaceae plant: a validation study

A novel spectral variable selection method, named as interval combination optimization (ICO), was proposed in the previous study of us. In the present study, ICO coupled with near infrared (NIR) spectroscopy was applied to the rapid determination of four primary constituents including total sugar, reducing sugar, total nitrogen and nicotine in Nicotiana plant. Partial least squares regressions was performed after ICO algorithm. The full spectrum was divided into forty equal-width intervals, and the interval with lower root mean squared error of cross-validation was selected for further analysis. As a result, only 155 variables were retained from 1555 variables for each constituent. Particularly, as a variables selection method, ICO improved the prediction accuracy of calibration model and obtained a satisfactory result compared with full-spectrum data. Results revealed that NIR combined with ICO could be efficiently used for rapid analysis of quality associated constituents of Nicotiana plant. Moreover, this study provided a supplementary verification of the proposed variable selection method for the further applications.

Advances in chemometric control of commercial diesel adulteration by kerosene using IR spectroscopy

Adulteration is a recurrent issue found in fuel screening. Commercial diesel contamination by kerosene is highly difficult to be detected via physicochemical methods applied in market. Although the contamination may affect diesel quality and storage stability, there is a lack of efficient methodologies for this evaluation. This paper assessed the use of IR spectroscopies (MIR and NIR) coupled with partial least squares (PLS) regression, support vector machine regression (SVR), and multivariate curve resolution with alternating least squares (MCR-ALS) calibration models for quantifying and identifying the presence of kerosene adulterant in commercial diesel. Moreover, principal component analysis (PCA), successive projections algorithm (SPA), and genetic algorithm (GA) tools coupled to linear discriminant analysis were used to observe the degradation behavior of 60 samples of pure and kerosene-added diesel fuel in different concentrations over 60 days of storage. Physicochemical properties of commercial diesel with 15% kerosene remained within conformity with Brazilian screening specifications; in addition, specified tests were not able to identify changes in the blends' performance over time. By using multivariate classification, the samples of pure and contaminated fuel were accurately classified by aging level into two well-defined groups, and some spectral features related to fuel degradation products were detected. PLS and SVR were accurate to quantify kerosene in the 2.5-40% (v/v) range, reaching RMSEC < 2.59% and RMSEP < 5.56%, with high correlation between real and predicted concentrations. MCR-ALS with correlation constraint was able to identify and recover the spectral profile of commercial diesel and kerosene adulterant from the IR spectra of contaminated blends.

Intelligence Algorithms for Protein Classification by Mass Spectrometry

Mass spectrometry (MS) is an important technique in protein research. Effective classification methods by MS data could contribute to early and less-invasive diagnosis and also facilitate developments in the bioinformatics field. As MS data is featured by high dimension, appropriate methods which can effectively deal with the large amount of MS data have been widely studied. In this paper, the applications of methods based on intelligence algorithms have been investigated. Firstly, classification and biomarker analysis methods using typical machine learning approaches have been discussed. Then those are followed by the Ensemble strategy algorithms. Clearly, simple and basic machine learning algorithms hardly addressed the various needs of protein MS classification. Preprocessing algorithms have been also studied, as these methods are useful for feature selection or feature extraction to improve classification performance. Protein MS data growing with data volume becomes complicated and large; improvements in classification methods in terms of classifier selection and combinations of different algorithms and preprocessing algorithms are more emphasized in further work.

Wisdom of artificial crowds feature selection in untargeted metabolomics: An application to the development of a blood-based diagnostic test for thrombotic myocardial infarction

INTRODUCTION

Heart disease remains a leading cause of global mortality. While acute myocardial infarction (colloquially: heart attack), has multiple proximate causes, proximate etiology cannot be determined by a blood-based diagnostic test. We enrolled a suitable patient cohort and conducted a non-targeted quantification of plasma metabolites by mass spectrometry for developing a test that can differentiate between thrombotic MI, non-thrombotic MI, and stable disease. A significant challenge in developing such a diagnostic test is solving the NP-hard problem of feature selection for constructing an optimal statistical classifier.

OBJECTIVE

We employed a Wisdom of Artificial Crowds (WoAC) strategy for solving the feature selection problem and evaluated the accuracy and parsimony of downstream classifiers in comparison with traditional feature selection techniques including the Lasso and selection using Random Forest variable importance criteria.

MATERIALS AND METHODS

Artificial Crowd Wisdom was generated via aggregation of the best solutions from independent and diverse genetic algorithm populations that were initialized with bootstrapping and a random subspaces constraint.

RESULTS/CONCLUSIONS

Strong evidence was observed that a statistical classifier utilizing WoAC feature selection can discriminate between human subjects presenting with thrombotic MI, non-thrombotic MI, and stable Coronary Artery Disease given abundances of selected plasma metabolites. Utilizing the abundances of twenty selected metabolites, a leave-one-out cross-validation estimated misclassification rate of 2.6% was observed. However, the WoAC feature selection strategy did not perform better than the Lasso over the current study.

Wavelength selection for portable noninvasive blood component measurement system based on spectral difference coefficient and dynamic spectrum

Noninvasive blood component analysis by spectroscopy has been a hotspot in biomedical engineering in recent years. Dynamic spectrum provides an excellent idea for noninvasive blood component measurement, but studies have been limited to the application of broadband light sources and high-resolution spectroscopy instruments. In order to remove redundant information, a more effective wavelength selection method has been presented in this paper. In contrast to many common wavelength selection methods, this method is based on sensing mechanism which has a clear mechanism and can effectively avoid the noise from acquisition system. The spectral difference coefficient was theoretically proved to have a guiding significance for wavelength selection. After theoretical analysis, the multi-band spectral difference coefficient-wavelength selection method combining with the dynamic spectrum was proposed. An experimental analysis based on clinical trial data from 200 volunteers has been conducted to illustrate the effectiveness of this method. The extreme learning machine was used to develop the calibration models between the dynamic spectrum data and hemoglobin concentration. The experiment result shows that the prediction precision of hemoglobin concentration using multi-band spectral difference coefficient-wavelength selection method is higher compared with other methods.

Determination of main fruits in adulterated nectars by ATR-FTIR spectroscopy combined with multivariate calibration and variable selection methods

Grape, orange, peach and passion fruit nectars were formulated and adulterated by dilution with syrup, apple and cashew juices at 10 levels for each adulterant. Attenuated total reflectance Fourier transform mid infrared (ATR-FTIR) spectra were obtained. Partial least squares (PLS) multivariate calibration models allied to different variable selection methods, such as interval partial least squares (iPLS), ordered predictors selection (OPS) and genetic algorithm (GA), were used to quantify the main fruits. PLS improved by iPLS-OPS variable selection showed the highest predictive capacity to quantify the main fruit contents. The selected variables in the final models varied from 72 to 100; the root mean square errors of prediction were estimated from 0.5 to 2.6%; the correlation coefficients of prediction ranged from 0.948 to 0.990; and, the mean relative errors of prediction varied from 3.0 to 6.7%. All of the developed models were validated.

Combinatorial design and virtual screening of potent anti-tubercular fluoroquinolone and isothiazoloquinolone compounds utilizing QSAR and pharmacophore modelling

The virulence of tuberculosis infections resistant to conventional combination drug regimens cries for the design of potent fluoroquinolone compounds to be used as second line antimycobacterial chemotherapeutics. One of the most effective in silico methods is combinatorial design and high throughput screening by a ligand-based pharmacophore prior to experiment. The combinatorial design of a series of 3850 fluoroquinolone and isothiazoloquinolone compounds was then screened virtually by applying a topological descriptor based quantitative structure activity relationship (QSAR) for predicting highly active congeneric quinolone leads against Mycobacterium fortuitum and Mycobacterium smegmatis. The predicted highly active congeneric hits were then subjected to a comparative study between existing lead sparfloxacin with fluoroquinolone FQ hits as well as ACH-702 with predicted active isothiazoloquinolones, utilizing pharmacophore modelling to focus on the mechanism of drug binding against mycobacterial DNA gyrase. Finally, 68 compounds including 34 FQ and 34 isothiazoloquinolones were screened through high throughput screening comprising QSAR, the Lipinski rule of five and ligand-based pharmacophore modelling.

Spectroscopy with computational analysis in virological studies: A decade (2006-2016)

This review presents a retrospective of the studies carried out in the last 10 years (2006–2016) using spectroscopic methods as a research tool in the field of virology. Spectroscopic analyses are sensitive to variations in the biochemical composition of the sample, are non-destructive, fast and require the least sample preparation, making spectroscopic techniques tools of great interest in biological studies. Herein important chemometric algorithms that have been used in virological studies are also evidenced as a good alternative for analyzing the spectra, discrimination and classification of samples. Techniques that have not yet been used in the field of virology are also suggested. This methodology emerges as a new and promising field of research, and may be used in the near future as diagnosis tools for detecting diseases caused by viruses.

•

A retrospective study of 2006–2016 using spectroscopic methods as a research tool in the field of virology.

•

Chemometric algorithms used in virological studies were evidenced.

•

This review emerges as a new and promising field of research in virology.

Investigating the discrimination potential of linear and nonlinear spectral multivariate calibrations for analysis of phenolic compounds in their binary and ternary mixtures and calculation pKa values

Vanillin (VA), vanillic acid (VAI) and syringaldehyde (SIA) are important food additives as flavor enhancers. The current study for the first time is devote to the application of partial least square (PLS-1), partial robust M-regression (PRM) and feed forward neural networks (FFNNs) as linear and nonlinear chemometric methods for the simultaneous detection of binary and ternary mixtures of VA, VAI and SIA using data extracted directly from UV-spectra with overlapped peaks of individual analytes. Under the optimum experimental conditions, for each compound a linear calibration was obtained in the concentration range of 0.61-20.99 [LOD=0.12], 0.67-23.19 [LOD=0.13] and 0.73-25.12 [LOD=0.15] μgmL(-1) for VA, VAI and SIA, respectively. Four calibration sets of standard samples were designed by combination of a full and fractional factorial designs with the use of the seven and three levels for each factor for binary and ternary mixtures, respectively. The results of this study reveal that both the methods of PLS-1 and PRM are similar in terms of predict ability each binary mixtures. The resolution of ternary mixture has been accomplished by FFNNs. Multivariate curve resolution-alternating least squares (MCR-ALS) was applied for the description of spectra from the acid-base titration systems each individual compound, i.e. the resolution of the complex overlapping spectra as well as to interpret the extracted spectral and concentration profiles of any pure chemical species identified. Evolving factor analysis (EFA) and singular value decomposition (SVD) were used to distinguish the number of chemical species. Subsequently, their corresponding dissociation constants were derived. Finally, FFNNs has been used to detection active compounds in real and spiked water samples.

Selection of clinical features for pattern recognition applied to gait analysis

This paper deals with the opportunity of extracting useful information from medical data retrieved directly from a stereophotogrammetric system applied to gait analysis. A feature selection method to exhaustively evaluate all the possible combinations of the gait parameters is presented, in order to find the best subset able to classify among diseased and healthy subjects. This procedure will be used for estimating the performance of widely used classification algorithms, whose performance has been ascertained in many real-world problems with respect to well-known classification benchmarks, both in terms of number of selected features and classification accuracy. Precisely, support vector machine, Naive Bayes and K nearest neighbor classifiers can obtain the lowest classification error, with an accuracy greater than 97 %. For the considered classification problem, the whole set of features will be proved to be redundant and it can be significantly pruned. Namely, groups of 3 or 5 features only are able to preserve high accuracy when the aim is to check the anomaly of a gait. The step length and the swing speed are the most informative features for the gait analysis, but also cadence and stride may add useful information for the movement evaluation.

Dissimilarity based Partial Least Squares (DPLS) for genomic prediction from SNPs

Background

Genomic prediction (GP) allows breeders to select plants and animals based on their breeding potential for desirable traits, without lengthy and expensive field trials or progeny testing. We have proposed to use Dissimilarity-based Partial Least Squares (DPLS) for GP. As a case study, we use the DPLS approach to predict Bacterial wilt (BW) in tomatoes using SNPs as predictors. The DPLS approach was compared with the Genomic Best-Linear Unbiased Prediction (GBLUP) and single-SNP regression with SNP as a fixed effect to assess the performance of DPLS.

Results

Eight genomic distance measures were used to quantify relationships between the tomato accessions from the SNPs. Subsequently, each of these distance measures was used to predict the BW using the DPLS prediction model. The DPLS model was found to be robust to the choice of distance measures; similar prediction performances were obtained for each distance measure. DPLS greatly outperformed the single-SNP regression approach, showing that BW is a comprehensive trait dependent on several loci. Next, the performance of the DPLS model was compared to that of GBLUP. Although GBLUP and DPLS are conceptually very different, the prediction quality (PQ) measured by DPLS models were similar to the prediction statistics obtained from GBLUP. A considerable advantage of DPLS is that the genotype-phenotype relationship can easily be visualized in a 2-D scatter plot. This so-called score-plot provides breeders an insight to select candidates for their future breeding program.

Conclusions

DPLS is a highly appropriate method for GP. The model prediction performance was similar to the GBLUP and far better than the single-SNP approach. The proposed method can be used in combination with a wide range of genomic dissimilarity measures and genotype representations such as allele-count, haplotypes or allele-intensity values. Additionally, the data can be insightfully visualized by the DPLS model, allowing for selection of desirable candidates from the breeding experiments. In this study, we have assessed the DPLS performance on a single trait.

In-silico combinatorial design and pharmacophore modeling of potent antimalarial 4-anilinoquinolines utilizing QSAR and computed descriptors

There are very few studies for combinatorial library design and high throughput screening of 4-anilinoquinoline antimalarial compounds having activities against parasitic strain of P. falciparum. Therefore, an attempt has been made in the present paper to design potent lead compounds in this congener utilizing quantitative structure activity relationship utilizing theoretical molecular descriptors. QSAR models for a series of 4-anilinoquinolines considering various theoretical molecular descriptors including topological, constitutional, geometrical, functional group and atom-centered fragments has been carried out by stepwise forward–backward variable selections assimilating multiple linear regression (MLR) methods showing the topological indices contribute maximum impact on parasitic P. falciparum strain. A combinatorial library of 2160 compounds has been generated and finally, 16 compounds were screened through high throughput screening as promising 4-anilinoquinoline antimalarial hits based on their predicted activities utilizing topological descriptor based validated QSAR model. Highly predicted active compounds were then undergone for pharmacophore modeling to predict mode of binding and to optimize leads having greater affinity towards malarial P. falciparum parasitic strain.

Understanding the foundations of the structural similarities between marketed drugs and endogenous human metabolites

BACKGROUND

A recent comparison showed the extensive similarities between the structural properties of metabolites in the reconstructed human metabolic network ("endogenites") and those of successful, marketed drugs ("drugs").

RESULTS

Clustering indicated the related but differential population of chemical space by endogenites and drugs. Differences between the drug-endogenite similarities resulting from various encodings and judged by Tanimoto similarity could be related simply to the fraction of the bitstrings set to 1. By extracting drug/endogenite substructures, we develop a novel family of fingerprints, the Drug Endogenite Substructure (DES) encodings, based on the ranked frequency of the various substructures. These provide a natural assessment of drug-endogenite likeness, and may be used as descriptors with which to derive quantitative structure-activity relationships (QSARs).

CONCLUSIONS

"Drug-endogenite likeness" seems to have utility, and leads to a simple, novel and interpretable substructure-based molecular encoding for cheminformatics.