Classification study of skin sensitizers based on support vector machine and linear discriminant analysis

SkinSensDB: a curated database for skin sensitization assays

Skin sensitization is an important toxicological endpoint for chemical hazard determination and safety assessment. Prediction of chemical skin sensitizer had traditionally relied on data from rodent models. The development of the adverse outcome pathway (AOP) and associated alternative in vitro assays have reshaped the assessment of skin sensitizers. The integration of multiple assays as key events in the AOP has been shown to have improved prediction performance. Current computational models to predict skin sensitization mainly based on in vivo assays without incorporating alternative in vitro assays. However, there are few freely available databases integrating both the in vivo and the in vitro skin sensitization assays for development of AOP-based skin sensitization prediction models. To facilitate the development of AOP-based prediction models, a skin sensitization database named SkinSensDB has been constructed by curating data from published AOP-related assays. In addition to providing datasets for developing computational models, SkinSensDB is equipped with browsing and search tools which enable the assessment of new compounds for their skin sensitization potentials based on data from structurally similar compounds. SkinSensDB is publicly available at http://cwtung.kmu.edu.tw/skinsensdb.

Integrated Computational Solution for Predicting Skin Sensitization Potential of Molecules

Introduction

Skin sensitization forms a major toxicological endpoint for dermatology and cosmetic products. Recent ban on animal testing for cosmetics demands for alternative methods. We developed an integrated computational solution (SkinSense) that offers a robust solution and addresses the limitations of existing computational tools i.e. high false positive rate and/or limited coverage.

Results

The key components of our solution include: QSAR models selected from a combinatorial set, similarity information and literature-derived sub-structure patterns of known skin protein reactive groups. Its prediction performance on a challenge set of molecules showed accuracy = 75.32%, CCR = 74.36%, sensitivity = 70.00% and specificity = 78.72%, which is better than several existing tools including VEGA (accuracy = 45.00% and CCR = 54.17% with ‘High’ reliability scoring), DEREK (accuracy = 72.73% and CCR = 71.44%) and TOPKAT (accuracy = 60.00% and CCR = 61.67%). Although, TIMES-SS showed higher predictive power (accuracy = 90.00% and CCR = 92.86%), the coverage was very low (only 10 out of 77 molecules were predicted reliably).

Conclusions

Owing to improved prediction performance and coverage, our solution can serve as a useful expert system towards Integrated Approaches to Testing and Assessment for skin sensitization. It would be invaluable to cosmetic/ dermatology industry for pre-screening their molecules, and reducing time, cost and animal testing.

Comparison of different chemometric methods in quantifying total volatile basic-nitrogen (TVB-N) content in chicken meat using a fabricated colorimetric sensor array

PSO-SVMR is an efficient chemometric tool to quantify TVB-N content in chicken.

A quantitative prediction of the viscosity of ionic liquids using Sσ-profilemolecular descriptors

A QSPR study of ILs using MLR and SVM algorithms based on COSMO-RS molecular descriptors (S_σ-profile).

Predicting full thickness skin sensitization using a support vector machine

To assess the public's propensity for allergic contact dermatitis (ACD), many alternatives to in vivo chemical screening have been developed which generally incorporate a small panel of cell surface and secreted dendritic cell biomarkers. However, given the underlying complexity of ACD, one cell type and limited cellular metrics may be insufficient to predict contact sensitizers accurately. To identify a molecular signature that can further characterize sensitization, we developed a novel system using RealSkin, a full thickness skin equivalent, in co-culture with MUTZ-3 derived Langerhan's cells. This system was used to distinguish a model moderate pro-hapten isoeugenol (IE) and a model strong pre-hapten p-phenylenediamine (PPD) from irritant, salicylic acid (SA). Commonly evaluated metrics such as CD86, CD54, and IL-8 secretion were assessed, in concert with a 27-cytokine multi-plex screen and a functional chemotaxis assay. Data were analyzed with feature selection methods using ANOVA, hierarchical cluster analysis, and a support vector machine to identify the best molecular signature for sensitization. A panel consisting of IL-12, IL-9, VEGF, and IFN-γ predicted sensitization with over 90% accuracy using this co-culture system analysis. Thus, a multi-metric approach that has the potential to identify a molecular signature may be more predictive of contact sensitization.

SVM-RFE based feature selection and Taguchi parameters optimization for multiclass SVM classifier

Recently, support vector machine (SVM) has excellent performance on classification and prediction and is widely used on disease diagnosis or medical assistance. However, SVM only functions well on two-group classification problems. This study combines feature selection and SVM recursive feature elimination (SVM-RFE) to investigate the classification accuracy of multiclass problems for Dermatology and Zoo databases. Dermatology dataset contains 33 feature variables, 1 class variable, and 366 testing instances; and the Zoo dataset contains 16 feature variables, 1 class variable, and 101 testing instances. The feature variables in the two datasets were sorted in descending order by explanatory power, and different feature sets were selected by SVM-RFE to explore classification accuracy. Meanwhile, Taguchi method was jointly combined with SVM classifier in order to optimize parameters C and γ to increase classification accuracy for multiclass classification. The experimental results show that the classification accuracy can be more than 95% after SVM-RFE feature selection and Taguchi parameter optimization for Dermatology and Zoo databases.

On the use of multivariate statistical methods for combining in-stream monitoring data and spatial analysis to characterize water quality conditions in the White River basin, Indiana, USA

Mechanistic hydrologic and water quality models provide useful alternatives for estimating water quality in unmonitored streams. However, developing these elaborate models for large watersheds can be time-consuming and expensive, in addition to challenges that arise during calibration when there is limited spatial and/or temporal monitored in-stream water quality data. The main objective of this research was to investigate different approaches for developing multivariate analysis models as alternative methods for rapidly assessing relationships between spatio-temporal physical attributes of the watershed and water quality conditions in monitored streams, and then using the developed relationships for estimating water quality conditions in unmonitored streams. The study compares the use of various statistical estimates (mean, geometric mean, trimmed mean, and median) of monitored water quality variables to represent annual and seasonal water quality conditions. The relationship between these estimates and the spatial data is then modeled via linear and non-linear multivariate methods. Overall, the non-linear techniques for classification outperformed the linear techniques with an average cross-validation accuracy of 79.7%. Additionally, the geometric mean based models outperformed models based on other statistical indicators with an average cross-validation accuracy of 80.2%. Dividing the data into annual and quarterly datasets also offered important insights into the behavior of certain water quality variables impacted by seasonal variations. The research provides useful guidance on the use and interpretation of the various statistical estimates and statistical models for multivariate water quality analyses.

Prediction of melatonin excretion patterns in the rat exposed to ELF magnetic fields based on support vector machine and linear discriminant analysis

Bioeffects of magnetic field exposure have been motivated accomplishing various studies. However, no consensus or guideline is available for experimental designs relating exposure conditions as yet. In the present work, in order to analyze and predict the melatonin excretion patterns in the rat exposed to extremely low frequency magnetic fields (ELF-MF), linear discriminate analysis (LDA) and support vector machines (SVMs) were utilized. Subsequently, performances of LDA and SVMs were compared through resubstitution and jackknife tests on a database containing 33 experiments. Predictor variables were more effective parameters including frequency, polarization, exposure duration and strength of magnetic fields. Also, five performance measures including accuracy, sensitivity, specificity, matthew's correlation coefficient (MCC) and normalized percentage better than random (S) were used to evaluate the performance of models. The LDA as a conventional model obtained poor prediction performance. On the other hand, SVMs as a more powerful model, which has not been introduced in predicting melatonin excretion patterns in the rat exposed to ELF-MF, showed 0.38 value of MCC through jackknife test that confirms the higher reliability of the SVMs.

Prediction of skin sensitization potential using D-optimal design and GA-kNN classification methods

Modelling of skin sensitization data of 255 diverse compounds and 450 calculated descriptors was performed to develop global predictive classification models that are applicable to whole chemical space. With this aim, we employed two automated procedures, (a) D-optimal design to select optimal members of the training and test sets and (b) k-Nearest Neighbour classification (kNN) method along with Genetic Algorithms (GA-kNN Classification) to select significant and independent descriptors in order to build the models. This methodology helped us to derive multiple models, M1-M5, that are stable and robust. The best among them, model M1 (CCR(train) = 84.3%, CCR(test) = 87.2% and CCR(ext) = 80.4%), is based on six neighbours and nine descriptors and further suggests that: (a) it is stable and robust and performs better than the reported models in literature, and (b) the combination of D-optimal design and GA-kNN classification approach is a very promising approach. Consensus prediction based on the models M1-M5 improved the CCR of training, test and external validation datasets by 3.8%, 4.45% and 3.85%, respectively, over M1. From the analysis of the physical meaning of the selected descriptors, it is inferred that the skin sensitization potential of small organic compounds can be accurately predicted using calculated descriptors that code for the following fundamental properties: (i) lipophilicity, (ii) atomic polarizability, (iii) shape, (iii) electrostatic interactions, and (iv) chemical reactivity.

Prediction of skin sensitization with a particle swarm optimized support vector machine

Skin sensitization is the most commonly reported occupational illness, causing much suffering to a wide range of people. Identification and labeling of environmental allergens is urgently required to protect people from skin sensitization. The guinea pig maximization test (GPMT) and murine local lymph node assay (LLNA) are the two most important in vivo models for identification of skin sensitizers. In order to reduce the number of animal tests, quantitative structure-activity relationships (QSARs) are strongly encouraged in the assessment of skin sensitization of chemicals. This paper has investigated the skin sensitization potential of 162 compounds with LLNA results and 92 compounds with GPMT results using a support vector machine. A particle swarm optimization algorithm was implemented for feature selection from a large number of molecular descriptors calculated by Dragon. For the LLNA data set, the classification accuracies are 95.37% and 88.89% for the training and the test sets, respectively. For the GPMT data set, the classification accuracies are 91.80% and 90.32% for the training and the test sets, respectively. The classification performances were greatly improved compared to those reported in the literature, indicating that the support vector machine optimized by particle swarm in this paper is competent for the identification of skin sensitizers.

Gamma-turn types prediction in proteins using the two-stage hybrid neural discriminant model

Due to the slightly success of protein secondary structure prediction using the various algorithmic and non-algorithmic techniques, similar techniques have been developed for predicting gamma-turns in proteins by Kaur and Raghava [2003. A neural-network based method for prediction of gamma-turns in proteins from multiple sequence alignment. Protein Sci. 12, 923-929]. However, the major limitation of previous methods was inability in predicting gamma-turn types. In a recent investigation we introduced a sequence based predictor model for predicting gamma-turn types in proteins [Jahandideh, S., Sabet Sarvestani, A., Abdolmaleki, P., Jahandideh, M., Barfeie, M, 2007a. gamma-turn types prediction in proteins using the support vector machines. J. Theor. Biol. 249, 785-790]. In the present work, in order to analyze the effect of sequence and structure in the formation of gamma-turn types and predicting gamma-turn types in proteins, we applied novel hybrid neural discriminant modeling procedure. As the result, this study clarified the efficiency of using the statistical model preprocessors in determining the effective parameters. Moreover, the optimal structure of neural network can be simplified by a preprocessor in the first stage of hybrid approach, thereby reducing the needed time for neural network training procedure in the second stage and the probability of overfitting occurrence decreased and a high precision and reliability obtained in this way.