Search

For the first time, a global regression quantitative structure–toxicity/activity relationship (QSTR/QSAR) model was developed for the toxicity of a large data set including 1236 chemicals towards Vibrio fischeri, by using random forest (RF) regression algorithm. The optimal RF model with RF parameters of mtry = 3, ntree = 150 and nodesize = 5 was based on 13 molecular descriptors. It can achieve accurate prediction for the toxicity of 99.1% of 1236 chemicals, and yield coefficients of determination R2 of 0.893 for 930 log(Mw/IBC50) in the training set, 0.723 for 306 log(Mw/IBC50) in the test se, and 0.865 for 1236 toxicity log(Mw/IBC50) in the total set. The optimal RF global model proposed in this work is comparable to other published local QSTR models on small datasets of the toxicity to Vibrio fischeri. [ABSTRACT FROM AUTHOR]

Due to the complexity of objective world, as well as the ambiguity of human thinking, the practical decision-making issues become more and more difficult. Pythagorean fuzzy set is an effective tool for depicting uncertainty of the multi-criteria decision-making problems. This study aims to develop a Pythagorean fuzzy multi-criteria decision-making approach to deal with decision-making problem under uncertainty circumstance. Firstly, the concept, representation and related properties of Spearman rank correlation coefficient (SRCC) originated from statistical theory between two PFSs are introduced, which is used to measure the closeness degree between ideal alternative and each alternative. Then, a multi-criteria decision-making approach with Pythagorean fuzzy environment is developed based on the proposed SRCC. Finally, to illustrate the applicability and effectiveness of the proposed method, a real-world infrastructure project decision-making was demonstrated. The result shows that the main advantage of the proposed decision rule would reduce the complexity of the decision-making problem both in theory and practice. [ABSTRACT FROM AUTHOR]

The nonnegative matrix factorization (NMF) has been widely used because it can accomplish both feature representation learning and dimension reduction. However, there are two critical and challenging issues affecting the performance of NMF models. One is the selection of matrix factorization rank, while most of the existing methods are based on experiments or experience. For tackling this issue, an adaptive and stable NMF model is constructed based on an adaptive factorization rank selection (AFRS) strategy, which skillfully and simply integrates a row constraint similar to the generalized elastic net. The other is the sensitivity to the initial value of the iteration, which seriously affects the result of matrix factorization. This issue is alleviated by complementing NMF and deep learning each other and avoiding complex network structure. The proposed NMF model is called deep AFRS-NMF model for short, and the corresponding optimization solution, convergence and stability are analyzed. Moreover, the statistical consistency is discussed between the rank obtained by the proposed model and the ideal rank. The performance of the proposed deep AFRS-NMF model is demonstrated by applying in genetic data-based tumor recognition. Experiments show that the factorization rank obtained by the deep AFRS-NMF model is stable and superior to classical and state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Risk assessment of pollutants to humans and ecosystems requires much toxicological data. However, experimental testing of compounds expends a large number of animals and is criticized for ethical reasons. The in silico method is playing an important role in filling the data gap. In this paper, the acute toxicity data of 1221 chemicals to Vibrio fischeri were collected. The global models obtained showed that there was a poor relationship between the toxicity data and the descriptors calculated based on linear and nonlinear regression analysis. This is due to the fact that the studied compounds contain not only non-reactive compounds but also reactive and specifically acting compounds with different modes of action (MOAs). MOAs are fundamental for the development of mechanistically based QSAR models and toxicity prediction. To investigate MOAs and develop MOA-based prediction models, the compounds were classified into baseline, less inert, reactive, and specifically acting compounds based on the modified Verhaar's classification scheme. Satisfactory models were established by multivariate linear regression (MLR) and support vector machine (SVM) analysis not only for baseline and less inert chemicals, but also for reactive and specifically acting compounds. Compared with linear models obtained by the MLR method, the nonlinear models obtained by the SVM method had better performance. The cross validation proved that all of the models were robust except for those for reactive chemicals with nN (number of nitrogen atoms) = 0 and n(C=O) (number of carbonyl groups) > 0 (Q2ext < 0.5). The application domains and outliers are discussed for those MOA-based models. The models developed in this paper are significantly helpful not only because the application domains for baseline and less inert compounds have been expended, but also the toxicity of reactive and specifically acting compounds can be successfully predicted. This work will promote understanding of toxic mechanisms and toxicity prediction for the chemicals with structural diversity, especially for reactive and specifically acting compounds. [ABSTRACT FROM AUTHOR]

The rapid development of the economy in China resulted in increasingly serious water pollution problem. A lot of water pollution treatment projects have been launched to improve the water environment quality. Water pollution treatment is a complex and long-term task. Considering the concept of water pollution with fuzziness and the factors affecting the effect of water pollution treatment (EWPT), this study constructs a multi-dimensional dynamic fuzzy comprehensive monitoring model. The model considers the vague boundaries in the representation of water pollution and various factors affecting the treatment effect, such as monitoring time, monitoring index, and monitoring location. In detail, firstly, existing methods for evaluating the EWPT are analyzed and reviewed. Then a multi-dimensional dynamic model is developed for monitoring the EWPT. Finally, the Yueya Lake of Henan Province in China is taken as an example to demonstrate the effectiveness and practicability of the proposed method. From the analysis of the results, to maintain the cleanliness of the water, efforts should still be made to eliminate and completely block the pollutants on the shore in order to fundamentally solve the problem. [ABSTRACT FROM AUTHOR]

The internal concentration represented by the critical body residue (CBR) is an ideal indicator to reflect the intrinsic toxicity of a chemical. Whilst some studies have been performed on CBR, the effect of exposure route on internal toxicity has not been investigated for fish. In this paper, acute toxicity data to fish comprising LC and LD values were used to investigate CBR. The results showed that exposure route can significantly affect the internal concentration. LD and CBR calculated from LC and BCF both vary independently of hydrophobicity as expressed by log K; conversely, LC is related to log K. A poor relationship was observed between LC and LD, but the relationship can be improved significantly by introduction of log K because log CBR is positively related to log LD. The parallel relationship of log CBR-log K and log LD-log K indicates that LD does not reflect the actual internal concentration. The average LD is close to the average CBR for less inert and reactive compounds, but greater than the average CBR for baseline compounds. This difference is due to the lipid fraction being the major storage site for most of the baseline compounds. Investigation on the calculated and observed CBRs shows that calculated CBRs are close to observed CBRs for most of compounds. However, systemic deviations of calculated CBRs have been observed for some compounds. The reasons for these systemic deviations may be attributed to BCF, equilibrium time and experimental error of LC. These factors are important and should be considered in the calculation of CBRs. [ABSTRACT FROM AUTHOR]

Quantitative structure-activity relationships (OSARs) for the toxicity of substituted phenols, anilines and nitrobenzenes to algae (Scenedesmus obliquus) have been studied. Chemicals were divided into polar narcotics and electrophiles based on structure criteria according to the toxicity mechanisms of action. A good relationship between the toxicity and octanol/water partition coefficient (logP) was observed for the polar narcotic compounds. The results confirmed that hydrophobicity is the determining physicochemical parameter in modelling narcotics. However, no statistically significant mono-parametric QSAR models could be made from descriptors calculated in this study for all the compounds. In order to investigate the mechanisms of toxic action and develop OSAR models for all compounds, Abraham solvation descriptors were calculated from the program of Pharma-Algorithms. Stepwise regression produced a three-parameter QSAR model (polarizability/dipolarity, hydrogen-bond basicity and molecular volume) with a good statistical fit for all the compounds studied, which had square of the correlation coefficient R²=0.83. We have also developed an interpretable QSAR model by using parameter of logP, which represents the bio-uptake process, and polarizability/dipolarity (8), which represents the irteraction of chemicals with the acceptor micromolecules (target receptor) at the active site. No hydrogen-bond interaction was observed between the chemicals and receptor micromolecules. [ABSTRACT FROM AUTHOR]