http://hdl.handle.net/10967/107 University of Tartu (Estonia), Institute of Chemistry, Molecular Technology 2026-04-18T12:19:43Z http://hdl.handle.net/10967/269 Some adverse effects of hydroxylated polychlorinated biphenyls (OH-PCBs) in humans are presumed to be initiated via thyroid hormone receptor (TR) binding. Due to the trial-and-error approach adopted for OH-PCB selection in previous studies, experiments designed to test the TR binding hypothesis mostly utilized inactive OH-PCBs, leading to considerable waste of time, effort and other material resources. In this paper, linear discriminant analysis (LDA) and binary logistic regression (LR) were used to develop classification models to group OH-PCBs into active and inactive TR agonists using radial distribution function (RDF) descriptors as predictor variables. The classifications made by both LDA and LR models on the training set compounds resulted in an accuracy of 84.3%, sensitivity of 72.2% and specificity of 90.9%. The areas under the ROC curves, constructed with the training set data, were found to be 0.872 and 0.880 for LDA and LR models, respectively. External validation of the models revealed that 76.5% of the test set compounds were correctly classified by both LDA and LR models. These findings suggest that the two models reported in this paper are good and reliable for classifying OH-PCB congeners into active and inactive TR agonists. 2025-05-07T13:27:28Z http://hdl.handle.net/10967/204 Theoretical investigation of congener-specific soil sorption of polychlorinated biphenyls by DFT computation and potent QSAR analyses Purpose: Few studies have been conducted to understand well the underlying soil sorptive mechanism due to the limited experimental determination for the enormous number of polychlorinated biphenyl (PCB) congeners. The objective of this paper was to obtain further insights into the soil sorption behavior of PCBs with exploration of the sorptive mechanism at the molecular level for sorption affinity, which could be anticipated to help explore the migration fates and assess the bioavailability of PCBs in soil. Materials and methods: Soil sorption coefficients of 52 kinds of PCB congeners were collected in this paper. The geometries of PCBs were fully optimized within Gaussian 03 suite of programs, and 27 molecular descriptors which describe the electronic and thermodynamic properties were finally determined with optimized structures after optimization. The quantitative structure–activity relationships (QSARs) for predicting the soil sorption of PCBs were developed by the combination of density functional theory (DFT) computation and partial least squares analyses, which maximized the correlation between the DFT-calculated properties and soil sorption of PCBs. The QSAR was critically validated with better performance in sensitivity, robustness, interpretation and prediction, and specific description of the applicability domain. Results and discussion: For the successfully developed QSAR, R-y,cum(adj)(2) and Q(cum)(2) were respectively recorded as 0.922 and 0.896, and R(EXT)(2) and Q(EXT)(2) were respectively recorded as 0.905 and 0.914, which demonstrated the stability and predictability of the model. The molecular electronegativity of PCBs by DFT was significantly indicative of a positive correlation with sorption potency, while polarizability had a negative correlation with it. QSAR analyses also revealed the favorable structural requirement of more chlorination at meta/para sites for soil sorption. It was implied that the soil sorption should be largely ascribed to the electrostatic interaction between PCBs and soil organic matter. Nevertheless, the thermodynamic stability and hydrophobicity related to the molecular entropy increment of PCBs were also beneficial to enhancing soil sorption. Conclusions: QSAR analyses particularly indicated the strong dependence of variation of soil sorption on molecular electronic properties, such as electronegativity and polarizability, which suggested the predominance of electrostatic interaction with soil organic matter. Meta/para chlorination was illustrated as a preferable structural requirement for soil sorption. In addition, the thermodynamic stability and hydrophobicity driven by entropy increment were also effective for soil sorption. These results contributed to predict the migration and fate of PCBs in soil system. 2018-06-11T14:39:05Z http://hdl.handle.net/10967/200 The experimental log EC(50) toxicity values of 104 compounds causing bioluminescent repression of the bacterium strain Pseudomonas isolated from an industrial wastewater were studied. Using the Best Multilinear Regression method implemented in CODESSA PRO, models with up to 8 theoretical descriptors were obtained. Utilizing a rigorous descriptor selection and validation procedure a reliable QSAR model with four parameters was selected as best. The proposed model emphasizes the importance of the halogen atoms presented in each compound, the possibility of H-bond formation and the flexibility and degree of branching of the molecules. As pointed out by many researchers, the contribution of the octanol water partition coefficient to the explanation of the toxicity effect was also found to be significant. In addition, the model currently proposed was compared to those reported earlier and its advantages were discussed in detail. 2017-05-23T11:23:07Z http://hdl.handle.net/10967/199 This paper describes the derivation of a simple QSAR model for the prediction of log BB from a set of 55 diverse organic compounds. The model contains two variables: polar surface area (PSA) and calculated logP, both of which can be rapidly computed. It therefore permits the prediction of log BB for large compound sets, such as virtual combinatorial libraries. The performance of this QSAR on two test sets taken from the literature is illustrated and compared with results from other reported computational approaches to log BB prediction. 2017-05-18T13:27:52Z