Quantitative Structure-Activity/Property Relationship Study Of The Toxicity Of Some Compounds

In general,the rapid development of kinds of compounds is accompanied by an increased burden of environmental toxicity.Therefore,it is especially important to find suitable methods to evaluate the toxicity of different compounds.Compared with the experimental methods,quantitative Structure-Activity/Property Relationship?QSAR/QSPR?has attracted increasing attention and has been widely used in the research field of compound toxicity in recent years for its simplicity,low cost,and easy implementation.QSAR/QSPR approaches have been widely used in physical chemistry,pharmaceutical chemistry,environmental chemistry,toxicology and other research fields.It can mainly analyze the mechanism of action of some known compounds or predict the physicochemical properties of unknown compounds by calculating the descriptors that can represent the molecular structure and further establish the model regarding the relationship between the descriptors and the properties/activities of the compounds.In this thesis,the topic focused on the toxicity of different types of compounds including the nano-materials and organic mixtures using the multiple linear regression?MLR?and radial basis function neural network?RBFNN?methods.Several models were established QSAR model,and the toxic effects were analyzed.The contents are as follows:?1?A further development of the QNAR model to predict the cellular uptake of nanoparticles by pancreatic cancer cells.To accurately predict the cellular uptake values of these compounds,QNAR models were performed by dividing them into three groups.Judging from the attained statistical results,our derived QNAR models have an acceptable overall accuracy and robustness,as well as good predictivity on the external data sets.Moreover,the results of this study provide some insights on how engineered nanomaterial features influence cellular responses.?2?A QSAR study towards predicting the adsorption of environmental pollutants by multi-walled carbon nanotubes.A QSAR modeling study was carried out for predicting the adsorption property of a set of 59 environmental pollutant aromatic compounds into multi-walled carbon nanotubes.The statistical parameters are:?MLR?Training set:N=47,R²=0.886,RMS=0.3374,F=351.71,?=11.75;External test set:N_ext=12,R²=0.894,RMS=0.3654,F=75.01;?RBFNN?Training set:N=47,R²=0.906,RMS=0.2903,F=403.84;External test set:N_ext=12,Q²=0.894,RMS=0.3654,F=75.01.Judging from the attained statistical results,our derived QSAR models have an acceptable overall accuracy and robustness,as well as good predictivity on external data.This QSAR study suggested also that the adsorption ability of these compounds is mainly explained by size,charge and hydrophobia factors.?3?Prediction of the toxicity of binary mixtures by QSAR approach.Organic compounds are often exposed to the environment and have an adverse effect on environment and human health in the form of mixtures rather than single chemicals.We try to establish the reliable and classical quantitative structure-activity relationship?QSAR?models to evaluate the toxicity of 99 binary mixtures.The results are as follows:For the MLR method,Training set:N=79,R²=0.869,LOOq²=0.864,F=165.494,RMS=0.599;External test set:N_ext=20,R²=0.853,q² _ext=0.825,F=30.861,RMS=0.691;For the RBFNN method,Training set:N=79,R²=0.925,LOOq²=0.924,F=950.686,RMS=0.447;External test set:N_ext=20,R²=0.896,q² _ext=0.890,F=155.424,RMS=0.547.The results confirmed that the models built are acceptable and can be used to predict the toxicity of the binary mixtures.?4?Estimation of the toxicity of different substituted aromatic compounds to the aquatic ciliate Tetrahymena pyriformis by QSAR approach.To estimate the toxicities of substituted aromatic compounds to Tetrahymena pyriformis,the QSAR models were established.Unlike other QSAR studies,according to the difference of functional groups?-NO₂,-X?,our overall data set was divided into three groups and further processed,modeled separately.The results of the built models are acceptable,robustness and with good stability and predictability,which proved that the models can predict the toxicity accurately.?5?Prediction of the toxicity of nano metal oxides by QNAR approach.In the present chapter,the computational chemistry tools??NanoBRIDGES?software?was used to predict and evaluate the cytotoxicity of some MOx nanoparticles both in BEAS-2B and RAW 264.7 cell lines.The results are:For the MTS assay in BEAS-2B cells,N=10,R²=0.9248,R² _adj=0.9462,SEE=5.2574,SDEP=10;For the ATP assay in BEAS-2B cells,N=8,R²=0.9390,R² _adj=0.9146,SEE=2.3919,SDEP=3.7443.For the MTS assay in RAW 264.7 cells,N=9,R²=0.8232,R² _adj=0.7171,SEE=3.4252,SDEP=7.7879.For the ATP assay in RAW 264.7 cells,N=11,R²=0.8263,R² _adj=0.7519,SEE=3.8162,SDEP=7.1379.It should be noted that the descriptive data values of the test set data can be brought into the original model to calculate the result value since the data set is small.The calculated values are compared with the original experimental values.and it is found that the establishment of the model is acceptable.