Phát triển và đánh giá QSARs cho thiết bị đầu cuối Ecotoxic: A. Bối cảnh Độc tính dữ liệu B. Hóa chất mô tả dữ liệu C. thống kê Phương pháp III. Vật liệu và phương pháp Hóa chất thử nghiệm A. B. C. dữ liệu sinh học phân tử descriptor D. E. Phân tích thống kê dữ liệu lựa chọn IV. Kết quả A. ban đầu Benzene đáp ứng bề mặt mô hình B. Đánh giá của các bề mặt đáp ứng-Mô hình Benzen C. Kết hợp mô hình đáp ứng bề mặt Benzene. | Section 4 QSARs for Environmental Toxicity and Fate 2004 by CRC Press LLC Chapter 12 Development and Evaluation of QSARs for Ecotoxic Endpoints The Benzene Response-Surface Model for Tetrahymena Toxicity T. Wayne Schultz and Tatiana I. Netzeva contents I. Introduction II. Background A. Toxicity Data B. Chemical Descriptor Data C. Statistical Methods III. Materials and Methods A. Test Chemicals B. Biological Data C. Molecular Descriptors D. Statistical Analyses E. Data Selection IV. Results A. Initial Benzene Response-Surface Model B. Evaluation of the Benzene Response-Surface Model C. Combined Benzene Response-Surface Model V. Discussion References I. introduction As the uses of toxicological-based quantitative structure-activity relationships QSARs move into the arenas of priority setting risk assessment and chemical classification and labeling the demands for a better understanding of the foundations of these QSARs are increasing. Specifically issues of quality transparency domain identification and validation have been recognized as topics of particular interest Schultz and Cronin 2003 . Quality QSAR can only be constructed and validated with quality data but quality in a QSAR is more than a high coefficient of determination. Transparency has several different meanings as it 2004 by CRC Press LLC applies to QSARs. First transparency means that the data both biological and chemical that are used in QSAR development and validation are available for examination. Second models which are developed with descriptors that quantify the pivotal aspects of toxic expression are considered to be mechanistic-based fundamental and more easily interpreted and thus transparent. Transparency can also mean the amount of process information obtainable from the statistical methodology it goes from the black boxes of neural networks to interpretable multiple linear regression. Since the use of a particular QSAR is only valid within its domain Schultz and Cronin 2003 identification of
