Tất cả quá trình mô hình yêu cầu dữ liệu. Đối với độc tố tiên đoán, nơi khởi đầu cho tất cả các mô hình là đối chiếu dữ liệu về độc. Chương này sẽ mô tả dữ liệu độc hại có thể đạt được và làm thế nào họ nên được tiếp cận. Người đọc cũng được gọi đến một số giấy tờ khác, nơi mà người đọc có thể có được thêm thông tin (1,2) và các bộ sưu tập tuyệt vời của các giấy tờ trong chất độc được công bố, Khối lượng 157, vấn đề. | 4 Toxicological Information for Use in Predictive Modeling Quality Sources and Databases MARK T. D. CRONIN School of Pharmacy and Chemistry John Moores University Liverpool . 1. INTRODUCTION All modeling processes require data. For predictive toxicology the starting place for all models is the collation of toxicological data. This chapter will describe where toxicological data may be obtained and how they should be approached. The reader is also referred to a number of other papers where the reader may obtain further information 1 2 and the excellent collections of papers published in Toxicology Volume 157 Issues 1-2 3-14 and Volume 173 Issues 1-2 15-28 . It is hoped that this chapter will provide a source of information for 93 2005 by Taylor Francis Group LLC 94 Cronin further investigation. It is well beyond the remit of one chapter and probably the knowledge of one scientist to describe all sources of information. However it is hoped that this chapter will provide a starting place to show how data may be obtained for predictive toxicology. . Fundamentals For the development of models data on a single chemical are normally required. It is generally not possible to create models and make predictions on mixtures of chemicals simply because the ethos behind the modeling process is the association of an activity with the structure and or properties of single chemicals. There are some exceptions to this rule quantitative structure-activity relationship QSAR analyses of surfactants for example have been developed for mixtures by assuming that surfactants may be represented as single structures. Thus for a mixture of surfactants with varying chain lengths an average of the chain lengths may be taken provided all components are assumed to be at approximately the same concentration . For properties and activities where chemical structure may be approximated in this manner this approach is relatively successful. However all discussion in this chapter will assume that
