Computational Systems Pharmacology and Toxicology
CHAPTER 9: In silico Toxicology: An Overview of Toxicity Databases, Prediction Methodologies, and Expert Review
Published:01 Mar 2017
Special Collection: 2017 ebook collectionSeries: Issues in Toxicology
D. Bower, K. P. Cross, S. Escher, G. J. Myatt, and D. P. Quigley, in Computational Systems Pharmacology and Toxicology, ed. R. J. Richardson and D. E. Johnson, The Royal Society of Chemistry, 2017, pp. 209-242.
Download citation file:
Understanding chemical toxicity is a necessary part of the R&D and regulatory approval process across many industries (e.g. pharmaceuticals, cosmetics, and pesticides). Toxicologists have an increasingly rich set of in vivo and in vitro methods to assess hazard and risk, which are being progressively supplemented with newer in silico approaches. The advantages and disadvantages of in silico methods are described alongside in vivo and in vitro tests. This chapter reviews a series of in silico methodologies for predicting toxicity and underpinning all in silico methodologies is the necessity to access high-quality and up-to-date toxicity study data from a variety of sources. Methods for organizing toxicity data in a harmonized manner (such as ToxML) are discussed to support combining toxicology data from different sources along with a number of commonly used toxicology databases. The three most commonly used methodologies for predicting toxicity—expert alerts, QSAR models and read-across—are reviewed. These complementary approaches provide different viewpoints concerning the structural and mechanistic basis for any prediction, alongside an analysis and rationale for supporting analog data. How this information can be then assimilated within an expert review to generate a final conclusion is discussed.