Computer assisted short-term test battery design: Some answers (reply to john ashby's commentary in environmental and molecular mutagenesis)
✍ Scribed by Ennever, Fanny K. ;Rosenkranz, Herbert S.
- Publisher
- Wiley (John Wiley & Sons)
- Year
- 1988
- Weight
- 299 KB
- Volume
- 12
- Category
- Article
- ISSN
- 0192-2521
No coin nor oath required. For personal study only.
✦ Synopsis
Once again, John Ashby has provided an insightful and thought-provoking perspective on the prospects for improvement in the "difficult period" in which he finds short-term tests [Ashby, 19881. Our primary disagreement with him is that we do not consider the "human judgment" for which he calls to be so very different from our "statistical" approach (the Carcinogenicity Prediction and Battery Selection method, or CPBSm).' In essence, Ashby feels that the interpretation of a chemical's response in a short-term test should be influenced by i) the structure of the chemical and ii) the known performance characteristics of the test. At present, our CPBS methodology takes only the second of these factors into account; however, nothing inherent in the theory of CPBS precludes the use of structural information. In fact, we are developing procedures which will eventually allow the integration of structural determinants into CPBS when we have achieved a better understanding of the way in which structure controls biological activity (see below). We do feel some concern that Ashby sees the statistical approach as competing with biologicakhemical investigations; we hope that it is complementary, and that simultaneous development will be mutually rewarding.
Undoubtedly, chemical structure ultimately determines biological activity or lack of activity, but evidently not in any simple way. In an area of computer-assisted research which is separate from but related to CPBS, we have developed the ability to identify structural determinants of activity for mutagens and carcinogens (computer-assisted structure evaluation, or CASE [Frierson et al., 19861). The CASE methodology can identify portions of molecules more complex than Ashby's "Alert from Structure" (his Table 2), and therefore in a class of chemicals superficially