The conditions and methods for constructing reliable QSARs are revised in relation to each component of a QSAR study: the selection of a training set out of a QSAR compatible series, the collection of data pertinent to the descriptors matrix (X) and to the effects matrix (Y), the analysis of data to connect X to Y by a regression model, and the validation of the model. In discussing these conditions, attention is given to the constraints that arise from the theoretical foundation of QSARs as analogy models of local validity and to the complexity and limited knowledge about the mechanisms of action. Hence, emphasis is placed on the need and importance to adopt multivariate methods for dealing with (1) the characterization of the structures, (2) the selection of a representative set of training compounds, and (3) analysis of the data. It is finally shown that the same integrated multivariate approach applies to the modeling of biological activities and other properties—chemical and biological—as well as to the modeling of correlations between batteries of data. The role of QSAR in risk assessment is addressed in the second part of the article. The framework of a strategy for an efficient screening assessment of toxic substances through the modeling of their exposure and toxicity-related properties is outlined. Applications of the strategy are reported that deal with two series of compounds. Examples of toxicity and persistency models are illustrated.
References
1.
1. Wold, S., and Dunn, W.J. (1983). Multivariate quantitative structure-activity relationships. Conditions for then-applicability. J. Chem. Inf. Comput. Sci.23, 6–15.
2.
2. Wold, S., Albano, C., Dunn, W.J., et al. (1984). Multivariate data analysis in chemistry. In: Kowalski, B.R. (ed.). Chemometrics and Statistics in Chemistry. NATO ASI Series C, No. 138. Dordrecht, Holland: D. Reidel Publ. Co., pp. 17–95.
3.
3. Tosato, M. L., Marchini, S., Passerini, L., et al. (1990). QSARs based on statistical design and their use for identifying chemicals for biological testing. Environ. Toxicol. Chem.9, 265–277.
4.
4. Clementi, S., Cruciani, G., Cesareo, D., and Tosato M.L. (1989). Comparison of chemometric methods in toxicology. Chim. Oggi3, 57–61.
5.
5. Tosato, M. L., Cesareo, D., Galassi, S., et al. (1988). Multivariate design in toxicological evaluation of chemical data. Chim. Oggi3, 41–45.
6.
6. Tosato, M.L. Vigano, L. Skagerberg, B. and Clementi, S. (1991). A new strategy for ranking chemical hazards. Framework and application. Environ. Sci. Technol. in press.
7.
7. Jonsson, J., Eriksson, L., SjöSTRÖM, M., Wold, S., and Tosato, M.L. (1989). A strategy for ranking environmentally occurring chemicals. Chemometrics Intel. Lab. System5, 169–186.
8.
8. Wold, S., Berntsson, P., Eriksson, L., et al. (1991) Informationally sound QSARs. In: Silipo, C. and Vittoria, A. (eds.). Proceedings of the 8th European Symposium on Quantitative Structure-Activity Relationships. Amsterdam: Elsevier, in press.
9.
9. McKIM, J. M., Bradbury, S.P., and Niemi G.J. (1987). Fish acute toxicity syndromes and their use in the QSAR approach to hazard assessment, Environ. Health Perspect.71, 171–186.
10.
10. Box, G.E.P., and Draper, N.R. (1987). Empirical Model Building and Response Surfaces. New York: John Wiley & Sons.
11.
11. Box, G.E.P., Hunter, W.J., and Hunter, J.S. (1978). Statistics for Experimenters. New York: John Wiley & Sons.
12.
12. Wold, S., SjöSTRÖM, M., Carlson, R., et al. (1986). Multivariate design. Anal. Chim. Acta191, 17–32.
13.
13. Tosato, M.L. and Geladi, P. (1990). Design: a way to optimise testing programmes in QSAR screening of toxic substances. In: Karcher, W., and Devillers, J. (eds.). Practical Application of Structure-Activity Relationships (QSAR) in Environmental Chemistry and Toxicology. Kluwer Acad. Publ., pp. 326–551.
14.
14. Wold, S., Esbensen, K., and Geladi, P. (1987). Principal component analysis. Chemometrics Intel. Lab. Systems2, 37–52.
15.
15. Skagerberg, B., Bonelli, D., Clementi, S., Cruciani, G., and Ebert, C. (1989). Principal properties of aromatic substituents. A multivariate approach to design in QSARs. Quant. Struct. Act. Relations8, 32–38.
16.
16. Geladi, P., and Kovalsky, Br. (1986). Partial least squares regression: A tutorial. Anal. Chem. Acta185, 1–7.
17.
17. HÖSKULDSSON, A. (1988). PLS regression methods. J. Chemometrics2, 211–228.
18.
18. Wold, S. (1978). Cross-validatory estimation of the number of components in factor and principal components models, Technometrics20, 397–405.
19.
19. Cruciani, G., Clementi, S., Baroni, M., Ebert, C., and Skagerberg, B. (1990). Comparison of chemometric models for QSARs. Quant. Struct. Act. Relations9, 101–107.
20.
20. Tosato, M.L. and Passerini, L. (1990). Group Structuring of ANNEX I Sustances (EEC Directive 67/54B). Final Report, EC Contract N. EG 8905426M.
21.
21. Tosato M.L., and Silano V. (1986) Systematic Investigation of Existing Chemicals. Outline of a Proposal for International Co-operation. Personal Communication to the OECD Chemicals Division, Environment Directorate, Paris.
22.
22. Tosato, M. L., Cesareo, D., Clementi, S., et al. (1987). A strategy for systematic analysis and priority ranking of chemicals. Report of a preliminary feasibility study. Background paper at the OECD Workshop: Co-operation on Existing Chemicals: Charting the course. Ottawa, November 1987, p. 107.
23.
23. Veith, G. D., and Broderius S.J. (1987). Structure-toxicity relationships for industrial chemicals causing type (II) narcosis syndromes. In: Kaiser, K.I.E. (ed.). QSARs in Environmental Toxicology, II. Dondrecht, Holland: Reidel Publishing Company, pp. 385–391.
24.
24. Tosato, M.L. and Marchini, S. (1990). QSAR Models for Estimating Missing Toxicological Data on Existing Chemicals. Part II. Final Report. EEC Contract N. B6614–88–78.
25.
25. Tosato, M. L., Marchini, S., Passerini, L., Pino, A., and Skagerberg, B. (1991). QSARs in aquatic toxicology. Performance of different ensembles of multivariate structural descriptors. In: Silipo, C. and Vittoria, A. (eds.). Proceedings of the 8th European Symposium on Quantitative Structure-Activity Relationships. Amsterdam: Elsevier, in press.
26.
26. Eriksson, L., Jonsson, J., Lindgren, F., et al. (1990). A strategy for ranking environmentally occurring chemicals. Part III. Multivariate QSARs for halogenated aliphatics. Environ. Toxicol. Chem.9, 1341–1353.
27.
27. Eriksson, L. Jonsson J. Hellberg, S. et al. (1991) A strategy for ranking environmentally occurring chemicals. Part V. The development of two genotoxicity QSARs for halogenated aliphatics. Environ. Toxicol. Chem. submitted.
28.
28. Eriksson, L. Jonsson, J. Hellberg, S. et al. (1991). A strategy for ranking environmentally occurring chemicals. Part V. A QSAR for a bacterial mutation of halogenated haliphatics. Scand. Chem. Acta
29.
29. Atkinson, R. (1986). Kinetics and mechanisms of the gas-phase reaction of the hydroxyl radical with organic compounds under atmospheric conditions. Chem. Rev.86, 69–201.
30.
30. Chiorboli, C., Bignozzi C., Maldotti, A., Giardini, P.F., Rossi, A., and Carassiti V. (1983). Rate constants for the gas-phase reaction of OH radical with beta-methylstyrene and acetone. Int. J. Chem. Kinetics15, 579–586.
31.
31. Tosato, M. L., and Clementi, S. (1988). Assessment of the photodegradation of chemicals in the atmosphere. Test methods and predictive approaches. In: Schiarello, M. (ed.). Photocatalysis and Environment. Trends and Applications. NATO ASi, vol. 237, Kluwer Acad. Publ., pp. 583–597.
32.
32. Tosato, M.L. Chiorboli, C. Eriksson, L. and Jonsson, J. (1991). Multivariate modelling of the rate constant of the gas-phase reaction of haloalkanes with the hydroxyl radical. Sci. Total Environ. in press.
