Summary
For structure-activity correlation, Partial Least Squares (PLS) has many advantages over regression, including the ability to robustly handle more descriptor variables than compounds, nonorthogonal descriptors and multiple biological results, while providing more predictive accuracy and a much lower risk of chance correlation. The major limitations are a higher risk of overlooking ‘real’ correlations and sensitivity to the relative scaling of the descriptor variables.
Similar content being viewed by others
References
Joreskog, K. and Wold, H. (Eds.) Systems Under Indirect Observation: Causality, Structure, Prediction, North-Holland, Amsterdam, 1982. PLS was developed in response to a request by the Swedish government for an objective method of handicapping state-run horse races (S. Wold, private communication).
Wold, S., Martens, H. and Wold, H., In Kagstrom, B. and Ruhe, A. (Eds.) Lecture Notes in Mathematics, Matrix Pencils, Vol. 973, Springer, Berlin, 1983, p. 286.
Wold, S., Ruhe, A., Wold, H. and Dunn III, W.J., SIAM J. Sci. Stat. Comput., 5 (1984) 735.
Cramer III, R.D., Patterson, D.E. and Bunce, J.D., J. Am. Chem. Soc., 110 (1988) 5959.
Lorber, A., Wangen, L.E. and Kowalski, B.R., J. Chemometrics, 1 (1987) 19.
Hoskuldsson, A., J. Chemometrics, 2 (1988) 231.
Wold, H., In Krishnaiah, P. (Ed.) Multivariate Analysis, Academic Press, New York, NY, 1966, p. 391. For one of several more efficient methods, see Lindgern, F., Geladi, P. and Wold, S., J. Chemometrics, 7 (1993) 45.
Cramer III, R.D., J. Am. Chem. Soc., 102 (1980) 1837.
Cramer III, R.D., Bunce, J.D., Patterson, D.E. and Frank, I.E., Quant. Struct.-Act. Relatsh., 7 (1988) 18.
Baroni, M., Costantino, G., Cruciani, G., Riganelli, D., Valigi, R. and Clementi, S., Quant. Struct.-Act. Relatsh., 12 (1993) 9.
Topliss, J.G. and Edwards, R.P., J. Med. Chem., 22 (1979) 1238.
Topliss, J.G. and Costello, R.J., J. Med. Chem., 15 (1971) 1066.
Clark, M. and Cramer III, R.D., Quant. Struct.-Act. Relatsh., 12 (1993) 137.
Wakeling, I.N. and Morris, J.J., J. Chemometrics, 7 (1993) 291.
Cramer III, R.D., DePriest, S.A., Patterson, D.E. and Hecht, P., In Kubinyi, H. (Ed.) 3D QSAR in Drug Design: Theory, Methods and Applications, ESCOM, Leiden, 1993, pp. 443–485.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Cramer, R.D. Partial Least Squares (PLS): Its strengths and limitations. Perspectives in Drug Discovery and Design 1, 269–278 (1993). https://doi.org/10.1007/BF02174528
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF02174528