Rational design of test series

The amount of information obtainable from any QSAR analysis, and hence ultimately its predictive power, largely depends on the initial design of the underlying experimental study. Only if the compounds included in the test data set are true representatives of the chemical class(es) concerned, can the maximum information be extracted from the data. The members of the training set, which are used to derive the model, should be as diverse as possible with respect to their structural features. In statistical terms this means that they should reveal minimum intercorrelation and maximum variance in the properties regarded relevant for the effects studied. 

The selection of the training set becomes more sophisticated if more than three descriptors are assumed to be relevant. A manual selection is not feasible with a parameter space that is more than three-dimensional and computational techniques have to be applied. These mostly use multivariate statistics to account for underlying multiple intercorrelations among variables, such as PCA and PLS (Hellberg, 1986 Tosato et a/., 1991 Lindgren et a/., 1995), to find an acceptable compromise with respect to collinearity, variance and syn-thetic/commercial accessibility of the chemicals. 

There is no excuse for not making an effort to choose the appropriate compounds to form the basis of the intended QSAR. Even if little information is available on the relevant descriptors, there is at least always the possibility of using a 7r/cr-substituent constants diagram of the candidate chemicals to assess the intercorrelations between lipophilic and electronic properties. 

---