Data structures and data transformations for clustering chemical data

The quality of a clustering of chemical data is determined by a proper choi ce of distance measures and data transformations. The latter aspect is ofte n neglected and its importance is shown here. It is also shown that the V-s haped data structure that is often obtained in a principal component analys is of chemical data may indicate that the clustering of the raw data can le ad to classifications that are not relevant from a chemical point of view a nd that the log double centering transform should be considered as a possib le alternative. (C) 2001 Published by Elsevier Science B.V.