I have been using the pvclust package to perform hierarchical clustering with bootstrapped p-values. I am using it to compare the numbers of clusters identified between different datasets. However, the numbers are not comparable because pvclust calculates p-values for the optimal number of clusters relative to the distances in that dataset. Is there any way of setting a scale for pvclust so the package knows whether the distances it has are close together or far apart, in order to compare the different datasets? Any help you could give me would be greatly appreciated, Hannah [[alternative HTML version deleted]]