hadilou
Hi, In the dendrogram function [link to scipy](https://docs.scipy.org/doc/scipy/reference/generated/scipy.cluster.hierarchy.linkage.html#scipy.cluster.hierarchy.linkage) the default metric function to calculate distance between pairs of points is `Euclidean`. Since we are dealing with nullity binary values, won't it be more convenient to use a *similarity metric* such as `Hamming` distance (i.e. proportion of disagrees), `Jaccard index`, etc? Thanks!