For my PhD project i am trying to use copy number variations (CNVs) to deduce phylogeny between 9 cancer cell lines... and it seem to work quite well. It is not hard to get to a 9x9 distance matrix showing the copy number variations between each of them (400-2000 CNVs out of 20000 genes in the genome). But do any of you have experience with how to draw a consensus tree with bootstrapping from this (and not just a tree based on the one distance matrix)? I have looked into the R-packages ape, phangorn and phyclust but all of them require you to start with sequences that you then align. They are not very happy about accepting a distance matrix alone (because bootstrapping require taking sub-samples from your dataset).
It is not hard to make e.g. 1000 new distance matrices based on extraction of e.g. 1/4 the dataset by random
So I believe the question is: Would it be possible to use 1000 distance matrices to create one concensus phylogenetic tree? R would be preferred but any input is appreciated.