I am trying to use the libshuff command after running either:

read.dist(phylip=sample.phylip.dist, group=sample.groups, name=sample.names)

or simply

read.dist(phylip=sample.phylip.dist, group=sample.groups)

In each case, I get an error saying there are more sequences in the group file than the distance file. Of course this is the case because of the unique.seqs steps run prior to the creation of the distance matrix. But the .names file does not seem to get around this. So mothur creates a new file group file called editted.groups which has the same number of sequences as the distance file.

Am I missing something or do you loose information in downstream applications using this editted.groups file?


sorry - libshuff doesn’t take the name file option. i’m not sure we’re likely to add it any time soon as libshuff is a very weird tool (see my ISMEJ paper). i suspect you will get significant p values if you have enough sequences to need to use the name file.