Hi,
some of my bootstrap values are over 100 (up to 200) when clustering at 0.01. How is this possible? I should mention it are unknown sequences at the domain level.
Thanks for bringing this to our attention. The confidence score is calculated as follows ceil((numberOfSeqsWithConsensusTax / totalNumberOfSeqs) * 100). Mothur mistakenly added 1 to numberOfSeqsWithConsensusTax in the special case of “unknown”. This will be fixed in the next release.