Hypothesis testing approaches

Hypothesis testing approaches allow you to determine whether there is sufficient evidence to suggest that the structuring of your data are more extreme than you would expect by chance. They cannot indicate a level of similarity - for those types of questions, you are best served by OTU-based approaches. Examples of each command are provided within their specific pages, but several users have provided several analysis examples, which use these commands. An exhaustive list of the commands found in mothur is available within the commands category index.

General commands

• dist.seqs - generate a distance matrix
• deunique.tree - reinserts redundant sequence identifiers into a unique tree
• clearcut

Analyses

• parsimony - the parsimony test, which was previously available through treeclimber
• unifrac.unweighted - the unweighted Unifrac algorithm
• unifrac.weighted - the weighted Unifrac algorithm
• libshuff - the integral and discrete forms of the libshuff method
• amova
• homova
• mantel
• anosim
• cooccurrence