I will try R, but my R skills aren't great yet. I am obsessed with using this plugin as it is really good for this kind of data. Ttest it is =)
Do I need to split it into two levels? I have four levels, with 1 level being a control? Or will it do pairwise associations between each level? Your R documentation seems to be a 2 level analysis.
When doing the t-test, it will have to be pairwise comparisons. The glm module is used for more than two conditions, and we are working to get this included in the qiime plugin as well! For now it is only available in the R package.
And thanks! I do have to mention that Greg Gloor is the brains behind this package!