In a nutshell

  1. Select a rank (e.g. Genus), set "Regress by" to "Taxa vs Envp" and press "Set Mode", followed by "Run Analysis". The displayed table shows the effect of the environmental variables on the abundance of taxa. For each taxon a regression model is fit of the form:
    abundance of taxon = env1 + env2 + env3 ...,
    where env1, env2, ... are the environmental variables. For each taxon and each environmental variable a p-value is shown, indicating if the variable impacts the abundance of the respective taxon. Select an environmental variable in the header of the displayed table to sort the p-values. If you have paired data, select the "Paired" checkbox and press "Run Analysis". In the case of paired data, a linear mixed effects model is fit of the form:
    abundance of taxon = env1 + env2 + env3 ... + pair1 + pair2 + ...,
    where the environmental variables are included as fixed effects and the pairs as random effect.
  2. Use the Taxa drop-down menu to select a specific taxon or OTU. Press "Run Analysis" to analyze the correlation between the abundance of the selected taxon/OTU and the environmental variables.
  3. Next, do a regression analysis to identify complex associations between environmental variables and the community diversity. First, set Taxa to "All" and "Regress by" to "Diversity vs Envp" and press "Set Mode". Next set the Index to either "Shannon" or "Richness" and press "Run Analysis". In the generated figure, first the p-values of the regression model are displayed. Additionally, for each environmental variable a two scatter-plots are shown. The p-values dispalyed in the scatter plot indicate if the Pearson correlation between the environmental variable and the diversity index is significant. Select the "Paired" checkbox for paired data.
  4. You can further do a regression analysis to identify environmental variables impacting the global community composition. Set "Regress by" to "Distance vs Envp" and press "Set Mode". Next, set Taxa to "All", "Distance Method" to "Jaccard" and press "Run Analysis".
  5. Calculate multiple linear regression model to explain the grouping of samples as specified in the meta data file. Set "Regress by" to Group and press "Set Mode". Select different groups for condition A and condition B. Set Taxa to "All". Press "Run Analysis".