Total sequence count was calculated for each sample. Using raw counts, a negative-binomial generalized linear model with and without a zero-inflation component was fitted for each genus with the glmmTMB R package v 1.0.2.1 using log(total sequence count) as an offset variable, and PD vs control as the independent variable. Results were extracted from the model with the lowest Akaike information criterion. P values were calculated using the base summary function in R and corrected for multiple testing using BH FDR method implemented in the p.adjust function from stats package.
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.