Data generated from amplicon sequencing is inherently compositional and provides relative abundances, which are generally independent of the total microbial load of the original sample ( Fig. 4).