Abstract
The composition of microbial communities is commonly determined by sequence analyses of one of the variable (V) regions in the bacterial 16S rRNA gene. We aimed to assess whether sequencing the full-length versus the V4 region of the 16S rRNA gene affected the results and interpretation of an experiment. To test this, mice were fed a diet without and with the prebiotic inulin and from cecum samples, two primary data sets were generated: (1) a 16S rRNA full-length data set generated by the PacBio platform; (2) a 16S rRNA V4 region data set generated by the Illumina MiSeq platform. A third derived data set was generated by in silico extracting the 16S rRNA V4 region data from the 16S rRNA full-length PacBio data set. Analyses of the primary and derived 16S rRNA V4 region data indicated similar bacterial abundances, and α- and β-diversity. However, comparison of the 16S rRNA full-length data with the primary and derived 16S rRNA V4 region data revealed differences in relative bacterial abundances, and α- and β-diversity. We conclude that the sequence length of 16S rRNA gene and not the sequence analysis platform affected the results and may lead to different interpretations of the effect of an intervention that affects the microbiota.