Analysis of the salivary microbiome using culture-independent techniques

Table 3 Similarity between the V3 16S and other datasets

Datasets to which the F and R 30-cycle V3 subsets were compared	Average similarity^a ± SD (%)	Pearson's correlation coefficient (F, R)^b
Itself (F vs R)	94.8	0.998
25-cycle V3	97.1 ± 0.3	0.998, 0.999
20-cycle V3	93.2 ± 1.5	0.974, 0.977
30-cycle V1	82.7 ± 0.8	0.987, 0.990
16S-WGS	68.5 ± 3.7	0.933, 0.958
WGS	64.1 ± 0.5	0.650, 0.652
V3 from other individuals^c	64.8 ± 3.5	0.453-0.871, 0.458-0.867

F, forward run; R, reverse run.
^a The genera frequency were square-root transformed and used to construct the Bray-Curtis similarity matrix. When applicable, the similarities were calculated separately for the pairs of forward and the pairs of reversed subsets prior to averaging.
^b Based on untransformed data.
^c Samples 1.1, 2.1, 3.1, 4.1 and 5.1 from a previous study [21] in which the initial PCR cycles were carried out in the same conditions as those in the present work. Then, 28 cycles were performed using bar-coded primers and the amplicons were sequenced by the Roche/454 platform. The average similarity observed among these five microbiomes was 65.2 ± 7.2%.