File S1 - The Bias Associated with Amplicon Sequencing Does Not Affect the Quantitative Assessment of Bacterial Community Dynamics

Supporting Information file. Table S1. Operational parameters of the wastewater treatment plant. Table S2. Numbers of reads obtained throughout the bioinformatic workflow. Table S3. Distribution of classified 16S rRNA sequences at the phylum level for 12 activated sludge samples (two technical replicates for each of the two variable regions). Figure S1. Heatmap of bacterial genera abundance in each time point, based on the analysis of the two 16S rRNA regions. The 25 most abundant genera are shown. Color scale refers to the square root of the abundance of each genus relative to all bacterial sequences. Figure S2. Estimates for bacterial richness are affected by the choice of primers. Rarefaction curves built for OTUs defined at 97% similarity. Each plot corresponds to a time point (month) for V1–V3 (▵) and V4 (○). Error bars show the 95% confidence interval derived from the 1000-time iterative sampling. Figure S3. Effect of the different primer sets on the estimation of alpha-diversity. (A) Observed richness and (B) Shannon index were calculated on the normalized samples as average between the two technical replicates. Gray bars represent the values obtained with primers F8-R534 (V1–V3) and white bars, with primers F563-R907 (V4). Error bars show SEM. Figure S4. Changes in the relative sequence abundance of bacterial core members. A) Time series of the relative abundances of core members for each 16S rRNA variable region data set. OTUs considered part of the core were detected in all twelve sampling points in at least one 16S rRNA region; all other OTUs were grouped as “not core”. Replicates within each 16S rRNA region were averaged for clarity of presentation. B) Moving-window analysis. Symbols indicate the mean of Bray-Curtis dissimilarities between consecutive sampling points within the V1–V3 (▵) and the V4 (○) core datasets. Error bars represent standard error. Figure S5. Comparison of beta diversity results based on V1–V3 and V4 16r RNA regions. Procrustes analysis of NMDS ordination plots based on V1–V3 and V4 datasets. Replicates within each 16S rRNA region were averaged for clarity of presentation. Blue lines connect paired samples on the target configuration indicated by circles (V1–V3 region), and the reference configuration (end of arrow, V4 region). Correlation r =  0.90, p < 0.001.
(PDF)