2014 Env Microbiol, Figure 1 (Quantitative differences between clustering methods)

<p>Quantitative differences between clustering methods.</p> <p>A. Differences in total OTU counts when clustering a global data set of 887 870 bacterial 16S rRNA gene sequences according to different methods. Note that uparse filtered for chimeric sequences differently than the other methods, which led to different numbers of sequences being clustered at different cut-offs (see Table S2 and Fig. S7). Moreover, uclust and uparse did not cluster to > 99% similarity, with additional missing data points for uparse (see Appendix S1).</p> <p>B. Differences in OTU size distributions between methods when clustering to 97% nominal sequence similarity.</p> <p>C. Differential dominance of singleton (1 sequence) and large OTUs (> 100 sequences) at 97% similarity. Methods differed in the fraction of total sequences (upper panel) and of total OTUs (lower panel) that fell into different OTU size categories.