Secondly, in the naïve diversity profile for the putative methyltransferase group, the lines representing the diversity of the 2007A, 2009B, and 2010B samples crossed each

other numerous times between q = 0 and q = 5 (Additional file 1: Figure S2). Lastly, in the naïve profile for the putative concanavalin A-like glucanases/lectins group, the 2010B samples were as diverse as or more diverse than the 2007A samples at q = 0, but the diversity of 2010B samples dropped sharply and remained lower than all other samples after approximately q = 0.5 (Additional file 1: Figure S3). In the case of viral diversity, ultra-rare taxa play an important role in rapid evolution to allow new viruses to infect hosts that are constantly evolving defense Tanespimycin mechanisms. Thus, diversity calculated at low values of q, which are sensitive to rare taxa, is the more appropriate measure of viral diversity. Figure 1 Hypersaline lake viruses Cluster 667 diversity profiles. (A) Naïve and (B) similarity-based (phylogenetic relatedness) diversity profiles calculated for Cluster 667 from the hypersaline lake viruses data. We see similar STI571 nmr results

for the acid mine drainage dataset. At q = 0 (species richness) in the naïve analysis, the Env-3 at growth stage 2 check details sample is the most diverse sample, but the sample’s diversity decreases and is surpassed by the growth stage 0 bioreactor sample and both Env-1 samples between Morin Hydrate q = 1 and q = 2 (Figure 2), demonstrating that the bioreactor and Env-1 samples were less even than the Env-3 sample at growth stage 2. Thus, for this dataset as well as for the hypersaline lake viruses dataset, evaluating the diversity of the microbial communities at multiple values of q leads to a different interpretation of the results and response to the original hypotheses (Table 1). Figure 2 Acid mine drainage bacteria and archaea (HiSeq) diversity profiles. (A) Naïve and (B) similarity-based (phylogenetic relatedness) diversity profiles calculated from the acid mine drainage bacteria and archaea HiSeq data. Diversity profiles do not always add new information

to analyses of natural microbial datasets. In some cases, such as with the naïve profiles of the subsurface bacteria dataset, the most diverse samples in a dataset were always calculated as the most diverse, across the entire range of q in the naïve profile (Figure 3). Thus, whether we quantified diversity using species richness, Shannon diversity, or diversity profiles, we would arrive at the same result. In general, our findings provide evidence for the utility of diversity profiles to analyze microbial datasets, even when similarity information is not taken into account, because they allow researchers to visualize multiple diversity indices across the range of q in the same place after just one calculation.