The tragedy of the uncommon: understanding limitations in the analysis of microbial diversity

Molecular microbial community analysis methods have revolutionized our understanding of the diversity and distribution of bacteria, archaea and microbial eukaryotes. The information obtained has adequately demonstrated that the analysis of microbial model systems can provide important insights into ecosystem function and stability. However, the terminology and metrics used in macroecology must be applied cautiously because the methods available to characterize microbial diversity are inherently limited in their ability to detect the many numerically minor constituents of microbial communities. In this review, we focus on the use of indices to quantify the diversity found in microbial communities, and on the methods used to generate the data from which those indices are calculated. Useful conclusions regarding diversity can only be deduced if the properties of the various methods used are well understood. The commonly used diversity metrics differ in the weight they give to organisms that differ in abundance, so understanding the properties of these metrics is essential. In this review, we illustrate important methodological and metric-dependent differences using simulated communities. We conclude that the assessment of richness in complex communities is futile without extensive sampling, and that some diversity indices can be estimated with reasonable accuracy through the analysis of clone libraries, but not from community fingerprint data.