Introducing DOTUR, a Computer Program for Defining Operational Taxonomic Units and Estimating Species Richness

Although copious qualitative information describes the members of the diverse microbial communities on Earth, statistical approaches for quantifying and comparing the numbers and compositions of lineages in communities are lacking. We present a method that addresses the challenge of assigning sequences to operational taxonomic units (OTUs) based on the genetic distances between sequences. We developed a computer program, DOTUR, which assigns sequences to OTUs by using either the furthest, average, or nearest neighbor algorithm for each distance level. DOTUR uses the frequency at which each OTU is observed to construct rarefaction and collector's curves for various measures of richness and diversity. We analyzed 16S rRNA gene libraries derived from Scottish and Amazonian soils and the Sargasso Sea with DOTUR, which assigned sequences to OTUs rapidly and reliably based on the genetic distances between sequences and identified previous inconsistencies and errors in assigning sequences to OTUs. An analysis of the two 16S rRNA gene libraries from soil demonstrated that they do not contain enough sequences to support a claim that they contain different numbers of bacterial lineages with statistical confidence (P > 0.05), nor do they contain enough sequences to provide a robust estimate of species richness when an OTU is defined as containing sequences that are no more than 3% different from each other. In contrast, the richness of OTUs at the 3% level in the Sargasso Sea collection began to plateau after the sampling of 690 sequences. We anticipate that an equivalent extent of sampling for soil would require sampling more than 10,000 sequences, almost 100 times the size of typical sequence collections obtained from soil.