Table 1.

Overview of sampled microbial lineages at various levels of OTU definitions

The table provides numerical details on the raw data and the results, and also illustrates the effects of changing the phylogenetic resolution at which the analysis is performed. For very narrowly defined OTUs, many lineages have to be discarded because they do not occur in a sufficiently large number of samples. Conversely, for very broadly defined OTUs, the statistical false discovery rate becomes too high, since many of the more abundant OTUs are seen to co-occur even after conservative randomization of sampling sites. NC, Not computed.