The status of the microbial census: an update

Abstract

A census is typically carried out for people at a national level; however, microbial ecologists have implemented a molecular census of bacteria and archaea by sequencing their 16S rRNA genes. We assessed how well the microbial census of full-length 16S rRNA gene sequences is proceeding in the context of recent advances in high throughput sequencing technologies. Among the 1,411,234 and 53,546 full-length bacterial and archaeal sequences sequences, 94.5% and 95.1% of the bacterial and archeaeal sequences, respectively, belonged to operational taxonomic units (OTUs) that have been observed more than once. Although these metrics suggest that the census is approaching completion, 29.2% of the bacterial and 38.5% of the archaeal OTUs have been observed more than once. Thus, there is still considerable microbial diversity to be explored. Unfortunately, the rate of new full-length sequences has been declining and new sequences are primarily being deposited by a small number of studies. Furthermore, sequences from soil and aquatic environments, which are known to be rich in bacterial diversity, only represent 7.8 and 16.5% of the census while sequences associated with zoonotic environments represent 55.0% of the census. Continued use of traditional approaches and new technologies such as single cell genomics and short read assembly are likely to improve our ability to sample rare OTUs if it is possible to overcome this sampling bias. The success of ongoing efforts to use short read sequencing to characterize microbial communities requires that researchers strive to expand the depth and breadth of the microbial census.\n\nImportanceThe biodiversity contained within the bacterial and archaeal domains dwarfs that of the eukaryotes and the services these organisms provide to the biosphere are critical. Surprisingly, we have done a relatively poor job of keeping track of the ongoing effort to characterize the biodiversity as represented in full-length 16S rRNA genes. By understanding how this census is proceeding, it is possible to suggest the best allocation of resources for advancing the census. We found that the ongoing effort has done an excellent job of sampling the most abundant organisms, but struggles to sample the more rare organisms. Through the use of new sequencing technologies we should be able to obtain full-length sequences from these rare organisms. Furthermore, we suggest that by allocating more resources to sampling environments known to have the greatest biodiversity we will be able to make significant advances in our characterization of microbial diversity.