The Earth Microbiome project: successes and aspirations

The Earth Microbiome Project (EMP) was launched in August 2010, with the ambitious aim of constructing a global catalogue of the uncultured microbial diversity of this planet. The primary vision of the Earth Microbiome Project, to process the microbial diversity and functional potential from approximately 200,000 environmental samples, marks it as an undertaking so massive that it was at first considered to be pure folly (as late as 2012, Jonathan Eisen was quoted in Nature as saying ‘Knight and Gilbert literally talk about sampling the entire planet. It is ludicrous and not feasible - yet they are doing it’ [1]).

The initial concept arose out of a Department of the Environment (DOE) sponsored workshop on the promise of terabase-scale sequencing in Snowbird, Utah, designed to inspire research ideas using new technology to revolutionize microbial ecology and our understanding of the microbial world [2]. Many other exciting projects also evolved from that meeting, including efforts to extend the sequencing of type strains of cultured bacterial taxa, which in itself has become the Microbial Earth Project [3]. In October 2010, EMP pioneers held a small workshop at Argonne National Laboratories to determine the most effective way to jumpstart such an initiative. At this meeting, we agreed that the only feasible route to acquire and process 200,000 samples was through crowdsourcing, soliciting donations of samples from researchers around the world. This was identified as a key flaw in the design, on the grounds that it would not be possible to convince researchers to part with samples that had been painstakingly collected for inclusion in a single effort [4]. Fortunately, the participants’ generosity has greatly exceeded what we could have hoped for, and the crowdsourcing approach has been a success.

We floated this strategy initially as a potentially viable approach based on the precedent of existing programs that followed broadly similar designs, especially the International Census of Marine Microbes [5] and the Human Microbiome Project [6]. The basic design was founded on the principle of coordinated sample collection, and standardization of contextual metadata acquisition, DNA extraction, PCR and amplicon and shotgun sequencing approaches, and an open-source analytical platform with free, unrestricted access to both the amplicon and metadata immediately following completion of the analysis. Initially the effort was funded primarily by unrestricted funds available to the principle investigators through Argonne National Laboratory, Lawrence Berkeley National Laboratory, the Howard Hughes Medical Institute, and donations from corporate sponsors. Under this effort, the Earth Microbiome Project committee developed the standard protocols [7], contacted and collaborated with researchers from numerous different microbial ecology disciplines, from human, animal, plant, terrestrial, marine, freshwater, sediment, air, built environment and every intersection of these ecosystems. By August 2012, less than 2 years since its initiation, the Earth Microbiome Project had processed approximately 7,000 environmental samples, generating 16S rRNA amplicon data and releasing these data using an open portal through the Quantitative Insights into Microbial Ecology (QIIME) database. In June 2013, the EMP received awards from the WM Keck Foundation and the John Templeton Foundation to support activities to bring the catalogue up to 50,000 samples processed, and as of July 2014 we have reached over 30,000 (compared with the phase 1 Human Microbiome Project amplicon analysis of 5,771 samples [8]). In its planning phase, the EMP proposed the co-analysis of samples using metagenomics and metabolic modeling of ecosystems, and these aims are still viable, but such efforts have to date been more targeted to specific environments and studies. As it stands, the EMP represents the largest effort to characterize the diversity, distribution, and structure of microbial ecosystems across the earth, achievable only through coordinated collaboration of all of the independent research projects (166) that comprise the EMP. Although each hypothesis-driven study provided by our collaborators can tell its own story, the real power of the EMP is through meta-analysis of these data, empowering researchers to develop and use samples acquired from myriad ecosystems to test hypotheses in microbial ecology. Importantly, this pooled data resource also provides an unparalleled opportunity to contextualize individual studies by defining the patterns they see in a global context. These large-scale meta-analyses can enable researchers to ask unique questions regarding the biogeography, dynamic dispersal, and ecology of the microbial planet.