MG-RAST, a Metagenomics Service for Analysis of Microbial Community Structure and Function

Abstract

Approaches in molecular biology, particularly those that deal with high-throughput sequencing of entire microbial communities (the field of metagenomics), are rapidly advancing our understanding of the composition and functional content of microbial communities involved in climate change, environmental pollution, human health, biotechnology, etc. Metagenomics provides researchers with the most complete picture of the taxonomic (i.e., what organisms are there) and functional (i.e., what are those organisms doing) composition of natively sampled microbial communities, making it possible to perform investigations that include organisms that were previously intractable to laboratory-controlled culturing; currently, these constitute the vast majority of all microbes on the planet. All organisms contained in environmental samples are sequenced in a culture-independent manner, most often with 16S ribosomal amplicon methods to investigate the taxonomic or whole-genome shotgun-based methods to investigate the functional content of sampled communities. Metagenomics allows researchers to characterize the community composition and functional content of microbial communities, but it cannot show which functional processes are active; however, near parallel developments in transcriptomics promise a dramatic increase in our knowledge in this area as well.

Since 2008, MG-RAST (Meyer et al., BMC Bioinformatics 9:386, 2008) has served as a public resource for annotation and analysis of metagenomic sequence data, providing a repository that currently houses more than 150,000 data sets (containing 60+ tera-base-pairs) with more than 23,000 publically available. MG-RAST, or the metagenomics RAST (rapid annotation using subsystems technology) server makes it possible for users to upload raw metagenomic sequence data in (preferably) fastq or fasta format. Assessments of sequence quality, annotation with respect to multiple reference databases, are performed automatically with minimal input from the user (see Subheading 4 at the end of this chapter for more details). Post-annotation analysis and visualization are also possible, directly through the web interface, or with tools like matR (metagenomic analysis tools for R, covered later in this chapter) that utilize the MG-RAST API (http://api.metagenomics.anl.gov/api.html) to easily download data from any stage in the MG-RAST processing pipeline. Over the years, MG-RAST has undergone substantial revisions to keep pace with the dramatic growth in the number, size, and types of sequence data that accompany constantly evolving developments in metagenomics and related -omic sciences (e.g., metatranscriptomics).