course of a human lifetime. For example, one key universal property is that unlike the human genome, the human microbiome is acquired anew each generation, with vaginally born babies acquiring different microbiomes than cesarean section (C-section) babies (Dominguez-Bello et al., 2010). Meanwhile, Proctor questioned whether certain other phenomena—namely, enterotypes, the notion of a “core” microbiome, and the idea that the presence of a pathogen indicates disease—are universal properties. None of these, in her opinion, are universal properties based on the evidence to date (e.g., Wu et al., 2011a).
Following Proctor’s presentation, Jennifer Russo Wortman, director of microbial informatics at the Broad Institute, described methodologies that HMP Consortium investigators are using to analyze the massive amount of genomic data that are accumulating. Most researchers are using one of two types of data: (1) 16S ribosomal ribonucleic acid (rRNA) data to determine what microbes are present (i.e., by using operational taxonomic units, or OTUs, as proxies for species) and (2) whole-genome shotgun reads to get a sense of what these microbes might be doing (i.e., by comparing sequences to functional databases). The data reveal varying levels of microbial diversity, depending on taxonomic level, among body sites (e.g., vaginal samples have less genus-level microbial diversity than other body sites, but more species-level diversity). Among individual hosts, scientists are observing greater compositional diversity (based on 16S rRNA reads) than putative functional diversity (based on shotgun reads). One of the greatest challenges in moving forward will be interpreting the massive amount of sequencing data that are accumulating, especially with respect to function, by integrating them with transcriptomic, proteomic, and metabolomic data into a systems-level approach to studying the microbiome.
Jeremy Nicholson, head of the Department of Surgery and Cancer at the Imperial College London, argued that not only is an integrative, systems-level approach necessary for understanding human health and disease, but studying the microbiome is central to that approach (Mirnezami et al., 2012; Nicholson, 2006). Only by understanding how gut microbes are signaling and otherwise functioning, especially with respect to their impact on their human host, will scientists ever be able to tease apart human biocomplexity enough to realize the vision of personalized health care. Nicholson discussed some of the ways that gut microbes influence human host metabolism and generate differential metabolic phenotypes (Holmes et al., 2008). For example, mouse and rat studies have demonstrated what Nicholson described as a “massive effect of the microbiome on bile acid metabolism,” with gut microbial activity impacting liver and colonic disease risk as a result (Martin et al., 2007; Swann et al., 2011).