What Will Human Microbiome Metagenomics Tell Us (Redux)?

Since I'm attending a human microbiome meeting today, I'll repost this question about the utility of metagenomics:

NIH, in about six months, will release a huge sum of money to fund the study of the human 'microbiome': those microorganisms that live on or in us. One of the things that will be done with this money is metagenomics which is "the study of genomes recovered from environmental samples as opposed to from clonal cultures." (In this case, the samples would be fecal, vaginal, your mouth, etc.). In other words, we can sample the 'community' of genomes (or perhaps we should call it the community ecology of genomes). So I have a question: what do we learn from metagenomics of the human microbiome?

Let me explain what I mean by this. There are lots of useful things that come from metagenomics. First, organisms that we currently can't culture will be discovered. Second, having some knowledge of genomes can help us develop better methods for tracking in situ community dynamics and interactions. Currently, most studies amplify 16S RNA genes and use them as a 'barcode' to follow species abundances. We could develop better markers (i.e., ecologically relevant genes) and build better technologies to study the human microbiome. We could also follow community-level gene expression patterns (which would be really useful). The list of tools and methods stemming from this could go on and on.

What I'm asking, however, is what do we learn from metagenomics per se about the ecology of the human microbiome that we wouldn't be able to glean from genetic barcodes, or for that matter, latin binomials?

In other words, if we sequenced enough human microbiomes (e.g., fecal samples from 200 people), and enough genomes per sample to develop the tools and methods I described above (and many others I didn't), would there be any need to sequence additional genomes? What exactly could we learn from the ecology of genomes?

Discuss.

Tags

More like this

Well, barcodes can't tell you anything about the prevalence of specific genes or plasmids. They can't measure horizontal gene transfer or the amount of selective pressure on different alleles.

It seems like all of these things could potentially be informative when combined with health information about the person from whom the sample was collected.

"Metagenomics" is appealing because it doesn't require doing anything difficult, such as obtaining clonal cultures of many individual microorganisms. It has a very fancy name, but it comes down to only looking for your keys under the streelight, even though you have no reason to believe that you specifically lost them there.