STC Poll #1 – The Definition of the "Microbiome"
The term "Microbiome" is being used in two ways; either to denote all microbes at a site or all their genomes.
Both can't be right, so let's have a poll!
(as of 2023, the poll is defunct, but we posted the results here)

Great idea, Elio...
But then I wonder: what does "microbiota" mean?
My spin is that "microbiota" refers to the organisms, and "microbiome" to their many genomes. Jon Eisen is allergic to the "omics" trend in many ways, but I think it has become part of our scientific lexicon.
Interesting to see what happens.
I question that we have to make this discrimination. The majority of the novel "species" has been detected by metagenomic studies and few of them actually cultivated. Thus they are known to us by their genomes — and their physiological properties deduced via e.g. bioinformatic pathway reconstructions. Admittedly, this is no longer microbiology sticking to Koch's postulates — who feels the vocation to provide an update?
All the species at a site: microbiota
All the genomes at a site: metagenome
Microbiome isn't an omics word at all, it's an ecology word — small biome (which would probably mean that @phylogenomics would have no problem with it at all, not bad omics). Neither of the above two definitions refer to a biome, which requires some mention of interaction of the biotic with the environment.
I lean towards parsing it as "micro-biome" (synonymous with "microbiota", I suppose) rather than "microbi-ome". We use metagenomic and other approaches to characterize the microbiome.
I agree with Mike Cox, who wrote:
"I question that we have to make this discrimination. The majority of the novel "species" has been detected by metagenomic studies and few of them actually cultivated. Thus they are known to us by their genomes — and their physiological properties deduced via e.g. bioinformatic pathway reconstructions."
If the originally-defined microbiome can be easily predicted from the 16S taxa, as stunningly evident in PICURSt, then I fail to see the need to distinguish between microbiota and microbiome. The latter is so universally used and recognized by the public, it is the place to start and end this debate.
Go with microbiome to capture any metagenomic study of microbial communities.
I voted "Something Else" because I would prefer for "microbiome" to mean an environment or habitat in which certain microorganisms live. We have other words, such as microbiota and metagenome, to refer to the organisms and genomes. But I realize that it's probably too late, and we'll be stuck with the current confusion for the foreseeable future.
I don't see the difference between "the totality of genomes at a site" and "the totality of microbial species (whatever that means) at a site." Aren't the species at a given site determined by the presence of their genomes? I also agree that the microbiota are the inhabitants of a micro-biome, with microbiome being an ecological term, especially in the context of a larger organism, ie the microbiome of the human gut implies the ecology of the gut and the host-microbe and microbe-microbe interactions in the gut.
Yeah, logically "microbiome" would refer to a small biome, or a biome in which microbes live, but I don't think that is actually how people parse the word — they treat it as if it was really "microbome" — the "ome" consisting of microbes in the same sense of proteome and the like.
I'm not sure it's a good idea to use "metagenome" to refer to this because the microbiome isn't the only case where metagenomics is used. Although interestingly enough I've seen the *opposite* case — a manuscript where the "microbiome of the Baltic Sea" was referred to.
There are already established ecological terms for these things, so molecular biologists do not need to make up new ones. Biome clearly refers to the ecosystem, not just the organisms. So I voted for Other. The totality of the species at a site should be called the Community. The total of the community's genomes would be the metagenome. As with many terms, context is important, especially defining the scale at which you are measuring. An ecosystem can be tiny or huge.
I'm still working on "gene" and "epigenetics."
–bks
I think some people here are making a lot confusion with terms.
"Microbiota" means simply "microorganisms". If it is applied to a specific habitat/space/whatever then we can refer to the totality of microorganims in that place as the "microbiome" of that place. The "place" can be whatever scale you want to refer to, from a small pond to an entire planet. Thus, "microbiome" refers to the totality of microbial taxa of a certain location. So yes, you can perfectly refer to the microbiome of the Baltic Sea.
"Metagenome" and "Metagenomics" are two different things, although they are related. A "Metagenome" is the totality of genomes/genes at a particular place. "Metagenomics" is a collection of techniques to obtain information about metagenomes.
So, to answer Elio's question, whoever is using the term "Microbiome" to denote all microbial genomes at a particular site is doing it wrong. In fact, I work in Metagenomics and I never came across such a definition.
I should add a further comment that I forgot: Vicky McKinley is right when he says that the term "Microbiome" also implies the relationships between the organisms, since the word derives from the ecological term "Biome". Sometimes it is used only to denote microbial diversity and structure of a certain community at a particular site, but strictly speaking it includes both the organismal relationships and abiotic environment on which the community is found. This is the subtle difference between "microbiota" and "microbiome".
Here are my definitions:
Microbiota: The assemblage of microorganisms present in a defined environment. i.e. a census of microorganisms (bacteria, archaea or lower eukaryotes...).
Metagenome: The collection of genomes and genes from the members of a microbiota.
Microbiome: This term refers to the entire habitat, including the microorganisms, their genomes (i.e., genes) and the surrounding environmental conditions. This definition is based on that of "biome", the biotic and abiotic factors of a given environments.
Also, let's stop referring 16S rRNA gene surveys as metagenomic studies!
See my post at: http://www.allthingsgenomics.com/blog/2013/1/11/the-vocabulary-used-to-describe-microbial-communities-microbiome-metagenome-microbiota
A note from Elio
I will soon write up the results of the poll. Please bear with me.
I agree with Jacques Ravel, although I have a minor disagreement, so to speek:
"""Microbiome: This term refers to the entire habitat, including the microorganisms, their genomes (i.e., genes) and the surrounding environmental conditions. This definition is based on that of "biome", the biotic and abiotic factors of a given environments."""
I think including genomes/genes in the definition of "microbiome" is probably not warranted. As Jacques says, "microbiome" is a special case of the term "biome" used in Ecology. However, traditionaly that term doesn't included any reference to genetic material present at a "biome". In my opinion, neither should it have such a reference when refering to "microbiomes". I understand that much in microbiology by necessity will have a greater focus on the metabolic/biochemical potential of microorganims, but genetics should be left out of the definition of microbiome. When one refers to the, say, African Savanna biome, one does not include in the definition the genomes of lions or elephants, and neither should we do so when refering to microbiomes, even if in practice we need to study that genetic potential when analysing the functions and roles of the microorganisms.
In the early days when the HMP was being formed, the NIH officially defined the human microbiome as 'all of the microorganisms and their genes and genomes which make up the microbial communities that inhabit the human body'. This definition was meant to reflect that we were not just considering microbial composition but also the genetic potential of these communities. Of course what this does not reflect is any relationship between the human host and its microbial inhabitants. This is where the 'human metagenome' (ala Jeff Gordon) does a better job of capturing the host/microbial community ecosystem.
Elio replies:
I will be posting the results of the poll soon. In the meanwhile, I am thankful to Lita for this information. I'm glad that the NIH has an official definition of microbiome. However, as Lita points out there is still room for discussion. For one thing, to say that the microbiome consists of "all the microorganisms and their genes" seems like a tautology or something. Of course the organism carry their genes. It goes without saying.