Saturday, September 20, 2014

Notes from 2007 for a blog post I should have written: How many microbial cells in humans?

Well sometimes you just screw up.  In 2007 I attended some planning meetings for the human microbiome project (see for example A human microbiome program? a post I wrote from one of the meetings in 2007).  And at those meetings I kept asking one question.  Where did this "fact" everyone kept citing that there were "10 times as many microbial cells in the human body as there were human cells" come from?  I could not find a citation.  So I started taking some notes for a blog post about this.  Here are those notes:

Wikipedia linkOnline textbook hereSears paper from Arizona site. She discusses only gut bacteria and cites a Gordon paper from 2001.
Seems to not be from this paper but really from here:
This in turn is not from there but apparently here

But, alas I got distracted.  And I did keep asking people - where did this "fact" come from.  And most people just brushed me off (and probably thought I was a bit of a crank ...). And nobody had a good answer.  Well, I was both pleased and sad (because I should have done it) to see Is your body mostly microbes? Actually, we have no idea by Peter Andrey Smith in the Boston Globe who addresses this issue in much much more detail that I ever could have done.  Everyone who works on the human microbiome and who is interested in "facts" and how they can get misreported should read this.  As a side note, Smith reports in the article that this is even given as a fact in Ted talks.  Sadly mine was one of them.  This is despite the fact (yes, the fact) that I swore to myself that I would NOT say that in my talk since I have been such a crank about this issue at meetings.  OMG - such truisms are so pervasive that even someone who actively questioned the truism still used it.  Uggh.  Oh well.  I really should have finished that draft post.

8 comments:

  1. I've often wondered about that "fact" too. Here's as far as I got last year: http://nsaunders.wordpress.com/2013/10/31/were-only-10-human-according-to-who/

    ReplyDelete
    Replies
    1. And I note that Judah Rosner, author of the letter to Microbe mentioned in the Boston Globe article, commented on my post at the time...

      Delete
  2. Does this go for the "microbial genes 100:1" truism as well? Seems like we might have a bit more data on that front.

    ReplyDelete
    Replies
    1. Well we have some data on that but I think the analysis is generally fundamentally flawed ... Most of the analysis I have seen would count a RecA gene in bacteria1 and being different than a RecA gene in bacteria 2. That seems to be a bad idea. I think protein family #s would be more useful ...

      Delete
    2. But then would we have to eliminate redundant/duplicate genes in human genome as well? At some point it's going to come down to semantics. So maybe my problem is that I don't even know what the point of citing one of these numbers is - what's the point that it's trying to make? That bacteria are important because there are a lot of them?

      Delete
    3. I should add - I have used these numbers in talks because they sounded impressive... I'm rather chagrined by that and not trying to get on a high horse about it.

      Delete
  3. I don't think it is completely about semantics. I think the sum total genetic potential in the microbiome vs. the host is an interesting question. But I hate it when numbers are kind of made up or misleading. In this case, if we really want to compare genetic potential then the following two genes: 1 RecA from E. coli and 1 RecA from Salmonella should not count the same as these two: 1 RecA from E. coli and 1 Photolyase from Salmonella. But in the methods being used they sometimes do..

    ReplyDelete