Several years ago, I was part of the Human Microbiome Project, so I’m not someone who’s disgruntled that the microbiomes are getting all the money. On the other hand, that has also led me to have a very skeptical opinion on most of the ‘your microbiome governs X’ stories. To say some of the work is a bit oversold would be an understatement. So this article by Sarah Zhang who had the microbiome of her apartment sequenced, to me, is pitch-perfect* (boldface mine):

Three years ago, the human microbiome was just bursting into the mainstream, and scientists were just refining new tools to study bacteria out in the world. Mysterious bugs to be found in my apartment? Sign me up!

When the bacterial data was first released to participants several months ago, I spoke with Rob Dunn, a North Carolina State biologist who heads up Wild Life of Our Homes. Dunn walked me through some of the thousand-plus kinds of bacteria found on my door frame, including Sphingomonadaceae, a soil microbe, and Corynebacteriaceae, common in armpits. There was also, of course, also fecal-associated bacteria, just as my roommate probably feared. This all seemed nice to know, but I could already tell you there was soil and skin in my apartment.

As I paged through the study published today aggregating data from 1,200 homes, I felt what has since become common reading microbiome papers: disappointment. Not because the science was bad but because the findings seemed underwhelming. This study found that bacteria in household dust varies depending on the people and pets living in the house, while the fungi varies depending on the outdoor environment. Not exactly groundbreaking, right?

That doesn’t mean these kinds of studies aren’t important:

The tools for sequencing microbes in the environment are so new that scientists are only beginning to know what they don’t know….

But metagenomics still has its limits. Those algorithms usually rely on reference libraries of known genomes; if a particular bacteria completely unknown to science shows up, the algorithms don’t really know to look for it. That’s a false negative, but metagenomics is also prone to false positives—like when plague supposedly showed up on the New York City subway.

Put another way, microbiology is still in its cataloguing phase, akin to 18th century botanists pressing flowers into their notebooks.

This does make it exciting, but most of the ‘we’ll change your microbiome (very soon) and make your life vastly better’ claims need to be scaled back. The science is cool enough–and in applied settings, important enough–as is.

*Though I’ll quibble about the claim that 90+% of microorganisms can’t be cultured; most can be, but most people are either too lazy or don’t know how.