Can You Use Your American Gut Results to Inform Major Lifestyle Changes?

By Embriette Hyde, PhD, Project Manager of the American Gut Project

Here at American Gut we have a big job. Not only are we amassing the world’s largest human microbiome sequencing and metadata dataset, and making that available to anyone around the world, but we are also navigating the choppy waters that inevitably always occur when something big has the potential to promise improved health in the future. Fecal transplants are a great example of this. They have been incredibly successful in the context of Clostridium difficile infection, and this together with identified associations between the microbiome and a growing number of diseases as varied as inflammatory bowel disease, autism, diabetes, and depression, had led many to wonder whether a fecal transplant is the answer for them. As the FDA only approves fecal transplants in the context of C. diff-and with good reason; we have no idea what the long-term effects of fecal transplants are-many are turning to self fecal transplants since they cannot receive them in a clinic setting. While we understand what drives people to these decisions, the American Gut Project does not condone or encourage this action; in fact, we discourage it. Also important-the results we produce for our participants are not sufficient to be used to make such a decision, and should not be used to inform major (or even minor) lifestyle changes.

I became very concerned when I read a pair of articles this morning. The first article was actually a blog post response written by Dr. Jonathan Eisen after reading an article published by Bicycling Magazine, in which it describes a young woman performing a DIY fecal transplant after submitting her sample to American Gut and then learning that “she was populated by 96% gram-negative pathogens so toxic that if they got into her blood stream they could kill her [and] had picked up bugs in the lab where [she] was working because [her] system was so weak and susceptible.” Jonathan Eisen rightly states “I don’t think the American Gut project can say anything about pathogens nor do I think they do say this. My guess is this is a misinterpretation by the scientist here or more likely the reporter. Also I doubt the American Gut data could be used to say anything about picking up bugs in the lab.” I appreciate Jonathan’s quick, correct, and responsible dialogue on this, and I want him to know that we thank him for enforcing accountability and setting the record straight regarding what American Gut can and cannot do. I also want to add my own input on this, as I feel this is incredibly important.

The American Gut Project is strictly a research project. The results we provide our participants are provided to them so that they have fun exploring what was found in their sample(s), and ultimately to get something back for their contribution to our research. The research benefits more from the data than the participants do at this point, and we are still in the data collection phase. Humans are complex-we are a bag of DNA (human and microbial), chemicals, and microbes, all of which work together and react to specific environmental inputs. My favorite example of this is a recent study showing that the same exact food can have completely opposite effects on blood sugar levels in two different bags of DNA, chemicals, and microbes (i.e., two different people). Of course we want to get to the point where all of this does change human healthcare, but we are still a long way off from that.

Most of the results produced by the Project are based on sequencing of a small portion of a single gene-the 16S ribosomal RNA gene. The information we are able to gather from this is limited to taxonomy and abundance, and even that is limited, as one cannot confidently assign species level taxonomy from 16S marker gene sequencing. Best case scenario, you can get genus level results. What does this mean? To give a bit of context, consider the following: the Bacillus genus contains a common cause of food poisoning (B. cereus), a commonly used biological pesticide (B. thuringiensis), and the causative agent of anthrax (B. anthracis). Genetically, these organisms should be considered the same species, with the major genetic differences giving them their different modes of action found on plasmids rather than their own chromosomal DNA. Now, the Bacillus genus also contains plenty of non-pathogenic species that can be found in humans. The fact is, the American Gut Project cannot say what specific Bacillus species was present in a person’s sample-pathogens could be present, but so could completely benign commensals. In addition to that, even the exact same species can have several variants-called strains-which are like the difference between you and a sibling. Genetically you are more similar to each other than either of you is to your next door neighbor, but you have certain differences in certain genes that make you slightly different from one another. So, while my brother and I are both of the species Hyde, we are definitely different strains! The anthrax present in a sheep field, for example, is not the same anthrax present in a research lab. What am I trying to say through all of this? It is impossible for one to look at their American Gut results and say that toxic pathogens, some of which were picked up from a laboratory setting, were present in a sample.

Am I concerned that the young woman in the article performed a DIY fecal transplant? Yes. However, she admits she took a risk, and she made her decision based on her own personal weighing of risks and potential benefits. What concerns me most is the way the American Gut Project was construed in the original Bicycling Magazine article, and I wanted to take steps to help ensure that this (which is already making quite a presence in the Twitterverse) doesn’t mislead current and future American Gut participants. If anyone has any questions or concerns about this topic or any other, please do not hesitate to contact me at