Your gut microbiome affects all aspects of your health, not just digestive health. It controls your mood and happiness, inflammatory responses, metabolism, hormonal health, and much more. The biome is a dynamic reservoir of approximately 22 million genes that allow you to adapt to different diets and environments.
Next-generation sequencing technologies have taught us a lot about the microbiome in recent years. These technologies, which are quickly adopted by biohackers and natural health practitioners, have also made it possible for consumers to test their stool microbiome.
Problems and potentially harmful consequences lie where the nuances get lost in echo chambers and marketing messages for such complex health topics. In this article, I’ll explain the full story, as a molecular biologist and functional health practitioner.
Gut Microbiome Testing Technologies: How Do They Work?
Before next-gen sequencing (NGS), scientists had to isolate and grow each bacteria strain to extract sufficient DNA to analyze them. However, up to 98% of bacteria on earth do not grow in the lab because microbiologists have not figured out ideal growth conditions for them.1
NGS is game-changing for microbiome research because we can now determine the genetic sequence of the microbes in a mixed sample like your stool. It is easy to isolate the combined DNA of the stool sample and determine the sequence without the need to grow the bacteria.
The use of NGS to determine the genetic and species composition of a mixed sample is called “meta-sequencing.” Meta-sequencing means determining the sequences of multiple small fragments in parallel and counting the number of occurrences of these small fragments. The number of occurrences can be an indirect measure of relative abundance or the number of bacteria each species.
There are many different competing next-gen sequencing technologies such as the following:
- Illumina is currently the most predominant NGS company on the market. Sequencing with Illumina involves binding each DNA fragment to a glass slide and replicating the DNA sequence on the slides with nucleotides that are labeled with different colors of fluorescent dyes. Then, the machine takes a series of images with fluorescent colored dots to determine the DNA sequence.
- Nanopore puts the DNA sequence through a protein channel and measures different proton charges for each nucleotide base.
- Sequencing with Roche 454 involves tethering each DNA fragment to a bead, which is then placed in its own small well before replicating the DNA. When the correct nucleotide is incorporated, the well emits a fluorescent signal. Then, the machine measures light emission to detect the DNA sequence.
The DNA sequences from these sequencers are then assembled into genomes and interpreted with computer algorithms. Typically, the analysis involves comparing the sequence to existing bacterial genomes and genes.
Different Ways to Test Your Microbiome
The 16S rRNA gene is a small gene of only 1550 base pairs and contains the unique signature for every bacteria species. Therefore, by choosing only to sequence this gene and count the number of occurrences of the same sequence, you can tell the composition of bacteria species in a sample.
The cost per base pair of NGS is going down rapidly, but it still depends on how many base pairs you’re sequencing in total. Therefore, 16S rRNA sequencing is the cheapest way to determine the species composition of a mixed microbiome sample. Most microbiome studies use this method.
A test that uses this method is Atlas Microbiome.
Whole-Genome Shotgun Sequencing
Your gut bacteria can acquire or lose genes, even though the number of cells and species remain the same. Therefore, a Lactobacillus in one person may have different genes than the next person. In contrast to 16S rRNA Sequencing, whole-genome sequencing won’t only tell the number of species you have, but also the composition of all the genes your stool bacteria have.
“Shotgun” means breaking down the DNA into numerous small fragments of roughly the same size. Then, the sequencing process determines each fragment before a computer program assembles them into bacterial genomes and another program predicts genes.
Tests that use this method are DayTwo and BiomeFX.
Simply because a gene exists in the gene pool doesn’t mean it’s being utilized. One way to measure if a gene is being utilized is to test for its transcripts or messenger RNA (mRNA). Transcriptome sequencing is a way to test for the total transcript readouts of the bacteria genome. In other words, it takes a collective snapshot of the gene activity of your gut bacteria.
A test that uses this method is Viome.
Atlas Biome vs. Viome vs. DayTwo Reviews
Because these different types of tests measure different things, it is totally possible and likely to get very different results on the same sample.2 Interestingly, each company claims that its test is best. The truth is that there are pros and cons for each of these technologies.
The microbiome and its activity are dynamic. The more data you collect, the more noise you will have. Transcriptome generates a larger dataset than whole-genome and 16S rRNA. While transcriptome sequencing may be the most reflective of your microbiome biochemical activities, it is also the most dynamic, and therefore tends to generate the most noise.
While these parameters may affect the accuracy of the data, none of what we know about each microbiome test reflects the accuracy of the health recommendations that these companies provide.
Viome now also offers the Health Intelligence Test, which tests for your genetic expression (transcriptome) in your blood in cellular, mitochondrial, immune, stress response, and biological age categories.
Microbiome Test Caveats: Should You Trust Them?
Your Stool Microbiome Does Not Equal Your Gut Microbiome
Most of your gut microbes reside in your large intestine, which is not a unified fermentation chamber. A healthy gut biome looks more like a multilayered lush jungle of the Amazon. Many bacteria build biofilms to colonize in the gut, while others simply bind to the gut lining and the mucus that covers it. Therefore, the microbes near your gut lining are distinct from those in the lumen of the gut. Also, those in the upper colon are different than those in the lower colon and the stool.3
Therefore, it’s extremely challenging to study the gut microbiome in a condition that fully reflects its complexity and the complexity of its interaction with your body.4 Some microbiologists have attempted to develop artificial chambers to simulate the human gut condition. In most other cases, the microbiome studies have relied on the stool microbiome.
As with many other medical treatments, the rodent microbiome studies tend to show remarkable results, which don’t always apply to humans. In fact, I’ve never observed any benefits from 99% of probiotic supplements both for myself and my health coaching clients, despite rodent studies and even some clinical studies showing their specific benefits.
Because the applicability of stool microbiome studies are questionable, their results need to be taken with a grain of salt. They also tend to contradict each other (which is also typical in science), suggesting that there are many unknown individualized factors we have yet to discover.
Although the stool microbiome is distinct from the gut microbiome, stool microbiome testing isn’t 100% useless. You need to look at your stool test result as an output of an ever-moving system, not as an actual representation of your gut microbiome. For example, your stool can have an abundance of a bacteria strain because the strain doesn’t colonize in your gut.
In the realm of lab test interpretation, looking at markers as an output of a system is not unheard of.
As a parallel example, currently, the best functional hormone test is the Dried Urine Test for Comprehensive Hormones (DUTCH), which gives a snapshot of your hormone metabolism. Instead of directly measuring hormones that reach your cells, it measures the hormones and hormone metabolites that you excrete in your urine.
In my opinion, output tests like stool microbiome and DUTCH require clinical expertise to correlate each finding with symptoms, history, and health goals. They give a lot of data, such as a list of bacteria strains, many of which have no known clinical significance or cannot be easily changed. Also, research has barely scratched the surface with understanding how each of these bacteria affects your health.
Your Microbiome Test Provides a Snapshot of a Changing System’s Output
Your microbiome is the extra gene reservoir that allows you to adapt to new foods and environments. Also, new bacteria can colonize your gut if you eat the food that it prefers. For example, only Japanese people who eat seaweed have bacteria with enzymes that can digest seaweed carbohydrates.
Also, you have a two-way relationship with your gut bacteria. They influence your food preference, mood, brain function, and health status. Conversely, your diet, lifestyle, mindset, and health state also determine the bacteria that thrive in your gut.
Changing your diet can entirely change your stool bacteria composition within about 24–48 hours!5 Therefore, what you eat is a major determinant of gut bacteria composition, even more so than any probiotic supplement can.
Your gut bacteria will reflect if you’re stressed out, sleep-deprived, lonely, or inflamed.6,7,8,9 If this explains you, your test will show dysbiosis because your body is less habitable to friendly bacteria. You may have completely different test results within a few days if you’ve changed the lifestyle parameters that influence your gut microbiome.
NGS Technologies are Still Error-Prone
NGS technologies are rapidly improving by the day, but they are still so error-prone that most medical organizations recommend against relying on them for diagnostic purposes. This is another reason to take your NGS stool microbiome test with a grain of salt.
Are Food Recommendations by Viome and DayTwo Reliable?
What you eat strongly influences your gut microbiome, which influences your food preference.5,10 Therefore, people often find that their Viome tests recommend the food they already like. However, no well-controlled study has been published to demonstrate that following their food recommendations is truly beneficial. Viome’s website indicates that this clinical trial is ongoing.
DayTwo spun off from a clinical study that involved 800 people. The study used a machine learning algorithm that integrated blood tests, dietary habits, anthropometric measurements, physical activity, and gut microbiome to predict post-meal blood sugar responses. They then tested the predictions on 100 subjects and performed a blinded, randomized, controlled trial to test their intervention. The study demonstrated that following the intervention based on this algorithm successfully lowered post-meal blood sugar and changed the gut microbiome composition.11
Thanks to NGS and improvements in computation, now is a truly exciting time for gut microbiome research and to harness the research findings to improve your health. However, the technologies are imperfect, and we still know very little about how to leverage our microbiome for health. Therefore, it is a good idea to take gut microbiome tests with a grain of salt and work with a qualified practitioner to integrate all pieces of data in your health journey.
- Wade, William. “Unculturable Bacteria—the Uncharacterized Organisms That Cause Oral Infections.” Journal of the Royal Society of Medicine 95, no. 2 (2002): 81–83. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1279316/
- Kehoe, Thomas David. “Are Microbiome Food Recommendations Real or Hype?” Medium (2018). https://tdkehoe.medium.com/my-viome-microbiome-test-results-afab2b3df894.
- Zmora, Niv, Gili Zilberman-Schapira, Jotham Suez, Uria Mor, Mally Dori-Bachash, Stavros Bashiardes, Eran Kotler, et al. “Personalized Gut Mucosal Colonization Resistance to Empiric Probiotics Is Associated with Unique Host and Microbiome Features.” Cell 174, no. 6 (2018): 1388–1405.e21. doi:10.1016/j.cell.2018.08.041.
- Tang, Qiang, Ge Jin, Gang Wang, Tianyu Liu, Xiang Liu, Bangmao Wang, and Hailong Cao. “Current Sampling Methods for Gut Microbiota: A Call for More Precise Devices.” Frontiers in Cellular and Infection Microbiology 10 (2020): 151. doi:10.3389/fcimb.2020.00151.
- David, Lawrence A., Corinne F. Maurice, Rachel N. Carmody, David B. Gootenberg, Julie E. Button, Benjamin E. Wolfe, Alisha V. Ling, et al. “Diet Rapidly and Reproducibly Alters the Human Gut Microbiome.” Nature 505, no. 7484 (2014): 559–63. doi:10.1038/nature12820.
- Madison, Annelise, and Janice K. Kiecolt-Glaser. “Stress, Depression, Diet, and the Gut Microbiota: Human-Bacteria Interactions at the Core of Psychoneuroimmunology and Nutrition.” Current Opinion in Behavioral Sciences 28 (2019): 105–10. doi:10.1016/j.cobeha.2019.01.011.
- Smith, Robert P., Cole Easson, Sarah M. Lyle, Ritishka Kapoor, Chase P. Donnelly, Eileen J. Davidson, Esha Parikh, Jose V. Lopez, and Jaime L. Tartar. “Gut Microbiome Diversity Is Associated with Sleep Physiology in Humans.” PloS One 14, no. 10 (2019): e0222394. doi:10.1371/journal.pone.0222394.
- Dill-McFarland, Kimberly A., Zheng-Zheng Tang, Julia H. Kemis, Robert L. Kerby, Guanhua Chen, Alberto Palloni, Thomas Sorenson, Federico E. Rey, and Pamela Herd. “Close Social Relationships Correlate with Human Gut Microbiota Composition.” Scientific Reports 9, no. 1 (2019): 703. doi:10.1038/s41598-018-37298-9.
- Zeng, M. Y., N. Inohara, and G. Nuñez. “Mechanisms of Inflammation-Driven Bacterial Dysbiosis in the Gut.” Mucosal Immunology 10, no. 1 (2017): 18–26. doi:10.1038/mi.2016.75.
- Alcock, Joe, Carlo C. Maley, and C. Athena Aktipis. “Is Eating Behavior Manipulated by the Gastrointestinal Microbiota? Evolutionary Pressures and Potential Mechanisms.” BioEssays: News and Reviews in Molecular, Cellular and Developmental Biology 36, no. 10 ( 2014): 940–49. doi:10.1002/bies.201400071.
- Zeevi, David, Tal Korem, Niv Zmora, David Israeli, Daphna Rothschild, Adina Weinberger, Orly Ben-Yacov, et al. “Personalized Nutrition by Prediction of Glycemic Responses.” Cell 163, no. 5 (2015): 1079–94. doi:10.1016/j.cell.2015.11.001.