Many thousands of studies about gut microbiota and 16s RNA analysis are available to us nowadays. And while many results and interesting concepts emerge from those studies, a gold standard protocol for data analysis is not yet available. Moreover, the studies are carried out by different teams from different parts of the world, their designs are not the same, DNA extraction protocols differ, and 16s RNA region sequences vary between projects. As such, sample processing variables and the numerous environmental factors influencing an individual’s gut microbiota mean association studies between the human genome and gut microbiota genome are less powerful.

Writing in Nature Microbiology, David A. Hughes and colleagues have created a new analytical pipeline that disentangles associations between human host genotype and gut microbiome variation in 3 distinct cohorts, paving the way for causal inference analyses in the field.

Researchers harmonized the analytical pipeline across three independent cohorts: the Flemish Gut Flora Project (n = 2,223) and two German cohorts (Food-Chain Plus, n=950; and PopGen, n= 717).

Using fecal 16s RNA gene sequencing, the researchers first estimated the proportion of gut microbiome variation explained by genetic variation (heritability) between individuals. In total, they identified 13 genera that were heritable. Eight were from the phylum Firmicutes, five of which were from the family Lachnospiraceae and two from Ruminococcaceae.

Dorea and Anaerostipes genera, which are short-chain fatty acid producers, along with Hespellia were three of the most heritable bacteria.


During a second stage, the researchers identified associations between bacteria species and human genes, encountering two strong associations:

  • The first one was between Ruminococcus and the rs150018970 gene. This gene is upstream of the RAPGEF1 gene, which encodes for a G-protein-coupled receptor involved in sensing metabolites from commensal bacteria in the gut. For this gene, heterozygous individuals were less likely to have Ruminococcus in their gut microbiota compared to homozygous individuals.
  • The second association was between Coprococcus and the rs561177583 gene. This gene is a non-protein-coding RNA LINC01787 and therefore, the result requires further investigation. Incidentally, heterozygous individuals are also less likely to have Coprococcus in their sample.


The two strong associations were followed by the discovery of 11 associations showing low heterogeneity. Among those 11 associations, they found a relationship between the butyrate-producing genus Butyricicoccus and the SLC5A11 gene, which is a sodium-dependent myo-inositol glucose cotransporter that is highly expressed in the brain and intestine. The findings are in agreement with previous studies suggesting that butyrate-producing bacteria are associated with blood glucose and appetite regulation.

Another association was identified between Veillonella and rs117338748. This gene is involved in regulating low-density lipoproteins (LDLs) and transporting high-density lipoproteins (HDLs). The researchers observed that the presence of Veillonella was associated with a drop in LDL-cholesterol.

Using a Mendelian randomization model, the researchers estimated relationships between 5 microbial pathways and 7 outcomes (diseases). For instance, Bifidobacterium was associated with body composition. However, in the absence of clear microbiome-driven effects, any interpretation requires caution. In other words, it could be the cause—less Bifidobacterium means the individual has a higher body mass index (BMI)—or the consequence—an individual with a higher BMI will present less Bifidobacterium in their gut.

In conclusion, this in-depth study on human genome-gut microbiome associations in 3 distinct cohorts generated a growing catalogue of genetic associations and showed better associations between the host’s genetics and its gut microbiota. Next steps should look at understanding the causation factors for a better understanding of gut microbiota function and association with outcomes.



Hughes DA, Bacigalupe R, Wang J, et al. Genome-wide associations of human gut microbiome variation and implications for causal inference analyses. Nat Microbiol. 2020. doi: 10.1038/s41564-020-0743-8.