PeerJ (Nov 2020)

Different analysis strategies of 16S rRNA gene data from rodent studies generate contrasting views of gut bacterial communities associated with diet, health and obesity

  • Jose F. Garcia-Mazcorro,
  • Jorge R. Kawas,
  • Cuauhtemoc Licona Cassani,
  • Susanne Mertens-Talcott,
  • Giuliana Noratto

DOI
https://doi.org/10.7717/peerj.10372
Journal volume & issue
Vol. 8
p. e10372

Abstract

Read online Read online

Background One of the main functions of diet is to nurture the gut microbiota and this relationship affects the health of the host. However, different analysis strategies can generate different views on the relative abundance of each microbial taxon, which can affect our conclusions about the significance of diet to gut health in lean and obese subjects. Here we explored the impact of using different analysis strategies to study the gut microbiota in a context of diet, health and obesity. Methods Over 15 million 16S rRNA gene sequences from published studies involving dietary interventions in obese laboratory rodents were analyzed. Three strategies were used to assign the 16S sequences to Operational Taxonomic Units (OTUs) based on the GreenGenes reference OTU sequence files clustered at 97% and 99% similarity. Results Different strategies to select OTUs influenced the relative abundance of all bacterial taxa, but the magnitude of this phenomenon showed a strong study effect. Different taxa showed up to 20% difference in relative abundance within the same study, depending on the analysis strategy. Very few OTUs were shared among the samples. ANOSIM test on unweighted UniFrac distances showed that study, sequencing technique, animal model, and dietary treatment (in that order) were the most important factors explaining the differences in bacterial communities. Except for obesity status, the contribution of diet and other factors to explain the variability in bacterial communities was lower when using weighted UniFrac distances. Predicted functional profile and high-level phenotypes of the microbiota showed that each study was associated with unique features and patterns. Conclusions The results confirm previous findings showing a strong study effect on gut microbial composition and raise concerns about the impact of analysis strategies on the membership and composition of the gut microbiota. This study may be helpful to guide future research aiming to investigate the relationship between diet, health, and the gut microbiota.

Keywords