Healthy human gut microbiome: Towards standardized research.

Objective: An increasing number of international researchers are focusing on the taxonomic composition of fecal microbiota and its correlation with disorders. Thousands of researchers compare conditionally healthy cohorts to those with specific diseases to identify potential markers. However, clinical application requires assessing the feasibility of synthesizing these findings and establishing reference intervals for normal gut flora, at least at higher taxonomic levels. Design: This study involves a systematic review and meta-analysis of human gut microbiota research based on 16S rRNA gene next-generation sequencing (NGS). Relevant research was sourced following the PRISMA guidelines. Descriptive statistics, linear regression analysis by weighted least squares method, Mann-Whitney test, and Benjamini-Hochberg procedure adjustments were employed. The study has been registered with PROSPERO (CRD42023431467). Results: Of the 4,346 studies initially identified, 86 publications involving 20,748 unique participants met the quality criteria and were included in the analysis of the impact of fecal sample preparation on taxonomic composition. The phylotype composition, in relation to preprocessing methods and cohort locations, are presented as relative abundances (%): Bacillota (median 49.5-59.6%), Bacteroidota (28.0-33.4%), Pseudomonadota (3.4-5.9%), Actinomycetota (2.3-3.7%), Verrucomicrobiota (0.5-1.0%), Fusobacteriota (maximum 4.6%), and Euryarchaeota (maximum 2.11%). The content of 27 key family-level representatives was also evaluated. The well-known hypothesis regarding the influence of the homogenization stage on taxonomic composition was examined using generalized results. Conclusion: While supported by a strong theoretical basis and evidence from individual practical cases, none of the phyla showed a statistically significant association and consistent relationship with sample preparation or cohort location when generalizing across studies after the two exceptionally large cohorts exclusion, both originating from a single research group. These findings underscore the need for strict methodological standardization in microbiome studies. Key features of the 16S NGS process accounting for these results are outlined, along with proposed optimizations for microbiome research.