An international team led by the University of California San Diego has spearheaded an effort to reconcile years of microbiome studies, bringing together two primary sequencing techniques—16S ribosomal RNA gene amplicon (16S) and shotgun metagenomics sequencing. In a significant advancement for microbiome research, the team recently announced the creation of a reference database, Greengenes2, that allows for the integration and comparison of microbiome data derived from either technique.
Published in Nature Biotechnology, the study introduces Greengenes2, a comprehensive database that promises to revitalize over a decade's worth of 16S data. The standardization of results across both methods is expected to significantly enhance the discovery of microbiome biomarkers for health and disease, according to senior author Rob Knight, Ph.D., a professor in several departments at UC San Diego.
Microbiome studies hinge on the identification of the microorganisms present in a given sample, which is achieved by sequencing the genetic information in the sample and comparing it to reference databases. Until now, data from the two main techniques, 16S and shotgun sequencing, were considered too disparate to be combined, a notion disproven by the new study.
The newly minted Greengenes2 builds upon its predecessor, the original Greengenes database, widely used in the microbiome field for over a decade. However, the original version had its limitations, as it relied on the sequence of a single gene, 16S, to identify organisms. Modern microbiome studies have since moved towards using shotgun sequencing, which provides more species-level specificity and functional insights about the microbes.
Differences between the 16S and shotgun techniques were previously attributed to variations in lab sample preparation. Though, the new study shows that these disparities arise from computational differences. By creating a more comprehensive reference database, the researchers have enabled the two methods to draw consistent conclusions, enhancing the reproducibility of microbiome research and the reusability of older data.
To develop Greengenes2, researchers expanded the Web of Life whole genome database and integrated existing high-quality full-length 16S sequences into the whole-genome phylogeny. They then placed 16S fragments from over 300,000 microbiome samples, resulting in an expansive reference database that both 16S and shotgun sequencing data could be mapped onto.
The successful standardization of findings from both sequencing techniques was demonstrated by analyzing 16S and shotgun sequencing data from identical human microbiome samples against the Greengenes2 phylogeny. This comparison yielded highly correlated diversity assessments, taxonomic profiles, and effect sizes—a first in the field.
This development, according to Daniel McDonald, Ph.D., the study's first author, signifies a major step in improving the reproducibility of microbiome studies and enhances clinicians' ability to draw clinical conclusions from microbiome data. With the advent of Greengenes2, a vast repository of 16S data can now be combined with modern shotgun data, thereby enriching future meta-analyses.