Comparison between 16S rRNA and shotgun sequencing data for the characterization of the human gut microbiota

The gut microbiome is an important aspect of human health given it is role in several homeostatic and pathologic processes. 16S rRNA gene sequencing and shotgun metagenomics are the two main approaches for microbiome analysis. The first part of this thesis compares approaches to shotgun metagenomics taxonomic profiling: I confronted three reference databases and compared assembly-free and assembly-based metagenomics. I concluded that assembly-free metagenomics against GTDB95 database was the most suitable for the subsequent research. The second part of the investigation compares 16S rRNA gene sequencing and shotgun metagenomics for enterotyping, taxonomic profiling and functional profiling. Simulated data analyses showed that shotgun metagenomics presents more false positives, but its overall score is higher at genus and species levels. Real data indicates that shotgun metagenomics identify more taxa, particularly at lower taxonomic levels. Even though the two methodologies return different abundances for a few taxa, overall, they present good relative abundance correlation. As for enterotyping, shotgun metagenomics returned a larger amount of information. In conclusion, shotgun metagenomics may be advantageous for enterotyping and other research that require higher taxonomic resolution, while 16S rRNA gene sequencing may be used for qualitative research up to family level.