Identification of microbiome-host genetics associations using full-length 16S rRNA gene sequencing
Human Genetics
Final Report Abstract
The human gut microbiota – which is the collection of all microorganisms living in the gut – is in constant interaction with its host and has been implied as a modulation factor in health and a multitude of diseases with (auto-)inflammatory components, such as inflammatory bowel disease, rheumatoid arthritis, and multiple sclerosis. While environmental and dietary factors are expected to be major drivers of the diversity and composition of the gut microbiome, genetically encoded host factors are likely to have profound influence on shaping host-microbiome interactions through a diversity of potential active and passive mechanisms. We used microbiome abundance data inferred from full-length 16S amplicon sequencing in combination with human genetic data from genotyping and imputation to perform statistical analysis to find associations of microbial abundance and prevalence patterns and human genetic variation across multiple German cohorts, a method often referred to microbiome-based genome wide association studies (mGWAS) or microbiome quantitative trait locus (mQTL) mapping. We aimed to sequence and analyze full-length 16S rRNA gene-based microbiome profiles of ~8,500 German individuals in five cohorts from Kiel, Greifswald and Potsdam. Data generation was delayed due to COVID-19 pandemic, however, was completed for roughly half of the initially included samples, is ongoing and will be completed in early 2025. Currently available profiles confirm robust workflows and processing steps in the sample preparation and show strong consistency of abundance profiles across cohorts with only minor influence of sample collection heterogeneity between cohorts. Preliminary mGWAS was carried out on a subset ~2,500 individuals for which matching fully processed genetic and microbiome data were available for a total of 308 traits derived from abundance and prevalence profiles of microbial species. The preliminary cohort-level results were combined in an inverse-variance meta-analysis which identified two novel human genomic loci with genome-wide significant associations on chromosomes 4 and 15, with an additional over 300 loci surpassing an intermediary threshold of p < 10^-5. These loci include genomic loci with previously identified association signals, further substantiating their potential involvement in host-microbiome interactions. Further modes of trait definition using co-abundance networks are set up and will be applied after completion of data generation. Downstream analyses using mendelian randomization to identify causal connections between microbiome and host traits based on GWAS summary statistics are in place and ready to be applied to the finalized set of results. In summary, preliminary data analysis shows high potential of replicating previous and identifying novel human genomic loci involve in host-microbiome interactions and will be highly relevant for the discovery of candidates, generation of new hypothesis and their functional investigation.
