Improved detection of microbiome-disease associations via population structure-aware generalized linear mixed effects models (microSLAM)

Microbiome association studies typically link host disease or other traits to summary statistics measured in metagenomics data, such as diversity or taxonomic composition. But identifying disease-associated species based on their relative abundance does not provide insight into why these microbes act as disease markers, and it overlooks cases where disease risk is related to specific strains with unique biological functions. To bridge this knowledge gap, we developed microSLAM, a mixed-effects model and an R package that performs association tests that connect host traits to the presence/absence of genes within each microbiome species, while accounting for strain genetic relatedness across hosts. Traits can be quantitative or binary (such as case/control). MicroSLAM is fit in three steps for each species. The first step estimates population structure across hosts. Step two calculates the association between population structure and the trait, enabling detection of species for which a subset of related strains confer risk. To identify specific genes whose presence/absence across diverse strains is associated with the trait, step three models the trait as a function of gene occurrence plus random effects estimated from step two. Applying microSLAM to 710 gut metagenomes from inflammatory bowel disease (IBD) samples, we discovered 49 species whose population structure correlates with IBD. In addition, after controlling for population structure, we found 57 microbial genes that are significantly more common in healthy individuals and 26 that are more common in IBD patients, including a seven-gene operon in Faecalibacterium prausnitzii that is involved in utilization of fructoselysine from the gut environment. Overall, microSLAM detected IBD associations for 45 species that were not detected using relative abundance tests, and it identified specific strains and genes underlying IBD associations for 13 other species. These findings highlight the importance of accounting for within-species genetic variation in microbiome studies. Author SummaryThe species composition of the human gut microbiome differs significantly between individuals and is associated with various diseases. Many studies have sought to understand this relationship by examining the relative amount of each bacterial species within metagenomic sequencing data from sick and healthy individuals. However, this approach makes it challenging to pinpoint which genes and pathways of a disease-associated species might actually contribute to disease risk, and it misses species where only certain strains are disease associated. To overcome these challenges, we developed an R package, called microSLAM, that uses mixed-effects modeling to associate within-species genetic variation to host traits. microSLAM performs two types of tests for each species: one for identifying strain-disease associations and another for identifying gene-disease associations. The gene tests account for the genetic relatedness of strains across hosts, making them particularly useful for detecting mobile genes. We applied microSLAM to hundreds of gut metagenomes from inflammatory bowel disease studies, identifying dozens of novel associations that were missed using relative abundance tests. MicroSLAM is a general modeling approach that can be applied to human traits beyond disease case/control studies and to microbiomes from other environments.