Spatially variable barley root microbiomes do not imply weak host genetic control at field scale

The plant microbiome has been proposed as ‘the second genome of the plant’, motivating interest in microbiome-assisted breeding. For microbiome-assisted breeding to become a reality, the plant genome must obviously have a role in regulation of the microbiome. However, agricultural fields are spatially heterogeneous, and it remains unclear how host genetics and soil heterogeneity jointly shape plant microbiomes under field conditions. Here, using 16S and ITS metabarcoding data from 2,400 root representing 386 barley lines across spring and winter field trials, we quantified host genetic effects on root-associated fungal and bacterial communities and their modulation by soil spatial heterogeneity. Barley genotype exerted a detectable and reproducible effect on overall microbial community composition and on a subset of bacterial and fungal biomarker taxa associated with plant biomass. However, these host genetic effects were strongly modulated by field-scale spatial heterogeneity. Spatial context influenced both whole-community structure and the abundance patterns of individual host-associated biomarkers. These biomarkers differed markedly in their occupancy consistency, spatial dependency, and host Single-Nucleotide Polymorphism (SNP)-by-soil spatial structure interaction patterns, supporting the view that host-associated microbiome recruitment is context dependent rather than a fixed filtering across field environments. Consistent with this, variation of SNP-associated positive biomarkers significantly correlated with barley biomass stability, but these associations were stronger at the level of total positive biomarker abundance than at the level of individuals. Together, our findings demonstrate that large-scale field experiments are essential for dissecting host-microbiome assembly patterns and reveal that host genetic control of the root microbiome in the field is detectable but strongly context dependent, with important implications for microbiome-assisted breeding.