Fig. 5

Environmental factors significantly discriminated the overall genotypes of B. longum, and arginine biosynthesis of B. longum was identified as a potential host age-associated factor. A Individual effect sizes of genome covariates (isolated province, longevous district status, age, and sex) determined by SNP-level db-RDA and PERMANOVA analysis of a local panel (Chinese strains). B Individual effect sizes of genome covariates based on a gene-level db-RDA and PERMANOVA analysis of a local panel. C Individual effect sizes of genome covariates based on SNP-level db-RDA and PERMANOVA analysis of a global panel (Chinese and Japanese strains). D Individual effect sizes of genome covariates based on gene-level db-RDA and PERMANOVA analysis of a global panel. E Manhattan plot of GWAS results for the associations of age with the genomic profiles of B. longum based on core genome SNPs (left) and genes (right). The significance threshold is indicated by a horizontal red line and was defined using Bonferroni correction with a required P-value of 0.05/number of tested variants. F Distribution of significant variants according to phylogeny and host age. G, H Triplots of the db-RDA analysis of genomic composition relative to the host province (or country), age, longevous district status, and sex. Different datasets were used for these analyses. G (left) SNP profiles of the local panel; G (right) gene profiles of the local panel; H (left) SNP profiles of the global panel, and H (right) gene profiles of the global panel. Among these analyzed phenotypes, the term “longevity/longevous district status” means whether a strain was isolated from the longevous districts or not. *P < 0.05, **P < 0.01, ***P < 0.001