Fig. 1: Characteristics of participants and microbiome composition across studies.

a, Density plots of age of participants in the discovery studies (SCAPIS, n = 8,733; SIMPLER-V, n = 4,515; SIMPLER-U, n = 981; MOS, n = 1,788; total n = 16,017 individuals) and in HUNT. Dashed line: combined discovery studies. b, Violin and boxplots of BMI of participants in the discovery studies (SCAPIS, n = 8,733; SIMPLER-V, n = 4,512; SIMPLER-U, n = 978; MOS, n = 1,788; total n = 16,011) and in HUNT (n = 12,652). Violin plots show the density distribution. The boxplots within the violin plots show the medians and the IQR, and whiskers extend to the values no larger than 1.5 times the IQR (upper whisker) or smaller than 1.5 times the IQR (lower whisker). Outliers are depicted as individual points. c, Map with the study sites for the discovery studies in Sweden (SCAPIS, SIMPLER-V, SIMPLER-U and MOS) and the replication cohort in Norway (HUNT), including the sample size and proportion of female participants in each study. d, Prevalence for the species analyzed with the logistic model. e, The log-transformed median abundance for the species analyzed with the linear model in the discovery studies. In d and e, each dot represents one species. Species are ranked by their prevalence and median abundance in SCAPIS.