Fig. 5: Social and microbiome strain niches.
From: Gut microbiome strain-sharing within isolated village social networks
a, People with a higher clustering coefficient are, on average, more similar to their first-degree connections (n = 1,753; two-sided Wilcoxon rank-sum tests). Data are represented as boxplots where the middle line is the median and the lower and upper hinges correspond to the first and third quartiles. The whiskers extend from the hinge to the largest or smallest value no further than 1.5× IQR from the hinge. b, Microbiome strain-sharing-rate Louvain clusters for the village of Basauri. Ties are weighted and sized according to the strain-sharing rate between the pair of people (modularity = 0.18). c, Social network Louvain clusters for the village of Basauri (modularity = 0.58). d, Microbiome cluster membership painted onto the social network. There is visual overlap between communities detected solely on the basis of shared microbe sets and communities detected based solely on social connections. e, P value distributions for clustering results. Histograms represent the null distribution of adjusted Rand index values from 10,000 microbiome permutations. All villages are significant (permutation test for Azpeitia, Basauri, Erandio, Ermua, Getxo, Hernani, Hondarribia, Irun, Lekeitio, Pasaia, Tolosa and Zumaia, P = 0; Bermeo, P = 1.3 × 10−3; Getaria, P = 4 × 10−4; Laguardia, P = 1 × 10−4; Mungia, P = 3 × 10−4; Sestao, P = 2 × 10−4; Zarautz, P = 2 × 10−4), with the observed overlap metric represented by the vertical red line. f,g, Examples of two differentially abundant species, Enterococcus faecium (f) and SGB14372 (g), within the village of Hondarribia. Nodes are scaled according to the log relative abundance of the species, with yellow indicating presence of the species and red indicating absence.
