Fig. 4
Identification of functional microbial and fungal communities associated with diet and disease and their association with transcriptome data. Functional microbial (a) and fungal (b) community (FMC, FFC, respectively)–trait correlations and adjusted P-values. Each value within the cell without the brackets shows the Spearman correlation coefficient between covariate of interest and eigenOTUs. The adjusted P-values are given within the brackets in each cell were derived using Wilcoxon test for sex (microbial, nfemales = 24 and nmales = 31; fungal, nfemales = 15 and nmales = 17) and disease at final stage (microbial, nyes = 22/nno = 31; fungal, nyes = 14/nno = 17), and the Kruskal–Wallis test for various disease stages (microbial, nnaive = 50, ntrans = 55, and nfinal = 53; fungal, nnaive = 29, ntrans = 32, and nfinal = 31) and diet (microbial, ncalorie restricted = 18, ncontrol = 17, and nwestern = 20; fungal, ncalorie restricted = 11, ncontrol = 11, and nwestern = 10) and were further adjusted for multiple comparisons across different modules using Benjamini-Hochberg correction procedure. The degree of correlation is indicated by the color of the cell in accordance with the color legend. c Heat map showing the correlation between FMC1 (rows) and FFC4 (columns) OTUs (species identified by both RDP and NCBI BLAST) calculated by the SparCC algorithm. The color codes and of the cells indicate either positive (purple) or negative (orange) correlations among the species (Padj < 0.05), while the size of the cell positively correlates with the degree of the correlation. d Heatmap describing differentially expressed genes intersecting between disease (n = 16) and no disease (n = 16), diet groups i.e calorie-reduced (n = 11), control (n = 11) and western (n = 10) and associated with FMC1 and FFC4. The color-coded heatmap shows purple as high expression and yellow as low expression of genes (rows) among the samples belonging to different groups (columns). Source data for (a–d) are provided in the Source Data file
