Fig. 3

Global analysis of intervention groups after 18-months reveals that the serum metabolome is influenced by sex. (A) PCA displayed some overlap of D + E males and females whereas (B) PLS-DA perfectly separated D + E males and females suggesting that the serum metabolome is potentially influenced by participant sex. (C) Fold change analysis identified 568 metabolite features that had a FC > 2 and were higher in abundance in D + E females. Conversely, 532 had a FC < -2 and were higher in abundance in D + E males. Similarly (D) PCA, (E) PLS-DA, and (F) fold change was applied and revealed a similar sexual dimorphic pattern when comparing male and female E participants. This same suite of analyses was applied to examine sex-associated metabolic patterns between male and female D participants (G–I). Populations of metabolite features distinguished by fold change analyses were then subjected to functional pathway enrichment analyses to pinpoint biological pathways. Considering differences in metabolic regulation associated with sex across D + E (A–C), E (D-F) and E (G-I) participants, it is evident that that serum metabolome of IDEA participants is influenced by sex. The colors in A–I correspond to: light and dark blue – female and male diet and exercise (A + E) participants; light and dark purple – female and male exercise (E) participants; light and dark green– female and male diet participants