Extended Data Fig.11: Individual responses of the plasma proteome of participants consuming the 4-fibre snack prototype.

a,b, Heatmaps plotting the projections on SV1 of changes in the representation of biological themes during consumption of the 4-fibre snack. c, Heatmap plotting the log₂ fold-change in the levels of plasma proteins enriched in the ‘Glucose Metabolism’ theme. Data for each participant is shown, normalized to the last day of the pretreatment phase on day 11. Note that the 4-fibre snack produced the greatest reduction in HOMA-IR among the three different snacks tested (Supplementary Table5c). However, this reduction did not achieve statistical significance (P=0.078, linear-mixed effects model) after the short period of snack consumption in this study. d-f, LC-QTOF-MS analysis of a biomarker of orange fibre consumption present in faecal samples obtained from gnotobiotic mice and humans. Panel d compares levels of the m/z 274.1442 analyte in colonized and germ-free mice fed the unsupplemented, orange fibre-supplemented or pea fibre-supplemented HiSF-LoFV diet for 10 days. The analyte is only detectable when orange fibre is consumed and is not dependent upon on the human donor microbiome for its generation. Bars represent mean values ± s.d. for biological replicates (n=5 mice/group). Panels d and e compare levels of the analyte in faecal samples obtained from participants in human study 2 on days 25 and 49 when they were consuming the maximum dose of the 2-fibre (pea and inulin) and 4-fibre (pea fibre, inulin, orange fibre plus barley bran) snack food prototypes. The bar graph in panel e represents mean values ± s.d. for technical replicates (n=14 participants). The horizontal dashed line in panel f denotes a baseline value operationally defined as the highest level of detection of the analyte in participants consuming the 2-fibre snack food prototype lacking orange fibre.