Fig. 6: Prevalence of OTUs in 43 rats according to dietary iron (Ferric citrate vs. haemoglobin) and calcium (control vs. mineral vs. dairy).
a The heatmap represents the abundance of the 486 OTUs detected in each rat (1 rat per column) and clusters of OTUs boxed in black correspond to OTUs whose changes induced by the addition of haemoglobin vs. ferric citrate are normalised by the addition of calcium, (b) regardless of its origin (Eisenbergiella spp., multi-affiliated Lachnospiraceae among others), (c) preferentially when it is from mineral origin (Bifidobacterium pseudolongum, Lachnoclostridium among others) or (d) from dairy origin (unknown Peptococcaceae, Lactobacillus intestinalis among others). For (b, c, d), abundances were obtained once OTUs were agglomerated at the species or genus levels. Individual normalised values were plotted within the box and whisker with “+” as means according to dietary iron (Ferric citrate vs. Haemoglobin) and calcium (Control vs. Mineral vs. Dairy) contents. For statistical analysis, two-way ANOVA, followed by Holm-Šídák’s multiple comparisons tests were performed. *padj < 0.05, **padj < 0.01, ****padj < 0.0001 according to dietary iron form. Detailed F-statistic and p values are provided as Supporting Information Table S1.
