Fig. 5: m⁶A modifications in the liver.

a Positions of detected m⁶A peaks on all methylated transcripts in liver were determined and visualized using the GUITAR package. b Motif enrichment in m⁶A modification was determined by calculating total occurrence of motifs in m⁶A peaks on the 5′UTR, CDS and 3′UTR of in the liver. Consensus motifs for m⁶A (RRACH) and m⁶Am (NBCAN) are highlighted in red. c Multidimensional scaling (MDS) plot of the peak log counts-per-million IP data of all differentially methylated peaks showing the positions of the samples in the space spanned by the first and second MDS dimensions. Samples are colored with respect to condition. CONV (n = 10), conventionally raised mouse (black); GF (n = 10), germ-free mouse (cyan); Am (n = 10), A.muciniphila- monocolonized mice (green), Lp (n = 3), L.plantarum-monocolonized mice (orange). d Heat map of the peak log counts-per-million IP data based on all differentially methylated peaks, hierarchical clustering was performed using euclidean distance and ward.D2 linkage. e m⁶A-peaks found to be differentially methylated compared to differential expression of transcripts in indicated mouse models. Differentially methylated peaks that are also differentially expressed on transcript level are in red, differentially methylated peaks that are unchanged on transcript level, are in blue. Methylation peaks that were not significantly changed are shown in gray. p-values were estimated from moderated t-statistics with empirical Bayes moderation using limma package, followed by Benjamini-Hochberg correction. Cut-offs for differential expression are log fold change (FC) −1 to 1 and Benjamini–Hochberg- corrected p-values < 0.05.