Fig. 1

Ribosome profiling around-the-clock in mouse liver and kidney. a Overview of the experimental design. Livers and kidneys for ribosome profiling were collected every 2 h for two daily cycles. Each timepoint sample was a pool of organs from two animals. Mice were kept under 12 h:12 h light-dark conditions, with Zeitgeber times ZT00 corresponding to lights-on and ZT12 to lights-off. b Read distribution to transcript features. RPF-seq (left; kidney in orange, liver in green) and RNA sequencing (RNA-seq) (right; blue and red for kidney and liver, respectively) compared with a distribution expected from the relative feature sizes (grey; the distributions based on feature sizes were highly similar for both organs, thus only that for kidney is shown). Note that RPF-seq footprints were enriched on the CDS and depleted from UTRs, whereas RNA-seq reads distributed more homogeneously along transcripts, according to feature size. Of note, the higher level of 3′ UTR footprints in kidney resulted mainly from differences in the efficiency with which stop codon footprints were captured, as described in (c). c Predicted position of the ribosome’s aminoacyl tRNA-site (A-site) of reads relative to the CDS start and stop codons. Read density at each position was averaged across single protein coding isoform genes (i.e., genes with one main expressed transcript isoform) that had an average RPF RPKM > 5, a CDS > 400 nt in length and were expressed in both organs (n = 3037 genes). This analysis revealed the trinucleotide periodicity of RPF-seq (but not RNA-seq) reads in both organs. Inset: frame analysis of CDS reads showed preference for the annotated reading frame (frame 1, the same frame as the start codon) in RPF but not in RNA reads. Violin plots extend to the range of the data (n = 3694 genes for liver, n = 4602 genes for kidney). A separate analysis of the higher level of stop codon footprints in kidney, that also led to the differences in 3′ UTR reads in B, can be found in Additional file 1: Figure S2A, B. d Principal component analysis (PCA) of kidney and liver RPF-seq and RNA-seq datasets, using the 4000 most variable genes. The first two components reflected the variability coming from organ (PC1, 64.21%) and from RPF/RNA origin of datasets (PC2, 28.35%). e PC3 vs. PC5 (together 12.5% of variation) resolved the factor time within each dataset, leading to a representation that resembled the face of a clock. Each dot represents one sample, timepoint replicates are joined by a line and timepoints within each dataset are sequentially coloured. The circular arrangement was larger for liver than kidney, suggesting a higher contribution of hepatic rhythmic genes to overall variability. Additional file 1: Figure S4 shows the scree plot for the ten first components