Fig. 2: Validation of predictions. | Nature Communications

Fig. 2: Validation of predictions.

From: Ribosome profiling at isoform level reveals evolutionary conserved impacts of differential splicing on the proteome

Fig. 2

a We show the distribution of the relative abundance in high polysome (left panels) and monosome (right panels) fractions of human embryonic stem cells (ESC) for translated isoforms and for isoforms with RNA expression (TPM > 0.1) but predicted as not-translated. The plot shows the results for three different ORF lengths: 200–1000nt (high polysome Wilcoxon test p-value 7.2e–67, translated n = 18,056 and not-translated n=9397, and monosome Wilcoxon test p-value 2.3e–77, translated n = 15,023 and not-translated n =8277), 1001–10000nt (high polysome Wilcoxon test p-value 2.5e–242, translated n = 19,169 and not-translated n = 3813, and monosome Wilcoxon test p-value 1.1e–94, translated n = 16,002 and not-translated n = 3179), and longer than 10000nt (high polysome Wilcoxon test p-value 5.1e–10, translated n = 200 and not-translated n = 69, and monosome Wilcoxon test p-vaue 5e–06, translated n = 178 and not-translated n = 66). The results for neuronal cells are given in Supplementary Fig. 3. Box boundaries correspond to the first and the third quartiles, the median is indicated by a thick black line, top and bottom whiskers extend up to 1.5 times the interquartile range to the highest and smallest values, respectively, and outliers are indicated as black dots. b Cross-species conservation of protein isoforms. Protein isoforms from each 1-to-1 orthologous gene pair are compared, and candidate orthologous pairs are assigned using an optimization approach (see the “Methods” section). c For the set of ORFs encoding a human–mouse orthologous protein pair (orthologous) and for those encoding proteins without an orthologous pair in mouse (non-orthologous), we plot the percentages that are predicted to be translated (translated) and the ones with RNA expression (TPM > 0.1) but predicted as not-translated (not-translated). We show here the results for human glia (Fisher’s test p-value = 1.41e–140, orthologous n = 12,801 and non-orthologous n = 17,111) and glioma (Fisher’s test sp-value = 3.63e–85, orthologous n = 12,453 and non-orthologous n = 16,278), and for mouse glia (Fisher’s test p-value = 1.143e–130, orthologous n = 12,792 and non-orthologous n = 7356) and glioma (Fisher’s test p-value = 7.462e–53, orthologous n = 12,015 and non-orthologous n = 5905). Other mouse samples are shown in Supplementary Fig. 10e.

Back to article page