Fig. 2

Relationship between synonymous bicodons, protein abundances (PA) and pause propensity. a Frequency distributions associated with the bicodons that encode the amino acid pair SK, computed using sequences from the low PA sample (red bars), and from the high PA sample (orange bars). Some bicodons are more frequent within lowly abundant proteins (such as TCCAAG), some within highly abundant ones (such as AGCAAG), and other bicodons have similar frequencies in both groups of proteins (such as TCAAAA, TCAAAG and TCGAAG). b Raster plot of the p-values versus the pause propensity for all bicodons. To improve the visualization of this correlation we plot −S log10[p-value] instead of p-value, where S takes the values +1 or −1 when the bicodon has preference for sequences with low or with high PA, respectively. c Raster plot of the residual scores χ ² versus the pause propensity and p-values for all bicodons. Small pause propensity values (orange zone) are related to bicodons with high PA, whereas large pause propensity values (red zone) are related to bicodons with low PA. Green lines represent the nine synonymous bicodon variants for the SK amino acid pair that involve a large change in pause propensity (Δ π≥0.754)