Extended Data Fig. 8: Refinement of the dN/dS calculation to remove skewed values.
From: The role of metabolism in shaping enzyme structures over 400 million years
a) dN (left) and dS (right) versus mean CR. Red dots denote a dS > 3, the solid line indicates the best linear fit and the dashed line denotes the axis median. b) dN/dS versus mean CR. Colouring according to a). The solid line indicates the best linear fit for the blue highlighted orthogroups and the dashed lines denote the axis median for the two groups. c + d) Significantly enriched pathways and GO slim terms of c) the blue (dS ≤ 3, n = 272) and red (dS > 3, n = 94) highlighted orthogroups with respect to the full set and d) orthogroups containing enzymes with the 25% highest dN/dS values. Dot size indicates the number of genes associated with the term or pathway in our selection, and colour indicates adjusted p-value. In addition the AUC is shown. e + f) Enrichment factor for e) enzyme classes and f) the GO term “lipid-metabolic process” according to dN/dS. The dashed line indicates random sampling. g) Probability of a residue being under neutral drift is inversely correlated with CR. h) dN/dS versus the mean CR for orthogroups with dS < 3. Orthogroups that contain proteins of the electron transport chain as well as Pox1p and Tdh1/2/3p are highlighted. The solid line indicates the best linear fit and the dashed line denotes the axis median.
