Table 1.

Studies aiming at measuring the relative rate of evolution of phosphorylation sites (comparison of phosphorylated and non-phosphorylated sites).

study	dataset	comparative method	conclusion
Gnad et al. [15]	large-scale phosphoproteomic experiments on human HeLa cells	examine whether sites are conserved across organisms (70 organisms from E. coli to mouse)	(i) phosphosites are more conserved than non-phosphorylated sites: 88% versus 81% for pS and S when compared with mouse and considering loop regions only
Malik et al. [28]	phosphoproteomics data derived from the human mitotic spindle apparatus	derived a normalized, relative score from the number of vertebrate species with a conserved residue (among six total species)	(i) phosphosites are more conserved than non-phosphorylated sites: 92% of pS/pT/pY conserved in mouse compared with 84% for S/T/Y (ii) phosphorylation sites with experimentally verified biological functions are significantly more conserved. Median conservation score: S/T/Y: −0.0893; pS/pT/pY: 0.1144; functional pS/pT/pY: 0.183
Boekhorst et al. [29]	comparisons of phosphosites for six eukaryote species (human, mouse, fly, yeast, plant, zebrafish)	all-against-all Smith–Waterman searches to measure overlap	(i) overlap of orthologous phosphosites among the datasets greater than expected by chance alone(ii) conserved phosphosites are slightly more likely to be found in globular domains (iii) three-way comparisons show conservation from 1.6 to 7.4 fold over the expectation
Landry et al. [25]	comparisons of rate of evolution of human phosphosites among vertebrates and yeast phosphosites among yeasts	measure relative rate of evolution across the proteome using concatenated multiple species alignment	(i) phosphorylation sites are significantly older that non-phosphorylated residues in both the yeast and human lineages (ii) phosphorylation sites with experimentally verified biological functions are significantly more conserved
Tan et al. [30]	human, fly, worm and yeast phosphorylation sites from large-scale and small scale experiments.	compare the conservation of phosphorylation in three model organisms, fly, worm and yeast and conservation of residues in other species.	(i) phosphorylation sites are more conserved than non-phosphorylated ones
Ba & Moses [31]	analysis of high-confidence annotated sites in budding yeast	compare rates of substitution at phosphosites and flanking sites by comparing with four closely related species	(i) overall, significant reduction in amino acid substitution at phosphosites compared with flanking regions (0.07 versus 0.12) (ii) sites grouped by protein kinases show that some sites evolve slower, though not all
Chen et al. [32]	human and mouse experimentally verified phosphosites.	measure the rate of evolution of phosphosites by inferring the ancestral phosphoproteome of human and mouse. Ancestral phosphorylation states are predicted	(i) phosphorylated S/T evolve slower than non-phosphorylated sites in both ordered and disordered regions (overall S: 0.097 versus 0.127; T: 0.121 versus 0.139) (ii) phosphorylated Tyr evolve at similar rate as non-phosphorylated Tyr
Boulais et al. [23]	compare the phosphorylation of Mouse, Drosophila and Dictyostelium phagosome proteins	compare the number of orthologous sites phosphorylated in two species with what is expected by chance	(i) phosphosites are more likely to occur at homologous positions between species
Wang et al. [33]	human phosphosites from large-scale and small-scale (annotated) studies	compare human phosphosites to orthologous sites in a second species	(i) phosphosites evolve slower than their flanking sequences for some functional categories and not others (ii) phosphosites of known function evolve slower than those of unknown function
Gray et al. [34]	human phosphosites, including annotated ones	estimate the absolute rates of evolution by mapping sequence differences among 44 species and dividing by the total time; P-sites of protein are compared with all equivalent sites of the same protein	(i) evidence for higher purifying selection in about 70% of phosphorylated sites; 30% of phosphorylation sites evolve at the same rate or faster than equivalent residues on the same proteins
Park et al. [35]	analysis of mouse phosphorylation sites in nine mouse tissues (36 000 sites)	measure conservation by comparison with rat, human and chicken separately	(i) found no significant difference in the rate of evolution of phosphorylated and non-phosphorylated S, T and Y