Table 2

Results of PAML analyses for the vertebrate dataset
Data set 2ΔlnL df P-value Proportion of sites with ω > 1 (average ω for these sites) Positively selected codonsa
All vertebrates (n= 77)
 M1a vs. M2a 15.22 2 P= 0.0005 0.017 (ω = 2.0)
 M7 vs. M8 34.89 2 P< 0.0001 0.050 (ω = 1.3) 97**, 98*, 104*, 286*
 M0 vs. free ratios 510.67 142 P< 0.0001
Mammals (n= 55)
 M1a vs. M2a 31.27 2 P< 0.0001 0.027 (ω = 2.3)
 M7 vs. M8 47.39 2 P< 0.0001 0.066 (ω = 1.6) 97**, 98*, 104*, 214*, 286**
 M0 vs. free ratios 265.71 100 P< 0.0001
Primates (n= 17)
 M1a vs. M2a 0 2 P = 1.0 none
 M7 vs. M8 0 2 P = 1.0 none
 M0 vs. free ratios 36.45 28 P = 0.13

n, number of sequences; 2ΔlnL, twice the log-likelihood difference of the models compared; df, degrees of freedom; P-value, level of significance (P-values < 0.05 are shown in bold).

a Sites pinpointed to be under positive selection under the selection model with the highest likelihood (M8 for the vertebrate and mammalian datasets) by Bayes Empirical Bayes (BEB) analysis. Only sites with posterior probability greater than 0.95 (*) or 0.99 (**) are shown. Sites are numbered according to the full human coding sequence.

Sabbagh et al.

Sabbagh et al. BMC Evolutionary Biology 2013 13:62   doi:10.1186/1471-2148-13-62

Open Data