Table 1.
Model-averaged estimates of the DFE parameters, for all species
| Species | Sd | b | Fraction of mutations −1 < Nes < 0 |
π 0/π4 | Model |
|---|---|---|---|---|---|
| Fagus sylvatica | −25,000 | 0.36 | 0.20 | 0.23 | − ɛanc |
| Quercus petraea | −9500 | 0.41 | 0.23 | 0.27 | +− ɛanc |
| Betula pendula | −190 | 1.59 | 0.17 | 0.20 | +− |
| Populus nigra | −571 | 0.39 | 0.24 | 0.26 | +− |
| Picea abies | −47,000 | 0.097 | 0.30 | 0.35 | − ɛanc |
| Pinus pinaster | −64 | 0.73 | 0.22 | 0.28 | +− |
Sd: the mean scaled strength of deleterious selection acting on new mutations rounded to two S. F., that is, the scale parameter of the gamma-shaped deleterious DFE; b: the shape parameter of the gamma-shaped deleterious DFE, which is inversely related to the coefficient of variation in the fitness effects of new deleterious mutations; the inferred fraction of mutations with fitness effects between −1 and 0, that is, the nearly neutral fraction of slightly deleterious mutations; and π0/π4. DFE parameters shown are model-averaged, such that estimates are weighted by model AIC. The best model, as ascertained using likelihood ratio tests, is indicated in the last column; whether fitting a deleterious-only DFE (−) or a full DFE including beneficial mutations (+−), and whether including the rate of error in the inference of the ancestral state improves the model fit (ɛanc).