Table 2.
N | blog3D | p-value | DE95%− | DE95%+ | |
---|---|---|---|---|---|
Region | |||||
NW_REF | 25 | 0.0071 | <0.001 | −0.0032 | 0.0034 |
NE_REF | 16 | −0.0011 | 0.707 | −0.0047 | 0.0056 |
S_REF | 7 | −0.0027 | 0.627 | −0.0125 | 0.0126 |
NE_EXP | 12 | −0.0006 | 0.556 | −0.0075 | 0.0081 |
S_EXP | 11 | 0.0076 | 0.157 | −0.0133 | 0.0138 |
Historical group | |||||
REF | 48 | 0.0023 | 0.003 | −0.0014 | 0.0015 |
EXP | 17 | 0.0016 | 0.275 | −0.0051 | 0.0051 |
PLANTED | 6 | 0.0016 | 0.278 | −0.0065 | 0.0056 |
All pop | 71 | 0.0032 | <0.001 | −0.0015 | 0.0016 |
Pairwise FST values were regressed on the logarithm of three-dimensional (3D) distances between plots. For each region/group we give the number of plots N, the slope of the regression (blog3D), its p-values and the 95% distribution envelope under the null hypothesis that there is no SGS (DE95%− and DE95%+).