Fig. 5

Detection ability for identifying adaptive mutations. A Each panel in the \(4\times 4\) array corresponds to one simulation (Methods section). Each point represents an adaptive mutation that occurred and established in the simulation. Points are colored according to whether they were identified by both methods (blue), only by FitMut2 (pink), or by neither (grey) (no point that only by FitMut1); their counts are shown in the right bottom corner of each panel. \(n_P\) represents the total number of established mutations for a given DFE. The three lines indicate the mean fitness (solid, \(s = \bar{s}(t)\)), the boundary above which mutations must occur in order to establish (dot-dashed, \(s = \bar{s}(t+\frac{1}{s})\)) and the boundary to be observed (short-dashed, \(s = \bar{s}(t+\frac{1}{s} + \frac{1}{s}\ln \left( \frac{s \bar{n}_0}{c}\right) )\)). The 5th column shows \(\mu (s)\) and the prior \(\tilde{\mu }(s)\) for each row. B Direct comparison of the detection ability between both algorithms