Each row corresponds to a simulated dataset and contains the parameter values used to simulate the dataset and summary statistics computed on the simulated dataset.
GATE was used to simulate 4DPET datasets corresponding to the acquired 4DMR images.
In bootstrapping, the initially estimated Rasch parameters are used to simulate further datasets that fit the Rasch model.
To check the frequency at which the true topology was reconstructed, we compared, for each dataset, the maximum a posteriori (MAP) topology of each segment against their respective true trees (i.e., the trees initially used to simulate the datasets).
For each pair, we calculated the test statistic from the original data, W. The original data were then used to estimate parameters under M1a (neutral), and the estimates were used to simulate 1000 datasets under M1a.
The QMSim software [ 6] was used to simulate all datasets.
The program msms (Ewing and Hermisson, 2010) was used to simulate population datasets under a neutral model (Neu, Neu_1 and Neu_2 after fission) and a single locus selection model (noSel, Sel, Sel_1 and Sel_2 after fission), respectively.
The ground truth of features used to simulate the dataset is indicated by solid (relevant) and dashed (irrelevant) lines in Fig. 4 a.
The 10 scenarios were defined by combinations of the number of genotypes (small = 177, large =698), marker effect variances [(small = (0.2019, 0.2019/10), large = (0.005892, 0.005892/10)], and multiples (5, 8, 10 for small datasets; 5, 10 for large datasets) of the SDs of the errors (small = 48.6728, large = 53.8715) used to simulate the phenotypic datasets.
Specifically we assume as true the model used to simulate the synthetic datasets and we check if IGpe* can be seen as a random realization from the IGpe distribution obtained for the model under testing.
