The process is repeated where for each repetition a random proportion of the data are used to train the model (default value is 0.75) while the remainder is used for testing the models.
Stepwise forward multiple regression method was utilized using the advanced variable selection and model building wizard for development of the 3D-QSAR model with default values.
We employed the WAG+I+G model with four rate categories for proteins (this was the optimal model selected with ProtTest 1.3 [ 46]), and the GTR+I+G model with four rate categories for nucleotides, with normally distributed priors for the parameters in models of molecular evolution, a Yule process tree model, and default values for all other settings.
The pedigree suggested an autosomal dominant mode of inheritance, and thus we assumed an autosomal dominant model with default values of full penetrance (f0, f1, f2) = (0, 1, 1) and disease allele frequency = 1e-05.
Maximum likelihood analyses were conducted on the combined data matrix with RAxML GUI [ 72- 74], using four partitions corresponding to the cox1, rrnL + trnL + nad1, SSU and LSU fragments with a GTR + I + G evolutionary model and default values for other parameters of the search [ 73].
Burn-in and sample size of 10, thinning 50, F-model and default values of pilot runs, priors and incremental values were used, reporting mode estimates from the run with lowest total deviance and acceptance rate of 25 - 45% after 10 replicates to ensure convergence [ 44, 100, 147].
The constants used in the original model were only used in this model as the default values for each of the model's parameters which in this model are now be regulated and controlled by mechanisms such as the TGF-β1 related internal decisions which include rules for the injury induced emergent behaviours.
The Appendix provides detailed information on the model's features, default values, assumptions and model runs.
