Genotypic descriptors as empirical functions of the rate
of clonality
To assess the relation between c and the genotypic descriptors,
we explored the mean results of simulations as a function of c .
Depending on the shape of the curves obtained with simulated data, we
tested the fit with basic functions (for example, simple sigmoids and
exponentially decreasing distributions) as well as with sigmoid and
parabolic curves. To assess the accuracy of our empirically inferred
formula to describe the relationships, we computed the mean absolute
error (MAE) and the root-mean-square deviation (RMSD) between
pseudo-observed simulated values and fitted formulae. These two
deviation measures aggregate the magnitudes of the errors of predictions
into a single measure of predictive accuracy. This measure represents
the mean deviation of predicted values with respect to the observed
values and has the advantage of sharing the same units as the model
variable under evaluation. Lower deviation measures indicate higher
accuracy of an analytical formula in the prediction of data. These
measures must be interpreted at the same scale as the mean value of the
studied parameter (PiƱeiro, Perelman, Guerschman, & Paruelo, 2008).
\begin{equation}
MAE=\frac{1}{n}.\sum_{1}^{n}\left|y_{s}-y_{f}\right|\nonumber \\
\end{equation}\begin{equation}
RMSD=\sqrt{\frac{1}{n}.\sum_{1}^{n}\left(y_{s}-y_{f}\right)^{2}}\nonumber \\
\end{equation}where n is the number of pseudo-observed simulations per
scenario, ys is the simulated value of the
genotypic descriptor under consideration, and yfis the calculated value of the genotypic descriptor using the fitted
formula.