Fig. 3.

Comparative assessment of breakpoints inference and CN calling for simulated data tests results. A–D Distribution of false positive rate (y-axis) as a function of cell count (x-axis) for all simulation scenarios: smaller tree size (first and second column) versus larger tree size (third and fourth column), low noise (first and third column) versus high noise (second and fourth column). E–H Distribution of false negative rat (y-axis) as a function of cell count (x-axis) for all scenarios. I–L Distribution of symmetric distance score between inferred and real breakpoints (y-axis) as a function of cell count (y-axis) for all scenarios. M–P Distribution of CN-RMSE (y-axis) depending on the cell number (x-axis) for all scenarios. Optimal values of the presented quality measures are marked with a green dashed line in each plot