In silico dilution of the sWGS data. (A) Example of dilution on chromosome 17 of sample ALG201322 (see supplemental Figure 10 for full legend and dilution settings). (B-C) Performances of the diluted data sets enforcing various minimal segment sizes. Top panel represents the number of segments in the original data set longer than minSize, sorted by copy number state (gain or loss). The second panel translates this coverage of the whole data set into the proportion of altered bins included in a segment longer than minSize. The third and fourth panels describe, respectively, the precision and recall in both diluted datasets, with 95% confidence intervals plotted in gray and thresholds of interest in green. The scales at the bottom provide 3 representations of minSize, in Mbp and as the number of consecutives bins for the 2 different diluted datasets. (B) Analysis focused on individual segments. (C) Analysis focused on regions, merging consecutive segments with different copy number states.