I'm trying CNVkit on FFPE targeted panel NGS data with 300 genes. I'm in the process of making a reference.cnn from 50 tumor samples.

I read the doc quickstart (https://cnvkit.readthedocs.io/en/stable/quickstart.html) and it said "Inspect the coverages of all samples with the metrics command, eliminating any poor-quality samples and choosing a larger or smaller antitarget bin size if necessary". After running 'metrics' I got these results:

sample                              segments  stdev     mad       iqr       bivar
TM24-0743_tumor.targetcoverage.cnn  -         0.5223    0.382844  0.519745  0.418463
TM24-0748_tumor.targetcoverage.cnn  -         0.574451  0.371302  0.50387   0.421322
TM24-0755_tumor.targetcoverage.cnn  -         0.436836  0.284481  0.3854    0.326914
...
sample                                  segments  stdev    mad       iqr       bivar
TM24-0743_tumor.antitargetcoverage.cnn  -         4.34981  0.422786  0.571706  0.391274
TM24-0748_tumor.antitargetcoverage.cnn  -         3.89586  0.731915  0.987979  0.680454
TM24-0755_tumor.antitargetcoverage.cnn  -         4.00702  0.642974  0.869162  0.608133
...

My question is:

• what is a normal range for these metrics?
• my antitargetcoverage metrics look like they are more variable than targetcoverage - is this concerning? Should I try "choosing a larger or smaller antitarget bin size"?

Thanks in advance