How does Picard's CollectVariantCallingMetrics calculates novel Ti/Tv per sample and as summary?
I am having a hard time making sense of the novel Ti/Tv values produced by Picard's
CollectVariantCallingMetrics when comparing the summary and the detailed reports.
The summary reports a novel
1.700175 (which is the same value found in tranche 90 of the
VariantRecalibrator's tranches plot). However, sample-wise, novel
Ti/Tv ranges from
I do not see how
1.7 summarizes a range of numbers from
1.9. What am I missing?
Thanks in advance and happy new year!