Clarify documentation on calling versus tumor fraction estimation

[lbeltrame]

Hello,

it is quite unclear if the parameters being set have an impact on copy number calling. We've been doing some benchmarking altering for example --txnE and its pair option, and while we can loosely understand the relationship they have with the calculation of tumor fraction, it is quite unclear how and if they affect copy number calling.

Is there any reference we can look up to see how it is actually being calculated and how it can be adjusted, if such a thing is possible?

[ilykos]

See if a reply by @yaotianran helps
broadinstitute#108

The formula he shows there is from Scalable whole-exome sequencing of cell-free DNA
reveals high concordance with metastatic tumors, Page 8

[lbeltrame]

I took a look, but my doubt remans: it's totally unclear how CN calls affect tumor fraction (if they do). In our experimentation, there seemed to be no relationship at all, on the surface.

[ilykos]

@lbeltrame Just curious, what values range did you probe?