I have been using SciClone to check clonality changes between pre- and post- treatment tumor. It is an impressive tools and the figure generated seems reasonable and easy to interpret.
However, SciClone paper claims that it only makes clonality inference from variants' VAF within CN neutral region. Then why do I see clusters on non-CN neutral region (1 copy and 3 copies)? How are they inferred? Is it that SciClone only uses VAF within CN neutral region to establish the baseline and later infers clusters on non-CN neutral region after adjusting the VAF based on the ploidy?
Then why do I see clusters on non-CN neutral region (1 copy and 3 copies)?
They're plotted by default in a 1d view (in the appropriate row), but not utilized for clustering. I'd need more information on exactly what your inputs and outputs are to offer more help.
My input is pretty standard. A variant list with VAF and a CNV list generated by VarScan. The output is 1D view and 2D view.
The 1 copy and 3 copy plot in the 1D view is exactly what I am talking ab out. So you mean the clusters on 1 / 3 copy plot are just plotting based on the variant and CNV input. No actually clustering is performed on them, right? So just need to focus on the 2 copy clustering for clonality changes.
Yep, you've got it. Just look at the 2 copy plot for clustering. There's also an option to show just the CN2 regions in the plotting command.
Log in to answer this question.
My input is pretty standard. A variant list with VAF and a CNV list generated by VarScan. The output is 1D view and 2D view.
The 1 copy and 3 copy plot in the 1D view is exactly what I am talking ab out. So you mean the clusters on 1 / 3 copy plot are just plotting based on the variant and CNV input. No actually clustering is performed on them, right? So just need to focus on the 2 copy clustering for clonality changes.