Thank you for your interest in HATCHet and I would be happy to help with your questions related to SNV analysis.

Using the resulting segmentation, you can get the allele- and clone-specific copy numbers for an SNV POS by retrieving the copy numbers of a segment in the same chromosome with START and END such that START <= POS < END. Note that the results are reported in a data frame and the results for the same segment in different samples are reported in different rows. Also, for every clone, the results report the corresponding allele-specific copy numbers (separated by |) and the proportion of the related clone; note that a proportion of 0 indicates that the clone and the corresponding copy numbers are not present in that sample.

Also, nearly all existing methods for computing CCFs and/or clustering SNVs do not use clone-specific information about copy numbers in input, but only the total proportion of the different allele-specific copy numbers. Thus, considering the following example for the copy numbers of a segment with a specific SNV:

Then you can conclude that the SNV is in a segment with copy numbers 1|1 for 70% of cells (including normal) and 2|0 in the remaining 30% of cells.

Moreover, we would like to inform you that we are about to release soon a new method called DeCiFer which clusters SNVs and infers related CCFs, and it should be easy to integrate with HATCHet; differently then other methods, DeCiFer can properly deal with subclonal CNAs.

Please do let us know if you have any further question.

Thanks for your response and looking forward to trying out DeCiFer!

I was looking into your script: https://github.com/raphael-group/hatchet-paper/blob/master/analysis/explainMutationsCCF.py
It worked for me and I was able to calculate CCF for each of my mutations and most were labelled as "Explained" which is promising. Ideally I would like to now cluster the mutations based on their CCF using something like SciClone to then build a clonal phylogenetic tree. To use SciClone though I would first need to convert my CCF values into adjusted VAF (CCF/2) and I am not sure if this is the most appropriate way to go about it. Was just looking to get your insight into this in case you have come across this type of question before? Otherwise feel free to close this issue. Thanks again!

If you would like to simply cluster the SNVs based on the CCFs computed by the explained-mutation analysis, I would recommend to consider using PyClone in the alternative mode, i.e. the mode in which the user can directly provide CCFs and let PyClone cluster those.

Unfortunately, I do not think that obtaining the adjusted VAF as CCF/2 is an appropriate approach with CNAs since the adjustment would depends on the number of mutated copies (copies of a segment with the SNV can be different than 1 for example when the SNV is occurring on an allele which is subsequently amplified).

I am closing the issue for now but please feel free to let us know if you have any other question or you would like to have further details about the above.