help in describing the input

[kzkedzierska]

Hi!

I am trying to use the tool but I run into a few problems and have questions regarding the inputs:

• I am not sure what you mean in the clusters data frame as VAF proportion - is that cancer cell fraction (CCF) of the cluster? Or is that VAF corrected for purity (if so what would be the recommended way of calculating it?)?
• I transformed the VCF to have the t_alt_count and t_ref_count but I do not have the AD/DP columns. Is that ok?
• Would it also be possible to see the vcf file that is being used here, the GRanges object I get when I apply the readVcf is different (my rowRanges have additional column paramRangeID) and it would be good to understand why.
• can the bb GRanges object be a subclone file from Battenberg? If so, is the clonal_frequency just a frac1_A or should that be corrected by purity?

My questions are sparked by the fact that despite my best efforts the function mutationTime produces only NAs and plots are empty (except for copy number). Which might hint that my vcf is different than what the functions expect.

I will be very grateful for suggestions and help!

[kzkedzierska]

I found some hints to the first question in #6, "All subclonal clusters have proportion < purity" but still not sure if that is CCF * purity or something else. I.e. what is the meaning of the VAF proportion?

I also found that #5 talks about the clonal_frequency, and if I understand correctly clonal_frequency is the fraction multiplied by purity.

[lincj1994]

Hi. I'm wondering if you have solved the question about the VCF input. Could you provide an input example of VCF?

[Salaheddinekacimi]

@lincj1994 @kzkedzierska I am wondering if you have solved the VCF input question. I am getting this error
mt <- mutationTime(vcf = vcf, cn = bb)
Error in geno(vcf)$AD[, 1, 2] : incorrect number of dimensions when I load my VCF and run mutationtime

[Salaheddinekacimi]

these are my dimensions

dim(geno(vcf)$AD)
[1] 233 1

[Salaheddinekacimi]

I fixed the error by making my vcf look similar to the one in the example repo

[tc86]

@Salaheddinekacimi Hi! Would you kindly elaborate briefly on how you formatted your vcf file similar to the example? I'm getting the same dimension error as you and was wondering how to format the vcf

[tc86]

Posting this if anyone else is facing the same issue. I think the dimension error occurs from the vcf format. If the vcf format is a collapsed vcf then the dimension error occurs when used as input. The expected vcf format is an expanded vcf file.
I was able to solve this error and run the tool smoothly on my vcf file by simply running vcf <- expand(vcf) function on the collapsed vcf file in R.

[avrajit]

Hi [kzkedzierska],
I also have simialr problem which you described here. My vcf file gnerated by using GATK somatic paired calling tool Mutect2. I processed my VCF files to meet the input parameter , however that dose not working. To process my somatic vcf

1. I removed adjacent normal sample
2. Kept only SNPs as I found MutationTime have issues to read AD information for complex variants
3. I keep only standard chromosomes (no contigs or ChrM)
4. For copy number data I have PURPLE somatic copy number output for same sample. Using this data for GRange object.
   Is this the correct way to make the input file ? please help