how to calculate BAF and LRR from VCF or BCF files?

Question

[xiaoguang]

Hi,

Do anyone know how to calculate BAF(B-allele frequency) and log R ratio from SNP calling pipeline( bcftoools mpileup & call) ?

I read some papers and I guess BAF can be calculated by (AD[2]/DP) and LRR can be calculated by [log2(cancer DP /normal DP)] , are that right?

I also want to know the level of allele specific CNV using ASCAT package. now I have paired samples (normal WGS and cancer WGS data), ASCAT need BAF and LRR from tumour and gemerline, but how to differ the tumour and gemerline variation?

[pmc.sa]

Hi @xiaoguang! Did you figure out if this is a valid way to calculate BAF and LRR?

Answer 1

[Matt]

@xiaoguang

hi, have you got the answer?

I want to calculate the tumor purity with ASCAT.

I have tumor.bam and normal.bam from a WES pipeline, but I don't how to generate the input files like Tumor_LogR.txt, Tumor_BAF.txt, GC_file.txt, RT_file.txt, from bam or vcf files.

Is there any tool could do this?

[xiaoguang]

Hi, The ASCAT new version has introduced a new function called ascat.prepareHTS, which can be used to prepare all files that are required for predicting purity.

[Nick]

I'm having the same problem. ascat.prepareHTS seems like the solution, except when I run it, it seems to be looking for those files that I want it to create, so it fails. Is it something I'm doing wrong? Thanks in advance for the help.

---

ascat.prepareHTS( tumourseqfile = "/data/tumor.mapped.sort.bam", normalseqfile = "/data/normal.mapped.sort.bam", tumourname = "Tumour_name", normalname = "Normal_name", allelecounter_exe = "/usr/bin/bin/alleleCounter", alleles.prefix = "/data/ascat_reference/G1000_allelesAll_hg38/G1000_alleles_hg38", loci.prefix = "/data/ascat_reference/G1000_lociAll_hg38/G1000_lociAll_hg38", gender = "XX", genomeVersion = "hg38", chrom_names = c('chr1', 'chr2', 'chr3', 'chr4', 'chr5', 'chr6', 'chr7', 'chr8', 'chr9', 'chr10', 'chr11', 'chr12', 'chr13', 'chr14', 'chr15', 'chr16', 'chr17', 'chr18', 'chr19', 'chr20', 'chr21', 'chr22', 'chrX'), nthreads = 2, tumourLogR_file = "/data/ascat_out/Tumor_LogR.txt", tumourBAF_file = "/data/ascat_out/Tumor_BAF.txt", normalLogR_file = "/data/ascat_out/Germline_LogR.txt", normalBAF_file = "/data/ascat_out/Germline_BAF.txt", skip_allele_counting_tumour = 0, skip_allele_counting_normal = 0)

Error in readAlleleCountFiles(tumourAlleleCountsFile.prefix, ".txt", chrom_names, : length(files) > 0 is not TRUE

[xav235]

Most likely, ascat is not able to find the allele file from the prefix you provided. From documentation alleles.prefix: Prefix path to the allele data (e.g. "G1000_alleles_chr").

https://github.com/VanLoo-lab/ascat/blob/master/ASCAT/R/ascat.prepareHTS.R

In readAllelesFiles function

files=paste0(prefix, chrom_names, suffix)

If your files are like: /data/ascat_reference/G1000_allelesAll_hg38/G1000_alleles_hg38_chr1.txt Your prefix should be something like: alleles.prefix = "/data/ascat_reference/G1000_allelesAll_hg38/G1000_alleles_hg38_"

The other reason could be that the read counts mapping to the regions in all the files are zero.