knausb / vcfR

Tools to work with variant call format files
248 stars 54 forks source link

write.vcf() not filtering variants #203

Closed mattbareno closed 2 years ago

mattbareno commented 2 years ago

Hello, i'm attempting to filter a vcf for quality, depth, maf and so on... I was following the steps laid out here https://pubmed.ncbi.nlm.nih.gov/27401132/ I followed the tutorial for a proof of concept with the following code in R: `library(vcfR)

library(pinfsc50)

vcf_file <- system.file('extdata','pinf_sc50.vcf.gz',package = 'pinfsc50')

vcf <- read.vcfR(vcf_file)

chrom <- create.chromR(name='Supercontig',vcf=vcf)

chrom <- masker(chrom, min_QUAL=0, min_DP=350,max_DP=650, min_MQ=59.5,max_MQ=60.5)

chrom <- proc.chromR(chrom)

chromoqc(chrom, dp.alpha=20)

chrom@vcf <- chrom@vcf[chrom@var.info$mask,]

write.vcf(x = chrom, file = "filtered.vcf.gz", mask = FALSE, APPEND = FALSE) `

And successfully generated a filtered vcf. However, when i load that filtered vcf into R again and analyse, it still has the same amount of variants. When i look at the things i supposedly filtered, they still remain (ie, there are variants with Mappingq quality < 59.5.

Is there some step i'm missing here or some other way to do this? Thank you very much!

knausb commented 2 years ago

Hi, I think you need to tell write.vcf() to use the mask.

write.vcf(x = chrom, file = "filtered.vcf.gz", mask = TRUE, APPEND = FALSE)

Does that give you the desired behavior? Brian

mattbareno commented 2 years ago

This resolved the issue, thank you very much