Error in seqVCF2GDS(vcf.files, gds.file, verbose = FALSE)

[Bulouere]

Hi,
I get an error when I'm trying to convert from a vcf file to a gds file using the SeqArray package. Here is my code and the error message.

gds.file <- tempfile()
seqVCF2GDS(vcf.files, gds.file, verbose=FALSE)
Error in seqVCF2GDS(vcf.files, gds.file, verbose = FALSE) :
FORMAT ID 'QSS' (Number=A) should have 1 value(s) but receives 2, please consider revising it to 'Number=.'.
FILE: C:\Users\bwodu1\OneDrive - Johns Hopkins\Yegna Lab\u01\test_vcf\55488_60663_Cancer_vs_55488_60664_Benign_mutect2_passed.vcf
LINE: 66, COLUMN: 10, 0/1:23,14:0.381

[zhengxwen]

As it said please consider revising it to 'Number=.', please revise the VCF file.

[amstilp]

I'm running into a similar error with structural variant VCFs from the All Of Us project:

Error in .DynamicForkCall(njobs, njobs, .fun = function(.jobidx, ...) { : 
  Error in seqVCF2GDS(vcf.fn, ptmpfn[i], header = oldheader, storage.option = storage.option,  : 
  INFO ID 'SOURCE' should have 1 value(s) but receives 2, please consider revising it to 'Number=.'.
FILE: /mnt/data/input/gs/fc-aou-datasets-controlled/v8/wgs/short_read/structural_variants/vcf/full/AoU_srWGS_SV.v8.chr2.vcf.gz
LINE: 133930, COLUMN: 8,  ..redacted..

Since I didn't create the VCFs, it's not as easy to revise them and I'd prefer to not pay additional cloud storage costs just to be able to convert them to GDS. Are there other ways I could fix this error?

[zhengxwen]

Please use

hr <- seqVCF_Header("your.vcf")
# hr$info is a data.frame
# revise its 'Number' to be '.' according 'ID=SOURCE'

# then pass the revised hr to seqVCF2GDS
seqVCF2GDS(, header=hr)