I got vcf files from Platypus variant caller. When I try to import them into R and the VariantAnnotation format VRanges I get the error message:

test = readVcf(path)
as(object = test,Class = "VRanges")
Error in end(ranges)[!is.na(meta$END)] <- meta$END[!is.na(meta$END)] : 
  invalid subscript type 'S4'

When I run the code of the as method line by line I see that in my case the END field in the meta variable is a <IntegerList> meanwhile in the example vcf file of the package (chr22.vcf.gz) is a <integer>.

By comparing the example file and my vcf file I can see that the END INFO field specification is different. Instead of Number=1 it says Number=..

I am not an expert in VCF field specification but I know they can be a mess. So this might not be a VariantAnnotation issue but the variant caller issue. Anyway, What should I do to be able to import those vcf files into R? I would like to avoid modifying the raw vcf files.

How to reproduce it?

Get the example chr22.vcf.gz file in the package. (use fl = system.file("extdata", "chr22.vcf.gz", package="VariantAnnotation") to locate where it is). Uncompress and modify the line:

##INFO=<ID=END,Number=1,Type=Integer,Description="End position of the variant described in this record">

to this

##INFO=<ID=END,Number=.,Type=Integer,Description="End position of the variant described in this record">

Then run the code:

> from = readVcf("path/to/R/win-library/3.5/VariantAnnotation/extdata/chr22.vcf") #this is the modified file
> as(object = from,Class = "VRanges")
Error in end(ranges)[!is.na(meta$END)] <- meta$END[!is.na(meta$END)] : 
  invalid subscript type 'S4'

Thanks, Should be fixed in 1.26.1 (also in devel).