Unable to load vcf file about variantannotation HOT 7 CLOSED

you did a lot of work. I think the problem is a stray blank character at the end of the #CHROM line. Once I remove that readVcf works.

from variantannotation.

thanks for the thorough report.

from variantannotation.

How did I find it out? I ran the example of readVcf, then did a readLines on the example file and on your excerpt of the problematic file. I noticed that the #CHROM lines differed as noted. A sad lack of resilience in the parsing infrastructure.

from variantannotation.

Thank you so much for your quick reply and solution! I didn't know about the readLines() function, but I just ran it and see the same space you did! After taking the space away vcf now loads.

I will close the issue now.

Thanks again!

from variantannotation.

you did a lot of work. I think the problem is a stray blank character at the end of the #CHROM line. Once I remove that readVcf works.

Hi, could you clarify what you mean by a stray blank character at the end of the #CHROM line? I'm having my own problems with bcftools not parsing header and think this might be my solution.

from variantannotation.

Hi,

If you're working with R you can use the readLines() command to look at all input lines of your vcf file, line by line (same thing as row by row). Your output should be something like this:

#CHROM\tPOS\tID\tREF\tALT\tQUAL\tFILTER

Notice that this is a tab separated file and there are no spaces anywhere, only the \t denoting the tab separation.

If you happen to see something like this:

#CHROM\tPOS\tID\tREF\tALT\tQUAL\t FILTER

Where there is a weird space somewhere, then it may be difficult to parse your vcf file. Also take note of the error messages. If I remember correctly, if there is an "INFO" column, then it must be proceeded by some columns with metadata (something to that effect). Also, if you look up the current vcf guidelines, the formats of vcf files are standardized. So perhaps your vcf file is somehow different than the standard format and for that reason it cannot be parsed/loaded.

from variantannotation.

I see you're working in python. In python you can use readlines()

from variantannotation.