Different format fields in VCF files

[Leonardo Caserta]

Hello, 

I have been using Lofreq combined with Variabel for low frequency variant calling of Nanopore data, as it seems to be the best option for this kind of data. However, the vcf output of Lofreq includes "AF" , but not "AD". Is there any way I could run MixviR without "AD" and use only "AF" that was already calculated in the input VCF file? 

I tried using the example provided in the vignette (bcftools variant calling), but looks like the ouput has a lot of spurious indel calls, since Nanopore data is much more noisy than Illumina. 

Thank you

Leonardo

[Mike Sovic]

Hi Leonardo,

Unfortunately, I don't have an immediate solution to this. As of the current version (3.5), there isn't any way around not having info from that "AD" field. This has me thinking about how to add that in as an option in the next version though, and I'll try to get to that soon. Do you have the "DP" field in your vcfs, and if so, are your "AF" and "DP" fields INFO or FORMAT entries? If you have both "DP" and "AF", it actually might be a pretty easy update. It's probably a bit more of a lift (though still doable) if only the "AF" field is present, and so you (or any user) just wouldn't want seq depths to be reported at all.

         Mike

[Leonardo Caserta]

Hi Mike, 

Yes, there is DP and AF under INFO entry. The thing is, it reports only the variants, I don't know if that would be an issue for MixviR, because the GATK command for variant calling reports all the positions.

Here's how it looks like:

#CHROM    POS    ID    REF    ALT    QUAL    FILTER    INFO
KX778101.1    289    .    T    A    18100    PASS    DP=3376;AF=0.493187;SB=14;DP4=530,612,907,892
KX778101.1    382    .    G    A    5753    PASS    DP=3314;AF=0.253470;SB=6;DP4=1185,1210,407,457
KX778101.1    785    .    C    T    4924    PASS    DP=3230;AF=0.274613;SB=7;DP4=1106,1154,419,487 

Thank you 

Leonardo