How unphased VCF is converted into ped file?

Question

[jingjin2203]

Hi All,

I have some ddRADseq data from a diploid organism I'm working on.

I've generated an unphased VCF file using freebayes that I wanted to convert into PED file. I was wondering how does VCF to PED conversion deal with unphased VCF data? Because when I further converted PED to FASTA, each of the sample had two reads, and the two reads for each sample were different. So how does the conversion program distinguish two alleles at a heterozygous site for each read?

Hope my question makes sense. Any answers or comments will be appreciated!

Thanks!

Answer 1

[Kevin Blighe]

I presume that you mean the running of plink --vcf on your file, i.e., in order to convert it to PLINK PED format?

The latest implementation of PLINK ignores phasing information. In the heterozygous situation, all variant alleles become A1 whilst the reference alleles become A2. In the homozygote situation, variant alleles are obviously set to both A1 and A2.

Kevin

[jingjin2203]

Thanks, Kevin! Yes, that's exactly what I was asking. Really appreciated your kind reply! Just a follow up question, do you know how I can convert phased vcf file to plink ped format with the phasing information incorporated?
Thank you!

[Kevin Blighe]

I'm not sure that phasing information is ever taken into account in PLINK. The person who will know is chrchang523

If you take a look here: https://www.cog-genomics.org/plink/1.9/input#vcf

--vcf loads a (possibly gzipped) VCF file, extracting information which can be represented by the PLINK 1 binary format and ignoring everything else (after applying the load filters described below). For example, phase and dosage information are currently discarded. (This situation will improve in the future, but we do not have plans to try to handle everything in the file.)

It says that phasing information is "discarded" ...

[jingjin2203]

Thank you, Kevin! Really appreciated it!