Using 1000 Genomes Vcf To Plink Conversion Script To Convert My Own Vcf Files?

Question

[Click downvote]

I have a vcf file I'd like to convert from vcf to ped. Luckily, 1000 genomes has a perl script that can be used to do this:

http://www.1000genomes.org/vcf-ped-converter#API%20Script

Problem is, that script requires a -sample_panel_file, but I have no such thing. The VCF I have is from one population only and it is not from one of those in 1000 genomes. Does this mean I can't use this script to convert my own vcf file or what?

Or if I find the ids of the people in the vcf, can I just roll my own -sample_panel_file by making a file like

id1, population_name_i_made_up
id2, population_name_i_made_up
...

All advice appreciated.

Ps. I know vcftools has a vcf to ped converter but it does not play well with ubuntu thanks to ulim. I also know GATK has a tool to do this conversion, which I am attempting to get to work atm. If you know any other ways to convert vcf to ped (no matter how circuitous, please do tell me!)

[zx8754]

PSEQ reads in VCF and can output plink format.

[Click downvote]

No, it only creates tped files (which vcftools also does swimmingly.) But upvote since I was too vague in the title.

[Click downvote]

Use the new plink, people: https://www.cog-genomics.org/plink2/

Answer 1

[Neilfws]

You can try the R package SNPRelate:

• the function snpgdsVCF2GDS() will convert VCF to GDS
• the function snpgdsGDS2PED() should convert GDS to PED

I say "should" since I have not tried the second function (I know that the first works fine).