Comparison of allele frequencies between population

Question

[elisheva]

Hi all,
I would like to examine whether the frequencies of many alleles differ between the African and North European populations.
I downloaded the 1000 genome project wgs vcf file via the FTP site: http://ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/20130502/ALL.wgs.phase3_shapeit2_mvncall_integrated_v5b.20130502.sites.vcf.gz
But the allele frequencies are given for super populations, thus the European population includes Spain and Italy.
Does anyone familiar with this kind of data and could help?

Thanks

Answer 1

[chrchang523]

1. Download the pedigree file from https://www.internationalgenome.org/faq/can-i-get-phenotype-gender-and-family-relationship-information-samples/ .
2. Define your European subpopulation, and create a file with just the sample IDs within that subpopulation with e.g. a short shell script.
3. Compute allele frequencies using just those samples. This can be done efficiently with e.g. plink2 --vcf <VCF file> --keep <sample ID file> --freq.

[elisheva]

Thanks for your response.
It seems that the versions of the vcf file and the pedigree file are different.
I couldn't find the matching version for both files.
Is there is any option to solve this?

[chrchang523]

The pedigree file contains some samples that were excluded from the phase 3 VCF, but that's okay, the extra sample IDs will just be ignored by --keep.

[elisheva]

Apparently, there are no sample Ids in the vcf file I posted, and I couldn't find any others.

[chrchang523]

Oh, that's because you downloaded the sites VCF instead of the full genotypes VCF. Download a .genotypes.vcf.gz file from http://ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/20130502/ instead, or the plink2-formatted version from https://www.cog-genomics.org/plink/2.0/resources#1kg_phase3 .