Reference & alternative alleles extracted from 1000G, Phase 3

Question

[Mr Locuace]

Hello, I extracted the first 5 columns from the vcf files of 1000G, Phase 3, for each chromosome:

http://hgdownload.cse.ucsc.edu/gbdb/hg19/1000Genomes/phase3/

This is the script I used, for instance, for chromosome 22:

awk 'BEGIN {OFS ="," ; FS = "\t"};{print $1, $2, $3, $4, $5}' chr22.vcf

However, for some SNPs I get the following info (this is one example):

CHROM,POS,ID,REF,ALT
1,886817,rs111748052;rs10465241,C,CATTTT,T

I am not sure which one is the REF and ALT allele in this case, since this example has 6 columns, and the last 3 columns have allele info.

Thanks very much

[Pierre Lindenbaum]

I extracted the first 5 columns

how about using another delimiter ?

[Mr Locuace]

Thank you @Pierre Lindenbaum. I will try the tab delimiter.

[Joe]

It's weird. Why did ID show two SNPs? I think your script extracting the 5 columns from vcf files made those mistakes. You can other scripts.

[Mr Locuace]

Hi @Joe, please see the solution I posted.

Answer 1

[Mr Locuace]

SOLUTION

The columns must be extracted using a tab as outer field separator (OFS), as one of the users suggested:

awk 'BEGIN {OFS ="\t" ; FS = "\t"};{print $1, $2, $3, $4, $5}' chr22.vcf

Doing this it becomes clear which columns have the REF & ALT alleles. For instance, for the following SNP, G and T are the ALT alleles:

22  18292633    rs430321    A   G,T