I am working with data from the HGDP public files. These are massive VCFs from whole genome sequencing. To begin with, I converted these to BEDs using plink (1.9). The next problem is that these files use rsid names instead of chrpos names. I would like to change to chrpos. I am wondering whether one can simply edit the BIMs or whether I have to do use plink's
--update-name call. The problem with the latter is that it doesn't support duplicates. The HGDP files are full of duplicate variant names because many of the variants don't have rsid's and are just named ".":
$> plink --bfile hgdp_chr21 --update-name updateids --make-bed --out hgdp_chr21_chrpos PLINK v1.90b6.17 64-bit (28 Apr 2020) www.cog-genomics.org/plink/1.9/ (C) 2005-2020 Shaun Purcell, Christopher Chang GNU General Public License v3 Logging to hgdp_chr21_chrpos.log. Options in effect: --bfile hgdp_chr21 --make-bed --out hgdp_chr21_chrpos --update-name updateids 64259 MB RAM detected; reserving 32129 MB for main workspace. 1086522 variants loaded from .bim file. 929 people (0 males, 0 females, 929 ambiguous) loaded from .fam. Ambiguous sex IDs written to hgdp_chr21_chrpos.nosex . Error: Duplicate ID '.'.
Thus, one gets an error when trying
--update-name and I am wondering if I can simply edit the BIM files to be chrpos instead of rsid. I can't find an answer to this anywhere. It seems that if the BED files don't contain a reference to the variant IDs, but only to the row in the BIM file, then this approach should be possible.
Here's the current BIM file:
user@desktop:/media/luks8tb1/data/genomics/HGDP$ head hgdp_hg38.bim 1 rs1331781659 0 10153 G A 1 rs1179689498 0 10163 C T 1 rs201694901 0 10180 C T 1 rs199706086 0 10250 C A 1 rs111200574 0 10257 C A 1 rs145427775 0 10291 T C 1 rs868413313 0 10297 T C 1 rs112750067 0 10327 C T 1 rs1035171912 0 10330 A C 1 rs868804735 0 10333 T C
As can be seen, very few lines have the "." variant name. In fact, in the first 100 lines there are 0.