I am trying to perform LD clumping on GWAS summary file. The problem is that the file lacks rsID and only has Chr and Pos columns. Hence I create a column named ID which is in the format #:Pos. I also run this command on the 1KG bed file:
./plink2 --bfile ~/genetics_2020/1kgtot_filt --set-all-var-ids @_# --make-bed --out ~/genetics_2020/1kgtot_filt.noid
So in both the 1kg bim file and in the gwas text file, the chr_pos column has this format: 1_10177
But when I try
./plink --bfile ~/genetics_2020/1kgtot_filt.noid --clump-r2 0.1 --clump-kb 250 --clump-field p -clump-p1 0.00000005 --clump ~/Documents/schizophrenia/PGCALZ2sumstatsExcluding23andMe.new.txt --out ~/Documents/schizophrenia/ALZA.r2.5e8
I get Error: No variant ID field found in
/home/davide/Documents/schizophrenia/PGCALZ2sumstatsExcluding23andMe.new.txt.
I tried running the code with the --clump-snp-field ID flag but I get the same error message.
How do I clump without rsID? Is it possible to do it using the position only?