Thanks for your quick response! I don't see a genomics tab in my cohort browser nor do I see a GENO filter. In this case, would the alternate option be to calculate it myself using PLINK (merge exome files and calculate freq?)? I currently have access to the 200K exome release
Yes, you will see GENO tab only if you have access to 450K WES data. In other case it is better to proceed with PLINK2, it is installed in Swiss Army Knife tool (you need to log into RAP to open this link) . You can then use following command in Command line parameter
Input file would be PLINK fileset (.bed + .bim + .fam), e.g. ukb23155_c1_b0_v1.bed, ukb23155_c1_b0_v1.bim, ukb23155_c1_b0_v1.fam. These files will be located in subdirectory of Bulk folder (e.g. /Bulk/Exome sequences/Population level exome OQFE variants, PLINK format - interim 200k release).
I can reproduce your problem. It is probably caused by presence of missing values for some samples. What solved the issue for me is ignoring phenotypes using --no-fam-pheno parameter, e.g.:
Comments
7 comments
In cohort browser, you can use Variant browser. Population Allele Frequency there is computed using 450k WES data
Thanks for your quick response! I don't see a genomics tab in my cohort browser nor do I see a GENO filter. In this case, would the alternate option be to calculate it myself using PLINK (merge exome files and calculate freq?)? I currently have access to the 200K exome release
Yes, you will see GENO tab only if you have access to 450K WES data. In other case it is better to proceed with PLINK2, it is installed in Swiss Army Knife tool (you need to log into RAP to open this link) . You can then use following command in Command line parameter
plink2 --bfile "file_prefix" --freq --out "file_prefix"
Input file would be PLINK fileset (.bed + .bim + .fam), e.g. ukb23155_c1_b0_v1.bed, ukb23155_c1_b0_v1.bim, ukb23155_c1_b0_v1.fam. These files will be located in subdirectory of Bulk folder (e.g. /Bulk/Exome sequences/Population level exome OQFE variants, PLINK format - interim 200k release).
Great, thanks!!
I get the following error when I ran the command:
+ eval 'plink2 --bfile ukb23155_c21_b0_v1 --freq --out ukb23155_c21_b0_v1'
++ plink2 --bfile ukb23155_c21_b0_v1 --freq --out ukb23155_c21_b0_v1
Error: 'PHENO1' entry on line 8353 of ukb23155_c21_b0_v1.fam is categorical,
while earlier entries are not.
(Case/control and quantitative phenotypes must all be numeric/'NA'.
Categorical phenotypes cannot be 'NA'--use e.g. 'NONE' to represent missing
Do you have any advice on how to format the fam file?
I can reproduce your problem. It is probably caused by presence of missing values for some samples. What solved the issue for me is ignoring phenotypes using --no-fam-pheno parameter, e.g.:
plink2 --bfile ukb23155_c21_b0_v1 --freq --no-fam-pheno --out ukb23155_c21_b0_v1
Thanks! That fixed the error ?
Please sign in to leave a comment.