Question: PLINK error: no --covar values loaded
0
gravatar for kindlychung
4.2 years ago by
kindlychung40
Netherlands
kindlychung40 wrote:

Here is the command that I used:

plink --allow-no-sex --bfile ./collr_task_Curliness-sex_age_s400_p1.00e-01/RS123_3069obs_shift_0000 --covar curliness_RS_withIID.tab --covar-name sex,age --linear hide-covar --out ./collr_task_Curliness-sex_age_s400_p1.00e-01/RS123_3069obs_shift_0000 --pheno curliness_RS_withIID.tab --pheno-name Curliness

And here is the log:

PLINK v1.90b2m 64-bit (15 Oct 2014)        https://www.cog-genomics.org/plink2
(C) 2005-2014 Shaun Purcell, Christopher Chang   GNU General Public License v3
Logging to ./collr_task_Curliness-sex_age_s400_p1.00e-01/RS123_3069obs_shift_0000.log.
96680 MB RAM detected; reserving 48340 MB for main workspace.
831138 variants loaded from .bim file.
2905 people (0 males, 0 females, 2905 ambiguous) loaded from .fam.
Ambiguous sex IDs written to
./collr_task_Curliness-sex_age_s400_p1.00e-01/RS123_3069obs_shift_0000.nosex .
2221 phenotype values present after --pheno.
Using 1 thread (no multithreaded calculations invoked).
Error: No --covar values loaded.

Note that I used the same file for --pheno and --covar, but plink didn't complain about --pheno at all.

The first 3 lines of the phenotype file looks like this:

FID IID ImageID Curliness sex age ergoid
1b 42326 102347101654 NA m 68.7527777777778 82002
1a 2112329 773125453 0 m 69.2194444444444 261001
plink gwas • 2.7k views
ADD COMMENTlink modified 4.2 years ago by chrchang5234.8k • written 4.2 years ago by kindlychung40

The first thing I'd check is the capitalization of "sex" and "age" in your phenotype file.

ADD REPLYlink written 4.2 years ago by chrchang5234.8k

They are both lower case, I have checked.

ADD REPLYlink written 4.2 years ago by kindlychung40
1

Oh, the actual problem is that sex is not a numeric covariate.  This should work after you recode it to e.g. 1 = male, 2 = female.

ADD REPLYlink written 4.2 years ago by chrchang5234.8k

Oh, my bad. I should have spotted this right away. Thanks! Could you put this into an answer so that I can accept it?

 

ADD REPLYlink written 4.2 years ago by kindlychung40
2
gravatar for chrchang523
4.2 years ago by
chrchang5234.8k
United States
chrchang5234.8k wrote:

--covar requires numeric phenotypes.  Representing sex as 1 = male and 2 = female, instead of m = male and f = female, should work.

ADD COMMENTlink written 4.2 years ago by chrchang5234.8k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1155 users visited in the last hour