PLINK --lasso Issue ('Warning: No Phenotypes Present') even with a covar file is present
1
0
Entering edit mode
19 months ago
ErickW • 0

Hello,

I'm utilizing PLINK --lasso for feature/SNP selection, running the following command:

plink --bfile SNPsFile --allow-no-sex --lasso 0.46 --covar pheno.txt  --covar-number 1-9 --make-bed --out Test


Resulting in the output below:

Even with covars uploaded, I am still getting this issue. Pheno.txt is a tab-delimited .txt file.

0
Entering edit mode
19 months ago

You're mixing up phenotype and covariate values.

0
Entering edit mode

Could you clarify that a bit? Like there is a difference between the two? I gathered that they mean the same from the PLINK documentation, unless I am missing something?

If so, then is there a process that I transform my phenotypes into covariates?

0
Entering edit mode

That only says that the file format is the same.

You need to take a step back and read about what phenotypes and covariates are in GWAS.

0
Entering edit mode

I get what you mean, but looking at various pieces of text and posts that speak about phenotypes and covariates, I don't feel like I've gotten much clarity on what I'm doing incorrectly. If for example, I was performing a GWAS with the phenotype of interest being BMI, I understand that gender may be a covariate of interest (based on what I understood from the following linked post and study: GWAS: when is it appropriate to add covariates? and Including Covars Can Reduce Power).

Looking at my example of BMI being the phenotype of interest and what is in my phenotype file, I have metrics of:

Age, Gender, # of times you work out per week, and BMI.

So if I believe gender and #workouts per week to be possible covars, is it wrong to have BMI in the same file? Or am I still misunderstanding something?

0
Entering edit mode

You can have them in the same file; that's why the file format is the same. But you have to tell plink which column is the phenotype you're interested in!