I have recently started doing GWAS and I chose to do it with SNPTEST. I would appreciate if you could help me with the following question,
I have 2500 samples. I have two phenotypes of interest - continuous (0 - 1.5) and discrete phenotype (numbers between 0 to 20). There are 500 missing values and 400 zeros in the continuous phenotype.
My questions are as follows,
SNPTEST with the continuous phenotype gives output file that has no P-values at all. It says "number of samples below limit". Is it due to the missing values and zeros. I still have more than 1000 samples, is it too low for a good fit?
I am not able to run SNPTEST with discrete phenotype (numbers between 0 to 20). I get error that "this is not one of the phenotypes in the sample files". However I checked and its there in the file categorized as Discrete (D). Does the positioning of the phenotype variables matter in the phenotype file?
I would appreciate for any help here!
best regards, Krishna