Hello.
I am a new in GWAS.
I have 70 samples. 35 samples are controls and 35 samples are cases.
What I'm curious about is whether the phenotype needed to do a GWAS can only have the numbers 0 or 1. In other words, can I perform GWAS by setting controls to 0 and cases to 1 in the data I have?
The phenotype that can be determined from the data I have is whether the sample is a control or a patient.
Yes, you can conduct GWAS using binary trait data, but it comes with a whole suite of complications. See this study for context and a benchmark of tools.
However, since you just say phenotype, it is unclear if the traits you have measured are suitable for GWAS. If you have a small number of traits, I suspect it would be simpler to look at allele frequencies of your SNP data and look at how SNPs segregating by phenotype are distributed. Are you sure your traits are heritable?
Finally, your sample sizes are on the small side for GWAS analyses, and you may find your analysis is underpowered.
