Hi
I have never used logistic regression but I would like to understand the below paragraph. Does it work by entering lots of SNPs that are either case or control as dependent variables, and then classify whether a SNP genotype (independent variable) is considered to be in the case or control according to the odd ratio by the p value from the standard error?
The key analysis in a GWAS for a disease is logistic regression with the dependent variable case-control status (1=case, 0=control) and a SNP genotype as an independent variable [coded as the number of copies of the minor or less frequent allele, 1 degree of freedom (df)]. The output of a logistic regression is identity of the reference allele and an odds ratio with its standard error (or confidence intervals) along with a statistic and a p value that tests whether the odds ratio differs from unity.
Many thanks